2 N E W S & R E S E A RC H Computers will support clinicians, not replace them AI solutions aid diagnoses and therapy choices Report: Cornelia Wels-Maug Neuroscientist Lynda Chin MD, Founder and CEO of Real-world Education Detection and Intervention, has lit- tle doubt: ‘Artificial intelligence to the rescue,’ she proclaimed in her keynote address at the Artificial Intelligence and Machine Learning Summit, held in Las Vegas this spring. ‘We need a system and ana- lytics to interpret data!’ she urged, despite being well aware that build- ing a suitable algorithm, as well as training it to discover patterns and identify relationships, is a tedious task. ‘We train an artificial intelli- gence (AI) system just as we would train a doctor’, she explained and Roland Berger of the strategy con- sultants and human venture capital firm Asgard, ‘Artificial Intelligence – A strategy for European startups’, the healthcare and FinTech industries together, after B2B services, attract most AI start-ups on a European and global level. Nonetheless, have those undertakings in AI, machine learning (ML) and deep learning effected improvements in healthcare on a larger scale? Solving real problems Although AI has become a buzz word in the healthcare world, its use is still in early days. AI is accredited with revealing clinical and opera- tional insights and contributing to impending cardiological problem. However, they are wrong in one in five cases, leading to patients being discharged and suffering a heart attack at home or having unneces- sary surgery. To train Ultromics, Leeson used the scans of 1,000 patients he had treated over the past seven years, along with information about whether those patients continued having heart problems. The solu- tion picks up details in the images doctors cannot see and gives recom- mendations as to whether a patient should be treated further. Ultromics has been tested in clini- cal trials in six cardiology units and, according to Leeson, the data indi- Radiomics discovers relationships between image features and clinical data. The ‘heatmap’ groups patients with similar features and represents them in areas of the same colour. By courtesy Fraunhofer MEVIS pondered, ‘What is the equivalent of a residency programme for an AI programme?’ Global interest in AI Work to derive actionable insights from ever-increasing data is the next logical step. Hence, self-learn- ing algorithms and their underlying neuronal networks have become a major focus in software develop- ment. As the development of AI is high- ly correlated with the availability of data, healthcare attracts a lot of AI activities. According to a study by spare patients from unnecessary procedures and worry. Some exam- ples on how AI enhances diagnos- tics and improves disease manage- ment include: Picking up illnesses earlier Ultromics, a new AI-based solution developed by Paul Leeson, Professor for Cardiovascular Medicine at the University of Oxford, diagnoses car- diac scans to determine a patient’s risk of an imminent cardiological problem. Currently, cardiologists can defer from the timing of the heartbeat in scans if there is an cates that the solution has greatly out-performed its fellow cardiolo- gists. The trial results suggest fur- ther that Ultromics could save lives and spare patients unnecessary sur- gery, which Leeson reckons could save England’s NHS more than £300 million per year. Another example is an AI/ ML-based system by the British start-up Optellium, which addresses the management of patients with large clumps of cells – nodules – in the lungs. According to Optellium, up to 30% of all patients scanned have such small nodules of which Paul Leeson, Professor for Cardiovascular Medicine at the University of Oxford the majority are harmless. However, radiologists often struggle to deter- mine if a nodule is cancerous, lead- ing to an indeterminate diagnosis that needs up to two years follow-up imaging to monitor potential growth and, in some cases, the performance of biopsies and surgeries. The solution, the firm reports, ‘provides an objective risk score of nodule malignancy learned from a database of thousands of examples with known ground-truth diagno- ses’, enabling clinicians to ‘confi- dently stratify lung nodule patients earlier, potentially on the basis of only one or two scans’. Optellium’s chief science and technology officer, Dr Timor Kadir, reports that trials of the system in Manchester imply that more than 4,000 lung cancer patients a year, in the UK alone, could be diagnosed much earlier and hence have a much greater chance of survival. Radiomics on the radar The Fraunhofer Institute for Medical Image Computing (MEVIS) and part- ners are deploying radiomics to dis- cover relationships between image features and clinical data within the Patient-Oriented Oncological Therapy Assistance (PANTHER) pro- ject. The aim is to develop a clinical decision support system for physi- cians to help determine the best course of therapy (medication, radi- ation, or a mix of both) for patients suffering from cancer of the colon or lymphoma. This will allow ben- efits to be maximised and curtail the side effects of a chosen therapy in a hitherto uncharted way by adjusting the course of prescribed therapy early in the treatment. Jan Hendrik Moltz PhD, Research Scientist at Fraunhofer MEVIS, explained: ‘So far, the size devel- opment of a tumour is the most important criterion in evaluating CT images. However, there is much more information hidden in those scans that has hardly been used.’ He refers to information on changes in the shape and texture of a tumour that occur during therapy, but that cannot be detected by physicians with their bare eyes. By deploying AI and deep learn- ing on historical data from the Medical Centre of the University of Munich, it is hoped to reveal insights into the effectiveness of the therapy of bowel cancer and lymphoma. Mid-term into the three-year-long project, Moltz reports good progress Jan Hendrik Moltz PhD, Research Scientist at the Fraunhofer Institute for Medical Image Computing (MEVIS). Image: Fraunhofer MEVIS on homogenising the data, as well as segmenting computed tomography scans – segmenting refers to mark- ing the structures to be analysed on the scans. ‘Analysing the spleen, for example, manually, is very time-con- suming,’ Moltz explained. We have managed to develop a segmentation algorithm based on deep learning that is already so effective that it almost needs no further manual cor- rections. Automating the segmenta- tion enables us to obtain outcomes much faster.’ The future role of the clinician Last November, ‘Xiaoyi’, meaning ‘lit- tle doctor’, an AI-fed robot, success- fully passed the medical approval test in China, even with above aver- age results. Will this be the future? Chin shook her head: ‘Data ana- lytics helps healthcare to be more efficient, but AI is just a tool. It needs clinical experts to understand the problem. Although computers will make therapy decisions, they will support clinicians, not replace them.’ Top Spanish innovation gains awar Tackling an epidur anaesthesia complication Report: Mélisande Rouger Four potentially game changing solu- tions selected for the annual com- mercial acceleration program of the Spanish foundation for innova- tion and prospective in healthcare (Spanish: FIPSE), included Duralock, a system that could ease post partum Carlos García (left) led a team of anaesthetists and engineer PDPH. A researcher at the Valencia foundation for healthcar among the Top 10 Spanish Talent list, which competed in the Eur EUROPEAN HOSPITAL Vol 27 Issue 4/18

The researchers developed a sen- sor that uses a photonic crystal cav- ity, a ‘trap’ of just a few micrometers into which incoming light falls if it has the right frequency. Fiore: ‘This trap is contained in a membrane, into which the captured light gener- ates a tiny electrical current that is precisely measured, retaining just a very tiny frequency interval and therefore measuring only light at that frequency.’ Two of the membranes were placed closely, one above the other, allowing them to influence one another: a slight change in dis- tance means the measured light frequency also changes. To ensure a varying distance between the mem- branes, the researchers incorporated a MEMS (micro-electromechanical system). A similar type of MEMS is already present in smartphones Sketch of proposed microspectrometer and other devices to sense free fall and put the sensitive parts in secure mode before ground impact. Fiore: ‘Using our present system, the sensor covers a wavelength range of around thirty nanometers, within which the spectrometer can discern some hundred thousand frequencies. This is exceptionally precise because we can control dis- tance between the membranes to just a few tens of femtometers’ (one femtometer is one millionth of a billionth of a meter, or about one million times smaller than typical atom size). Applications Measuring light gives a spectrom- eter a wide range of applications – two of the most interesting being in food and medical industries, says Fiore, used, for example, to measure CO2, check the pill you take is cor- rect, measure blood sugar level via Report: Madeleine van de Wouw Checking a lump for malignancy, finding out if food is fresh, just with your smartphone? It’s possible, according to Eindhoven University of Technology researchers in the Netherlands. Their recently presented spec- trometer is small enough to insert into a smartphone. This device is not yet ready for use on a big scale, Professor Andrea Fiore, supervisor of the Eindhoven research team points out. Spectrometry is the analysis of visible and invisible light. Every material and tissue has its own spectrum of light absorption and reflection. Some information on the spectrum of visible light is already gathered by our eyes – it gives the colour of the material. The invisible part of the spectrum also carries a lot of useful information, and can be ‘seen’ by a spectrometer. However, current devices are large due to the work they need to do: splitting light into frequencies (different colours). Each frequency must be measured separately, which can only be made some tens of cen- timetres after the splitting. Small v. big Created by Zarko Zobenica, a PhD candidate at Eindhoven under the supervision of Professor Fiore and colleague Dr Rob van der Heijden, the new microscale device is a small version of normal tabletop spec- trometers currently used in scientific labs. The difference is the way the small spectrometer works. Tackling an epidural anaesthesia complication Continued from page 3 Above: Optical-microscope image of fabricated microspectrometer Below: Electron-microscope image of the active area, with zoom-in on the photonic crystal cavity thin skin on the earlobes. Another application is to gain skin and tis- sue information, for example during surgery, using the spectrometer to determine whether exposed tissue is cancerous and help decide how much tissue to be removed, he sug- gests. The frequency range is too small to use on a big scale, Fiore explains. ‘The sensor covers only a few per- cent of visible and near-infrared light, the most used part of the spectrum. We need to expand this range for more general use. We also will need to integrate a light source so that the smartphone sensor can function independent of external sources. So, it will take years before the new spectrometer will be inte- grated within a smartphone and at least ten years before it’s on the market.’ Prof Andrea Fiore graduated in Electrical Engineering and Physics from the University of Rome ‘La Sapienza’. Between 1994-97, he wrote his PhD thesis on nonlinear frequency conversion in semiconductor waveguides at Thomson CSF Central Research Laboratory (Orsay, France). He held postdoctoral appointments at University of California in Santa Barbara, at the Ecole Polytechnique Fédérale de Lausanne in France and a research position at the Italian National Research Council. Between 2002-07, as assistant professor he led Quantum Devices activities at the Ecole Polytechnique Fédérale de Lausanne. Since October 2007, he has been Chair of Nanophotonics at Eindhoven University of Technology, the Netherlands. Nothing gives so much inf 4 N E W S & R E S E A RC H This app will become as normal as having a camera Smartphone to measure light spectrum The future Many companies and scientists are developing microspectometers, with different approaches. The Eindhoven team is also working to enlarge the measured spectrum width. ‘The technology will get there,’ Fiore assures. ‘The question is only whether a market for these microspectrometers exists and, if so, which company will produce them. If I’m correct, the future will show that having a spectrometer in your smartphone is as normal as having a camera. The only thing is the future must wait a couple of years.’ Being able to seal this breach by placing an implant can help prevent the occurrence of invalidating and sometimes long lasting headaches, according to Carlos García Vitoria, an anaesthesiologist from Valencia, who long thought about a way to get past the problem. ‘Epidural puncture aims to relieve pain, but paradoxically in 2% of cases it can cause a more intense and lasting pain than the original discomfort, which motivated drug administration. We need to evalu- ate how to act before headache occurrence to discard this invalidat- ing clinical scenario,’ said García, who led a team of anaesthesiolo- gists and engineers from Dr Preset University Hospital and Valencia Biomechanical Institute to develop Duralock, a sealing system that can help avoid PDPH. The system comes in a kit that includes an implant and tools nec- essary for its insertion through a needle that is conventionally used in epidural puncture. The implant can be applied in 30 seconds with just one hand in a completely ster- ile way, and does not present any risk of additional dural perforation, García pointed out. Besides impairing the patient’s life, PDPH can extend hospital stay by five days and generate higher healthcare costs. In 2005 in Spain, Doctors working in the eight-bed Paediatric Intensive Care Unit (PICU) at the Ramón y Cajal University Hospital in Madrid extensive- ly use point-of-care ultrasound to evaluate the condition of critically ill children, and they find it essential in their work, as Dr José Luis Vázquez Martínez, Head of PICU at Hospital Ramón y Cajal, with over 25 years’ experience in paediatric intensive care medicine, explained. P o i n t - o f - c a r e u l t r a s o u n d is used (POCUS) extensively in our unit, allow- ing comprehensive, head-to-toe assessment of critically ill children, including respiratory, oncology and post-operative car- diac patients, as well as those being treated for sepsis or multiple trau- ma. The POCUS approach allows not only an initial diagnosis, but also routine monitoring of treatment to see whether or not a patient’s condition changes, enabling alterna- tive strategies to be implemented if there is no improvement. POCUS helps paediatric doctors in many ways. For example, ultra- sound scans enable evaluation of a patient’s haemodynamic state, look- ing at their heart function and blood volume to see if these factors are contributing to respiratory failure. Conversely, doctors can see if a lung problem, such as pneumonia, is affecting the heart. For a patient in a coma due to multiple trauma, ultrasound is used to look for signs of bleeding – a potential cause of unexplained anaemia – and to assess the intracranial pressure. It’s also used to monitor kidney function in children with blood pressure problems, and visualise intestinal indications of sepsis. In addition, ultrasound guidance can be used for endotracheal intuba- tion. In short, broader applications that we did not anticipate until very recently. We have used ultrasound in our PICU for more than a decade, and have always had SonoSite systems, upgrading them as new technol- ogy is introduced. In the beginning, when my knowledge was more limited, the aim was to perform clinical echocardiography but, when the SonoSite representative showed me the linear probe and the various EUROPEAN HOSPITAL Vol 27 Issue 4/18 accidents occurred during 3,425 dural punctures. García estimated that Duralock could help to save as much as €10,500,000 annually for the nation’s health service. Due to the award, received in March in Madrid, the product can benefit from the medical industry’s expertise to help speed up release to market, notably by helping recipi- ents to remodel strategy and busi- The team’s solution, named Duralock, comes in a kit that includes an implant and necessary tools for insertion through a needle ness plans. Among the winning initiatives, Duralock will be featured at BioBoston 2018 and JP Morgan Healthcare 2019 in San Francisco, to boost its visibility in the North American market. 50 μm1 μm500nm

How can DBT image reading time be decreased? ‘If I had an answer to that, I’d prob- ably be rich. It’s a difficult question with no clear-cut solution. However, there are several promising studies and approaches. One way to short- en reading time could be to use the so-called slabbing technique, where thinner image slices are combined in thicker slices to reduce the image volume. ‘Hotly discussed are also devel- opments including AI and Deep Learning. I myself would like to nav- igate easier in the tomo-stack, so a CAD system that indicates important features or lesions would be much appreciated. I could also imagine a system trained on a big data set that gives you a rating of probability of malignancy to be beneficial. So CAD and AI are very likely the way forward to assist radiologists find- ing the relevant parts in the DBT volume more quickly. ‘Another interesting experi- ment was conducted recently: Our research group used an eye tracker to see how fast readers were find- ing lesions in differently presented DBT images, and found out that, when DBT images were displayed horizontally, the radiologists read them a lot faster, because their eyes were more adjusted to the horizon- tal alignment. ‘There is powerful potential for improvements out there, but we also should not forget that reading images has always been a matter of training and experience. It’s not unthinkable that we can reduce time and improve sensitivity and specific- ity significantly simply by working with DBT on an every-day basis. It has almost become a proverb by now, but we really need more stud- ies in this area.’ Future plans ‘There’s still a lot of information to analyse from the trial, such as bio- logical tumour characteristics. But, we are also planning to execute a Associate Professor Sophia Zackrisson MD PhD is senior consultant radiologist at Skåne University Hospital in Malmö, Sweden, and a senior lecturer in radiology at Lund University and research group leader of diagnostic radiology. Her research focus lies on mammography screening and breast cancer. She received her medical degree from Lund University in 1999 and her PhD in Epidemiology in 2006. Following her residency in radiology in Skåne, from 2012-17 she headed the department of oncological imaging at Skåne University Hospital. meta-analysis of three studies in the first step – the STORM, Oslo and Malmo trials – including even more trials along the way to get influen- tial results and see if we can show, for example, a common decrease in interval cancers. ‘The question that always comes up is: are we going to imple- ment tomosynthesis in screening? Ultimately, I don’t know, but I am positive that it could happen. Breast cancer screening is one of the best screening programs we have in Sweden, but DBT could enhance it even more. It will be a long journey, but I’m convinced that, if you want to improve, tomosynthesis is an easy way forward. You just replace a unit in the same building in the same room with the same technologies and receive better mammography. People only need to be convinced of the benefits. ‘If you start using DBT in your own practice, you soon notice how many cancers you miss with 2-D only and usually that’s enough to convince people. Tomosynthesis is not perfect and we should not have blind faith in it, but it’s a great screening tool for a vast majority of average risk women. Personally, I think that a grass roots approach would help. ‘Start in several places by install- ing units in different centres and invite radiologists to try them out. When using it yourself, you really understand what the technology can give.’ n e d e w S , ö m l a M l a t i p s o H y t i s r e v i n U e n å k S : y s e t r u o c e g a m I 3-D mammography (tomosynthesis) to the left shows a 1 cm large tumour which is not as clearly visible on the 2-D mammography image to the right. The breast is quite fatty and it is evident that not much overlapping tissue is needed to compromise lesion visibility on 2-D mammography and hence the case illustrates the superiority of the tomosynthesis technique. dose for one-view DBT than two- view DM. Though radiation expo- sure has lost much of its signifi- cance in discussions, since we could strongly reduce dosage over recent years, it is still a valuable accom- plishment if we can improve our results and lower radiation exposure at the same time. 6 R A D I O LO G Y Digital breast tomosynthesis improves diagnosis DBT could boost breast screening Tomosynthesis is under international review, with a surprising number of enticing studies carried out in Northern European countries, among them one headed by Professor Sophia Zackrisson at Lund University, Sweden. When interviewed by Lena Petzold of European Hospital, she not only revealed surprising trial results, but also shared her thoughts on practical implementation and unusual speed-reading methods. ‘Our institute’s connection to tomo- synthesis goes a long way back,’ explained Sophia Zackrisson, sen- ior consultant radiologist at Skåne University Hospital. ‘We began to try out a prototype unit from Siemens as early as 2006 under my men- tor Ingvar Andersson, a well-known breast radiologist and researcher. We were, and still are, positively surprised by the technology’s ben- efits, especially in terms of eliminat- ing the so-called overlapping tissue effect. ‘We were quite sure that tomo- synthesis could have a major impact on screening, so we started some initial smaller studies. Yet, to really prove its effects, we needed data on a much larger scale, so we set out to implement a prospective trial. ‘When we decided on the basic outline, we reviewed the principles of screening: it should be acceptable for the population, should be fast, and should be safe. So, we decided to use a very simple protocol and show that it is still feasible to gain the same or even better results with one-view digital breast tomosyn- thesis (DBT) than two-view DBT + digital mammography (acquired or synthetic) as compared with two- view digital mammography (DM). We used the Siemens Mammomat Inspiration, but actually read the images on our regular Sectra PACS system. ‘We screened nearly 15,000 women over several years. Our participants were recruited from the regular screening population in Malmö, where according to Swedish recom- mendation every woman between 40 and 74 years of age is invited in different intervals for a preventive screening. Participants received the regular two-view mammogram first and then were screened with one- view DBT and reduced compres- sion, a special feature of the trial.’ Results ‘We collected a huge amount of data; and our most important results were that we not only found more predominantly invasive cancers in all density categories but that the detection rate with one-view DBT was 34 percent higher than with DM. When starting out, we were hoping for a 10 or maybe even 15 percent increase, so we were very happy when the results showed such a powerful augmentation. ‘We were also astonished that the detection was better in all densities including quite fatty breasts and not just in dense breasts. It shows that you do not need much overlapping tissue to hide a spiculated tumour. Furthermore, we were happy that we could reduce the compression in the DBT part of the trial by up to forty percent without losing valuable image quality. This could positively affect future screening attendance rates, since many women refrain from going because they fear the painful part of the examination. Dose ‘Another positive effect we came across concerned the radiation dose. We were able to use a 15 percent lower so-called average glandular Like a box of pumpkins, data contains big v Machine learning Standardised and well-structured data, as well as the definition of clear objectives, are indispensable prerequisites for artificial intelli- gence implementation into clinical processes. Report: Michael Krassnitzer ‘Ask not what artificial intelligence can do for you but what you can do for artificial intelligence!’ This vari- ation on John F Kennedy’s famous quote comes from Dr Ben Glocker, Senior Lecturer in Medical Image Computing at Imperial College London, UK. The occasion was a lec- ture given by the computer scientist at the European Radiology Congress (ECR 2018) on how machine learn- ing can be routinely implemented in medical image analysis. Artificial Intelligence (AI) was among the main topics at the con- gress; this also includes machine learning and its subdivision ‘deep learning’, which is based on so- called neural networks. AI entails an artificial network learning from examples and recognising inherent patterns and regularities by itself. Based on this, numerous applica- tions are already used in imag- ing, such as image segmentation, where an algorithm precisely marks out certain anatomic or pathologi- cal structures from their environ- ment (e.g. healthy brain tissue from tumour tissue in MRI brain scans, or whole body MRI scans of organs). Machine learning also facilitates the detection of new disease patterns, the extraction of clinically relevant information from multi-dimensional, multimodal image data that would otherwise remain concealed. Clearly defined objective needed ‘It’s basically a mathematical func- tion,’ says Glocker explaining machine learning. ‘This function is fairly complex, and we also cannot say exactly what is looks like, but like every function it has an input and an output.’ According to this UK-based German computer scien- tist, there are frequently misleading ideas around the necessary input. EUROPEAN HOSPITAL Vol 27 Issue 4/18

8 R A D I O LO G Y Machine learning: a valuable radiologists’ assistant AI will make them data scientists Machine learning is increasingly helping radiologists to acquire faster and better quality images, and measure heart function. This is just the tip of the iceberg; artificial intelligence has far more to bring to the heart, explained Daniel Rueckert, Head of the Department of Computing at Imperial College London, during CMR 2018. Convolutional Neural Networks for Image Segmentation artificial intelligence-based solutions. This, Rueckert is sure, will definitely not happen. ‘ML is a tool which radiologists will all use in the future when they look at images, and it will help to make better decisions based on better images. But I repeat, it will not replace them,’ he said. Artificial intelligence has the poten- tial to boost the radiologist’s con- tribution and value in the chain of healthcare – for example by bring- ing in quantitative assessment to the radiologist’s report, an objective information that is currently missing. ‘When radiologists look at the image and then make a report, they give you a very qualitative assessment of what they have seen. A blood test will give you a reading value, a quantitative measure, but an imag- ing examination won’t give you that for the moment. I think we will see much more quantitative radiology. The report will give you a number and that number might be a criteria of information, Rueckert suggested. Report: Mélisande Rouger Machine learning (ML) is becoming a valuable assistant for radiologists performing cardiovascular MRI. ML tools mainly help clinicians to acquire better images of the heart in shorter episodes of time, a key factor that can improve examination quality and increase patient comfort, according to Rueckert. ‘Typically, patients need to hold their breath several times while the examination is being per- formed. The sicker you are, the more complicated it is to hold your breath, and this may impact examination outcome. So here the machine simply helps you get the images faster and in a more comfortable way,’ he said. ML also enables functional param- eters measurement, for instance of how much blood the heart can pump per heartbeat. To do that, radiologists typically measure the heart when fully relaxed and when contracted. The difference indicates how much blood the heart can pump. The operation requires labelling every pixel in the image depending on which part of the heart is being observed, so one can actually com- pute that measure; but with machine learning, this can now be done auto- matically, Rueckert said. ‘You can effectively analyse the images and © Bai et al. automatically perform segmenta- tion. Then you show the clinician what segmentation looks like, and if they’re happy, they can accept that and get these measures automati- cally.’ For the moment, machine learning use is limited to assisting the radiolo- gist in doing ‘basic’ tasks. But this is just the bottom of the pyramid, said Rueckert, who used the reference to illustrate the machine’s contribution to medical imaging. ‘In the future ML may well help the radiologist to perform high-end tasks, for instance directly diagnose disease from an image. Research is underway in this field and applications are emerging, but it may be a while before they effectively serve the radiologist,’ he said. The solutions developed so far will not replace radiologists, but rather make things easier for them at work, he added. ‘A lot of what has been developed in recent years are techniques that would make working easier for the radiologist, but are not designed to replace the radiologist.’ Rueckert’s comments followed debates about radiologists potentially losing their jobs to ML and other AI will not replace radiologists This will help establish the prob- ability for a diagnosis, and may help clinicians to diagnose earlier and more confidently. Quantitative meas- ures may also help identify further conditions, which often go unnoticed when screening for a particular dis- ease, Rueckert added. ‘One thing that is quite overlooked is that it’s quite common for people to have a multitude of diseases; you often have a mix of things. Taking in a probability for a certain disease may be the right thing to do, because it can reflect that you have certain other diseases,’ he said. A myriad of potential applications could benefit from ML, not only risk identification and disease prediction, but also treatment guidance. ‘If you want to perform heart surgery you need imaging to guide you; so, ML Intelligence: augmented rather than artificial Artificial intelligence (AI) will increase efficiency and improve quality as well as clinical outcomes – and thus strengthen rather than weaken the role of radiologists, said Dr Joon Beom Seo at ECR 2018, as EH correspondent Michael Krassnitzer reports. A spectre is haunting radiologists – the spectre of artificial intelligence. Is AI about to replace radiologists? Wrong question,’ declared radiolo- gist Dr Joon Beom Seo, professor at the Artificial Intelligence for Medical Imaging Research & Development Centre, University of Ulsan in Seoul, South Korea, during his presentation at ECR 2018. He responded with an alternative question: ‘How can we improve and reposition radiology by integrating AI technologies?’ The term artificial intelligence in the context of radiology usu- ally describes applications based on deep learning: an artificial system learns from examples by autono- mously recognising patterns and structures. To date, such applica- tions are already available for all areas of image analysis, be it image conversion and segmentation of lesions and organs or the detection and classification of lesions. Seo cited some examples of image conversion applications: images acquired with a certain filter can be converted in a way that they seem to have been acquired with a different filter. While in low-dose CT deep learning applications can reduce noise; in Diffusion Tensor Imaging (DTI) they can increase image qual- ity such as spatial resolution from 2.5 mm to 1.25 mm and the number of gradients from 15 to 90. ‘Image conversion using deep learning methods is a promising area,’ Seo AI is a tool for the radiologist, not a role replacement pointed out, because it might, inter alia, solve the issue of heterogene- ous data in multi-centre studies. More automation = less time, less effort ‘AI-assisted segmentation of lesions and organs can already be car- ried out in a fully automatic way,’ the South Korean expert suggested. For example, the segmentation of organs in abdominal CT, or the y t t a b n a e S / y a b a x i P © detailed visualisation of the res- piratory tract in thoracic CT. ‘Deep learning methods display the organs as well as manual post-processing. However, the computer will do in two minutes what the radiologist does in two hours.’ Thus, AI offers a significant reduction in time and effort. ‘Introducing this technology into clinical routine will be the dawn of a new era,’ Seo said, with enthusiasm. The same, he believes, holds true for detection and classification of lesions. AI already delivers great results in the case of CT detection of lung nodules. This could also be introduced to clinical applications, said Seo: the computer might check the scans for suspicious areas and report them for review before the physician looks at the images. Nevertheless, at this point AI still operates in a rather restrictive envi- ronment, Seo emphasised, point- ing at political and legal concerns regarding privacy and commercial use, but also at technological issues. Most importantly, however, many radiologists are worried about the fact that deep learning systems are black boxes: ‘We do not know how a system arrives at its results,’ Seo explained. Professor Daniel Rueckert heads the department of computing at Imperial College London, which he joined as a lecturer in 1999 and became senior lecturer there in 2003. He has been professor of visual information processing and led the biomedical image analysis group since 2005. could help use imaging more appro- priately in this setting as well,’ he explained. Machines may not threaten radi- ologists, but their jobs will be trans- formed. ‘ML will probably make life a little bit more complicated for radiologists, who in the future may become something more like data scientists – people who need to interpret a lot of information from different sources, from imaging and maybe from non-imaging examina- tions – genetics tests for instance. So, their work might become more com- plicated, but technology will support them in that change.’ Radiologists will need proper train- ing to be able to use these tools, communicate with computer scien- tists and understand when to and not to trust the machine. ‘Clinicians need to understand what is being said and to know what they can rely on; when should they trust the machine and when it is just an opinion. The machine cannot give you a definite answer but it can help in case of doubts.’ l u o e S e r t n e C l a c i d e M n a s A f o y s e t r u o C Joon Beom Seo MD PhD, is professor at the University of Ulsan College of Medicine in Seoul, South Korea, and attending physician in radiology at the Asan Medical Centre in Seoul. Who teaches the machine? Currently, AI applications in medical imaging are still based on super- vised learning processes, i.e. human specialist input is necessary. This raises the issue of who should or may train the systems. ‘If you look for example at the detection of dia- betic retinopathy you will see that there are large interobserver differ- ences,’ he said. Which goes to show that even a deep learning system depends on the quality of the data it uses to learn. ‘Artificial intelligence won’t replace radiologists, but rather strengthen their role by improv- ing quality and clinical outcomes and increasing efficiency,’ Seo said, returning to the initial question. AI thus is a tool not a replacement of the radiologist: ‘We should,’ he concluded, ‘call it augmented rather than artificial intelligence.’ EUROPEAN HOSPITAL Vol 27 Issue 4/18

1 0 CA R D I O LO G Y Intracardiac echocardiography (ICE) has benef Refined guidance with no need for anaesthetic Intracardiac echocardiography (ICE) is an increasingly important guiding tool for structural heart disease interventions – without gen- eral anaesthesia. José Ribeiro, who works in the thorax and circula- tion unit at Gaia Hospital Centre, Portugal, who has worked with this technology for the past two years, explained its benefits and limitations in an exclusive interview with Daniela Zimmermann of European Hospital. Discussing developments in Intracardiac echocardiography (ICE), José Ribeiro, cardiologist at the thorax and circu- lation unit in Gaia Hospital Centre, Portugal, explained that recently the need for a different ultrasound tool to guide patient treatment beyond transoesophageal echocardiography (TEE) became clear. ‘Consequently,’ he added, ‘a significant number of interventional cardiologists have start- ed to use ICE. ‘We still have limitations with ICE for structural heart disease, because we don’t see all the structure in the same plan and need to navigate with a catheter inside the heart. That’s why it’s so important for 3-D imaging to guide procedures. ‘We don’t need too much imaging to guide the intervention for struc- tural heart disease. But we need to have a good pre-procedure evaluation and to plan the procedure, and after that we only need specific steps to ensure procedure quality and check the results. If we can get the cardiac structures on 3-D, we have a signifi- cant advantage for guidance. ‘We initially had a narrow angle catheter, which only enabled us to view small volumes of the heart. This is not enough to image entire structures, for instance a valve, left appendage or oval fossa. So we devel- oped a new device with a wide open- ing angle; it’s a 12.5-Fr catheter and this enables us to view significant vol- umes of the cardiac tissue, including the whole mitral valve. This develop- ment appears to be a great advantage for guidance.’ DZ: When using ICE, is the image in front of you and can you see the relation between the structures? ‘Yes. When we have a volume, we can look inside and decompound it in a different 2-D plan to navigate more easily, which enables us to be more confident when doing the procedure. ‘With a 3-D wide angle ICE cath- eter, we have the same benefits as with 2-D ICE, meaning we don’t need an anaesthetist, the interventional car- diologist can do the intervention him- or herself, by putting the catheter in the right place to see the heart. ‘But we can also obtain a signifi- cantly higher amount of information and anatomy, so that the interven- Images of intracardiac echocardiography obtained with Acunav V wide angle catheter (Siemens Healthineers); on 3D images we can see entire cardiac structures: on top right the fossa ovalis, on top left the left atrial appendage opening (LAA), on bottom right the mitral valve with anterior (AL) and posterior (PL) leaflets and on bottom left the device occluding LAA. More people need nuclear cardiology training Ischaemia: Advances in nuclear imaging Experts outlined approaches to ischaemia imaging during the recent British Cardiovascular Society confer- ence. In a ‘Detection of ischaemia by cardiac imaging in 2018’ session, comparisons were made between solid state SPECT cameras, whether spatial resolution or visual assessment was of the greater importance, if CT-FFR offered advantages over CT perfusion, and the challenges in defining a gold standard of imaging ischaemia Discussing ‘Advances in nuclear ischae- mic testing, from SPECT to PET and beyond’, Dr Kshama Wechalekar, who heads Nuclear medicine and PET at the Royal Brompton Hospital in London, and is President of the British Nuclear Cardiology Society (BNCS), told delegates that advances in nucle- ar imaging with solid state technol- ogy offered improved ability to detect ischaemia. ‘There is improved spatial resolution from multiple solid state CZT (Cadmium Zinc Telluride) detec- tors and therefore sensitivity is very high,’ she explained. ‘You can reduce the acquisition time at least by half President of British Nuclear Cardiology Society Dr Kshama Wechalekar leads nuclear medicine and PET at the Royal Brompton Hospital in London, where she specialises in heart/lung nuclear imaging. Her main interests lie in using hybrid- imaging techniques, such as SPECT-CT and PET-CT, to improve understanding of pathophysiological processes affecting the heart and lungs. She has special interest in cardiac sarcoidosis and other inflammatory conditions of the heart. with excellent quality and the equip- ment has a small footprint. The advan- tages of solid state detector cameras is that you can reduce the radiation dose by one third, have high sensitivity and resolution, an open design suitable for claustrophobic patients, and good image quality even in obese patients.’ SPECT, PET and CMR Recent studies have shown ability to do dynamic imaging offering potential in quantitative myocardial perfusion with SPECT, Wechalekar pointed out, adding that SPECT is less expensive than current PET and MRI. ‘The future of SPECT Nuclear car- diac imaging,’ she concluded, ‘is in solid-state technology. Dynamic imag- ing, although technically challenging, can add value to MPI in the detec- tion of ischaemia. Whilst PET is the most accurate imaging technique for ischaemia assessment and prognosis, it remains expensive and less acces- sible.’ She also felt that the new trac- er, Flurpiridaz, with results of phase III trials in the UK pending, might change the future of PET MPI. One area of concern was how to persuade more people to train in nuclear cardiology with falling num- bers in the field. ‘The BNCS Council is working hard to improve curricu- lum, organise level 1 and 2 training courses, and to identify centres that can offer nuclear cardiology training across the country that is easily acces- sible for trainees,’ she said. Dr Chiara Bucciarelli-Ducci, in Lecturer Consultant Senior Dr Chiara Bucciarelli-Ducci is Consultant Senior Lecturer in Cardiology/non Invasive Imaging at the Bristol Heart Institute, University of Bristol, and co-Director of the Clinical Research and Imaging Centre (CRIC Bristol). She is currently one of the vice-presidents and chair of cardiac MRI of the European Association of Cardiovascular Imaging (EACVI). Invasive Cardiology/non Imaging Bristol Heart Institute, University of Bristol, explored the issue of quan- titative versus visual assessment in CMR stress perfusion. She explained that stress CMR has been included in the ESC guidelines since 2014 (ESC revascularisation guidelines) based on evidence using visual assessment of ischaemia, rather than quantitative. Bucciarelli-Ducci discussed pros and cons of both visual and quan- titative assessment, limitations and opportunities to increase spatial reso- lution, and very recent studies show- ing that there is no difference in diag- nostic accuracy visual vs. quantitative. Quantitative perfusion is promising, but the acquisitions and analysis need Single photon vs. Positron ET simplification to meet the need of a busy clinical service. ‘CMR perfusion (visual) is a good clinical tool already,’ she concluded, ‘but can get better while quantita- tive CMR perfusion is evolving into faster and robust tools. While several methods are available, more in-vivo and clinical validation is needed with a number of studies in the pipeline.’ Function addition can improve specificity Dr Marc Dweck, BHF Reader in Cardiology and Consultant Cardiologist at the University of Edinburgh and the Edinburgh Heart Centre, posed the question ‘CT-FFR/CT perfusion - nei- ther or both?’ ‘CT Perfusion,’ he acknowledged, ‘is interesting, but I’m not sure how we are going to use it in clinical practice. With CT-FFR you get beautiful pic- tures, where you can look down the coronary arteries and see areas that are not getting enough blood. The advantage of this technique is that you can use it on a post-hoc basis, on scans where you are not sure if a lesion is obstructive or not, without any extra radiation or medication for the patient. This may be useful in low- ering the rates of patients being sent EUROPEAN HOSPITAL Vol 27 Issue 4/18 Dr Marc Dweck is Senior Lecturer and Consultant Cardiologist at the University of Edinburgh and the Edinburgh Heart Centre. A British Heart Foundation Intermediate Clinical Research Fellow, he is a keen advocate of multi-modality cardiovascular imaging and is trained in echocardigraphy, carotid ultrasound, computed tomography (CT), cardiovascular magnetic resonance (CMR), PET/CT and PET/MR imaging.

1 2 CA R D I O LO G Y Coronary angiography will lose diagnostic value The changing face of imaging in cardiology While the question is still debated as to whether MRI is the better CT, along comes a potential game changer – a new data based 3-D reconstruction method of heart anatomy and function that aims to replace diagnostic coronary angiography. In the near future not only adult patients with coronary heart disease could benefit from this new technique but also children with complex congenital heart defects. Meanwhile imaging is conquering the cardiac operating room (OR). Report: Emilie Hofstetter Long before coronary heart disease (CHD) manifests its presence on an ECG, CT and MRI can detect it due to low perfusion caused by a stenosis of the coronary vessels. Dr Bettina Baessler, radiologist and researcher at the University Hospital Cologne, Germany, looks into mul- tiparametric imaging strategies. She considers both techniques comple- ment one another although MRI definitely produces images that are ‘more beautiful, almost works of art’. Professor Ulf Teichgräber, Head of Radiology at University Hospital Jena, Germany, agrees and thus predicts the demise of cardiac angi- ography. His opinion is corrobo- rated by the recently completed SYNTAX III study, whose results will be presented at the Transcatheter Cardiovascular Therapeutics From Vision to Impact. That’s Excellence for Life. From First German Pacemaker to Connected Cardiac Care. biotronik.com flow and can thus show whether a haemodynamically relevant block- age is present, i.e. whether the patient needs a stent or a bypass. In 2015, Professor Pamela S Douglas, cardiologist and Head of Multimodal Imaging at the Duke Clinical Research Institute in Durham, North Carolina, USA, showed the potential benefit of this method using 584 patient cases from 11 hospitals. Ten patients with suspected CHD underwent diagnos- tic cardiac catheterisation, but the suspicion was confirmed only in three patients – seven underwent unnecessary catheterization. Six out of ten patients with suspected CHD, whose FFRCT was determined first, did not need angiography. In three out of the four patients who did receive angio, the suspicion was confirmed – i.e. only one patient underwent an unnecessary angiog- raphy. ‘This feasible and safe meth- od shows a significantly lower rate of unnecessary invasive angiogra- phies,’ Douglas confirmed. Investors seem to buy in: HeartFlow, which today is already cooperating with the Big Three – GE, Siemens, Philips – recently raised USD 240 million to further develop the technology, launch new studies and drive com- mercialisation of its product. To establish 3-D imaging in con- genital heart disease treatment, paediatric cardiologists Animesh Tandoon and Tarique Hussein founded VARYFII Imaging, LLC, in Dallas, USA. They construct complex anatomical models of the individual patient’s pathologies using MRI or CT data. Cardiologists as well as sur- geons can enter the virtual and aug- mented realities of the anatomical models with the help of data head- sets to lift certain structures, analyse and reposition and thus devise the best strategy them 3-D reconstruction of a coronary system, detached from the heart muscle. Fractional flow reserve in the individual vessel sections is colour-coded. Courtesy: HeartFlow, Inc. Y-conduit of right and left internal thoracic artery in epicardial ultrasound. A: 2-D Long axis view, B: 2-D short axis view, C: Colour Flow Mapping long axis, D: Colour flow short axis. Courtesy: Di Giammarco to correct the heart defect prior to surgical intervention. ‘Our heart beats in 3-D, so why not examine it in 3-D?’ asks Dr Sandy Engelhardt, researcher at the Computer-Assisted Surgery Group at the Department of Simulation and Graphics in Otto von Guericke University, Germany. In addition to treatment planning and education she envisages a further application of this new technology: informing the parents of the young patients. The flow must continue Imaging has arrived in cardiac surgery – during the intervention itself and combined with flow measurements. Professor Gabriele Di Giammarco, cardiac surgeon at Gabriele D’Annunzio University Hospital in Chieti, Italy, considers the combination of high-frequency epicardial ultrasound (ECUS) and transit flow measurement (TTFM) in a single device ‘deci- sion making’ and explains: ‘Hard calcifications in the aorta, I can feel. I do not feel the dangerous soft plaques. With MiraQ, I see them in intraoperative ultrasound, can adapt my strategy and perform surgery in no-touch technique and off-pump.’ time Dr Daniel Wendt, Managing Senior Physician at the Cardiac Surgery Department of University Hospital Essen, Germany, uses intraoperative flow measurement of newly cre- ated bypasses not only for quality assurance purposes – he records a follow-up intervention rate of slightly below three percent – but also for training purposes: ‘It’s a tool to improve your skills, helps flat- tening the learning curve.’ The combination of risk minimisation and qual- ity assurance has proved suc- In cessful. 2017, the Oslo-based m a n u - f a c t u r e r M e d i s t i m sold prod- ucts and procedures worth NOK 229.8 million, up 14.6 percent over the previ- ous year – another chapter in the success story of surgical interven- tion in CHD patients. EUROPEAN HOSPITAL Vol 27 Issue 4/18 The physician, using smart glasses, in the virtual space has just removed the aorta at its root from the heart to examine it separately. Observers can follow on conventional screens. Courtesy: S. Engelhardt Symposium 2018 in San Diego in September. A team comprised of a radiologist, cardiologist and sur- geon (Heart Team A) evaluated the angiogram of a patient, calculated the SYNTAX II score and decided on the type of therapy, either invasive or non-invasive. The team members then saw the multislice CT (MSCT) scan with 3-D reconstructed coronary vessels and the relevant fractional flow reserve (FFRCT) and could either confirm or revise their decision. A second team (Heart Team B) of those pro- fessionals received CT and FFRCT of the same patient first. The team members calculated the Syntax III score, decided on the type of ther- apy and then saw the angiogram in order to either confirm or revise their decision. ‘The Syntax score was designed to inform the decision “invasive or non-invasive”, based exclusively on anatomical features,’ Teichgräber explained. ‘Syntax II took comorbidities into account and now Syntax III includes a functional component – FFRCT. Thus coronary angiography will lose importance in diagnostics and therapy planning.’ Non-invasive first To date, only the California-based HeartFlow Inc. can calculate FFRCT. Based on data obtained in a conven- tional CT, the company’s software, using flow mechanics, can recon- struct heart, aorta and coronary ves- sels in terms of geometrics as well as pathophysiology and function in 3-D. Moreover it visualises the

1 4 CA R D I O LO G Y Arrhythmia news from CMR 18 Multidisciplinary cardiology Eminent Spanish cardiologist highlights the evolving role of imaging in ventricle arrhythmias treatment Report: Melisande Rouger Intervention in ventricular arrhyth- mia has improved dramatical- ly over the past three decades thanks to advances in imaging and co operation between cardiology and radiology, according to Professor Josep Brugada MD, director of the paediatric arrhythmia unit at Sant Joan de Déu Hospital in Barcelona. information from imaging. In the late 1990s, CT enabled the visualisation of more complex arrhythmia substrates, such as ven- tricular tachycardia. ‘It was not enough to use echocardiography anymore. At that time we needed to have information on the position and size of different structures, for instance the pulmonary veins. We used the CT scan to help us to do anatomy of pathways in the muscle is crucial at the time of eliminating arrhythmia, Brugada explained. ‘Using MRI, to extract anatomic information, and our mapping sys- tem, which enables us to identify where the electricity goes through, we can determine the exact point that may be responsible for arrhyth- mia. We can then target this point with radiofrequency and burn it, l e d o p x E / m o c . k c o t s r e t t u h S © ‘Echocardiography, CT and MRI, combined with cardiology,’ he said, ‘have revolutionised the field into what it is now a truly multidiscipli- nary field.’ Cardiologists have learned to use echocardiography. Collaboration with radiologists has enabled them to understand ventricular tachycar- dia and ventricular fibrillation, two life-threatening diseases that are most commonly associated with heart attacks or scarring of the heart muscle from previous heart attack. As a result, knowledge of physio- pathology and arrhythmia mecha- nisms has grown over the past thirty years, first with understanding the location of arrhythmia, a parameter that remains crucial for treatment. Decades of developments ‘Thirty years ago, we were perform- ing diagnostic electro-physiological studies, and at that time our rela- tionship with imaging was very sim- ple – we just used X-ray. But in the late 1980s, we understood that we could start treating these patients and that we could start targeting the points that we identified as the points of arrhythmia, and with more and more accuracy to under- stand where these points were,’ he explained. Using X-ray and echocardiog- raphy combined with the electric signal first enabled very simple arrhythmic substrates to be located. Cardiologists soon understood that they needed, and could get, more just that. That was a very important step,’ Brugada pointed out. Inside cardiac tissue Ten years later, around 2010, the medical team realised that they had to go even further in their under- standing of arrhythmia, beyond ana- tomic knowledge; they also needed to know what was inside the heart tissue. ‘We wanted to see how the tissue is built in the heart, to assess tissue structures and understand how dif- ferent elements of cardiac tissue can be identified as normal or abnormal areas. This can only be done with nuclear magnetic resonance imag- ing. Only then did we begin to understand what the true nature of the tissue is,’ Brugada said. MRI enables us to depict details in both ventricles. Checking the left ventricle, especially, remains crucial because of the various potential origins of arrhythmia. ‘If you have a myocardial infarction, we know the abnormal tissue can be located in the endocardium site and even the mid-myocardium site. So you need to see the structure of this tissue. That’s why all the efforts are now focused on understanding how the tissue is structured and built in the ventricles.’ MRI provides information on the nature of tissue, and, used in combi- nation with electrical mapping sys- tem, helps identify electrical path- ways or channels that may cause arrhythmia. Unveiling the electrical and thus prevent the electricity going through the pathways that cause arrhythmia.’ Major contribution MRI is also used to guide interven- tion and correct potential error; this is a new and major contribution of imaging to arrhythmia treatment. ‘You need to integrate this imag- Editor-in-Chief: Brenda Marsh Art Director: Olaf Skrober Editorial team: Wolfgang Behrends, Sascha Keutel, Brigitte Dinkloh, Marcel Rasch Senior Writer: John Brosky Executive Director: Daniela Zimmermann Founded by Heinz-Jürgen Witzke ISSN 0942-9085 Correspondents Austria: Michael Kraßnitzer, Christian Pruszinsky.China: Nat Whitney France: Jane MacDougall. Germany: Anja Behringer, Eva Britsch, Annette Bus, Walter Depner, Emelie Hofstetter, Cornelia Wels-Maug. Great Britain: Session highlights cardiac arrest in w MRI has a central role in picking up myocardial infarc nary disease, a condition that particularly af with potentially fatal outcome. Heart attack in women presents dif- ferently than in men and requires a different approach when it comes to detection and prevention, accord- ing to Allison Hays, a cardiolo- gist and assistant professor at the Johns Hopkins University School of Medicine, speaking at CMR 2018 meeting in Barcelona. ‘Women don’t present typically with chest pain, but rather with tiredness or shortness of breath. When they eventually come in for diagnosis, women have much less rates of having abnormal cardiac catheterisation test, which shows degree of stenosis in coronary arter- ies. So, much more commonly, they Allison Hays MD is a general cardiologist and interim Director of Echocardiography Programs at the Johns Hopkins Heart and Vascular Institute. She is also an assistant professor at the Johns Hopkins University School of Medicine. Having graduated from Stanford University she received medical training at the Columbia University College of Physicians and Surgeons. Following her residency at the New York Presbyterian Hospital at Columbia Hays pursued cardiology fellowships at New York University Medical Center and at the Johns Hopkins University School of Medicine. Today she studies ways to use non-invasive imaging to detect cardiovascular disease. In terms of research, Hays uses cardiac MRI as a tool to study coronary and systemic endothelial function. Professor Josep Brugada studied medicine at the University of Barcelona, Spain, before moving to the University of Montpellier, France, to specialise in cardiology. He continued training in Maastricht, Netherlands, to specialise further in the clinical and basic aspects of cardiac arrhythmias, and he worked simultaneously in the basic and clinical electrophysiology laboratory. Brugada then became Assistant Professor at the University of Limburg and was the first foreign established investigator of the Dutch Royal Academy of Arts and Sciences. In 1991 he returned to the hospital at the University of Barcelona, where he became professor of medicine, head of the arrhythmia unit, and of cardiology, director of the thorax institute and finally medical director. He is now professor of medicine, director of the paediatric arrhythmia unit at Sant Joan de Déu Hospital in Barcelona. ing and perform 3-D imaging of the heart because you want to see exactly what is happening,’ Brugada explained. ‘That’s a technique we use every day, and increasingly to treat atrial arrhythmia in atrial fibril- lation,’ The main interventional treatment of AF consists in blocking the pul- monary veins using radiofrequency, i.e. by burning around the veins to create a line of electrical block. During the procedure, gaps may occur in these lines of block; these breaches can only be spotted on MRI. Being able to pick these gaps will lead to redoing the procedure and improving treatment success and patient outcome. In the future, MRI will continue to improve treat- ment efficiency in AF, a disease that affects millions of patients world- wide, and ventricular tachycardia patients, whose only chance of sur- vival is to receive ad equate treat- ment. ‘There are fewer cases of ventricular tachycardia than AF, but these are very severe patients who need a cure. It is fundamental for their survival and the only thing we can do right now is to target the right electrical pathways.’ Brenda Marsh, Mark Nicholls. Malta: Moira Mizzi. Spain: Mélisande Rouger, Eduardo de la Sota. The Netherlands: Madeleine van de Wouw. USA: Cynthia E. Keen, i.t. Communications, Lisa Chamoff. Subscriptions Liane Kaiser, Theodor-Althoff-Str. 45, 45133 Essen, Germany Subscription rate 6 issues: 42 Euro, Single copy: 7 Euro. Send order and cheque to: European Hospital Subscription Dept Printed by: WVD, Möhrfelden, Germany Publication frequency: bi-monthly Representatives China & Hongkong: Gavin Hua, Sun China Media Co, Ltd. Phone: +86-0755-81 324 036 E-Mail: gavin_hua@163.com Germany, Austria, Switzerland: Ralf Mateblowski Phone: +49 6735 912 993, E-Mail: rm@european-hospital.com France, Italy, Spain: Eric Jund Phone: +33 493 58 77 43, E-Mail: ej@european-hospital.com GB, Scandinavia, BeNeLux: Simon Kramer Phone/Fax: +31 180 6200 20 E-Mail: sk@european-hospital.com Israel: Hannah Wizer, International Media Dep. of El-Ron Adv. & PR Co., Ltd., Phone: +972-3-6 955 367 E-Mail: hw@european-hospital.com South Korea: CH Park, MCI Phone: +82 2 730 1234, E-Mail: mci@unitel.co.kr Taiwan: Charles Yang, Phone: +886 4 232 236 33, E-Mail: medianet@ms13.hinet.net USA & Canada: Hanna Politis, Media International Phone: +1 301 869 66 10, E-Mail: hanna@media-intl.com All company, brand and product names in this publication are the property of their respective holders. Users must obtain permis- sion from those holders before copying or using the owner’s trademarks, product and company names or logos. EUROPEAN HOSPITAL Vol 27 Issue 4/18

1 6 CA R D I O LO G Y HemoDynamic Analytics in Ultrasound A new tool box enhances heart failure diagnosis One of the challenges for every echo- cardiography lab is the technically difficult patient. Conventionally, labs use contrast agents to enhance endocardial border visualization. The application of contrast agents increases the exam time, resources and costs. Additionally, the use of contrast turns a previously non- invasive exam into an invasive pro- cedure. Hitachi Healthcare has now devel- oped a collection of cardiovascular analytic tools called HemoDynamic Analytics (HDAnalytics). These tools can be used for evaluation of the left ventricle (LV) when visualization is limited. One of the main applica- tions of the collection, LV eFlow was designed to demonstrate the discrimination between the blood flow and the cardiac tissue and offer an alternative to contrast agent use in some cases. LV eFlow is a high-definition left ventricular cavity blood flow imaging mode which substantially improves spatial and temporal reso- lution for a better visualization of the endocardial border in the left ventricle. The new tool operates with higher sensitivity and resolu- tion than conventional methods. LV eFlow may change a technically dif- ficult study into a diagnostic exam without using contrast agent. Head-to-head with echo contrast Dr. Zuyue Wang and technolo- gist Marvin Tyson of MedStar Washington Hospital Center had an opportunity to use this tech- nology in their practice over a period of 3 months. Their protocol included identifying patients that were candidates for contrast agents due to the difficulty in visualizing the endocardial border of the left ventricle. LV eFlow was added to the exam protocol for this patient set. Following the exam, the quality of the endocardial border delineation was evaluated by comparing the LV eFlow images with the images using contrast agents. Dr. Wang and Marvin Tyson com- pared LV eFlow and echo contrast agents in patients with suboptimal image quality and were impressed with the results. They found that “LV eFlow was comparable to echo contrast in improving visualization of difficult-to-image segments in selected patients”. Additionally, they found “a markedly more precise endocardial border delineation” and stated that “contrast agents should only be utilized when LV eFlow fails to enhance the endocardial borders”. Left: LV eFLOW – technically difficult patient; right: VFM – relative pressure with a dilated cardiomyopathy. Source: Hitachi Medical Systems Europe Vector Flow Mapping & Dual Gate Doppler Another tool in the HDAnalytics collection is Vector Flow Mapping (VFM), a novel and validated appli- cation that allows users to assess cardiovascular blood flow distribu- tion in an observation plane. This non-invasive technique is derived from the Color Doppler velocity data and generates the velocity fields on the 2D image. This allows to visual- ize, measure and analyze different parameters from the blood flow dis- tribution. For example, energy loss which is the rate of energy dissipa- tion due to blood viscosity, increas- es where turbulence flow occurs. In addition, wall shear stress, relative pressure and vortex characteristics can be evaluated. The Dual Gate Doppler (iDGD) generates a full Fast Fourier Transform (FFT) analysis and dis- play from two separate sample gates allowing measurements from two different locations during the same cardiac cycle. Hitachi Artificial Intelligence technology enables automatic sample gate placement and measurement at appropriate heart beats, resulting in 5 seconds to get E/e’ (83% shorten time com- pared with conventional measure- ment). Furthermore, iDGD works well for PW/PW and TDI/TDI com- binations. 2D tissue tracking (i2DTT) With 2D Tissue Tracking (i2DTT), the HDAnalytics set also provides an advanced tool which allows users to track the displacement of the cardiac tissue by using a novel and accurate algorithm of “Speckle Tracking”. Doppler based methods such as TDI are limited in evaluat- ing the displacement velocity of the tissue due to angle dependency. i2DTT allows the detection of veloc- ity components perpendicular to the beam which is impossible with conventional Doppler techniques. Tracking image by image, the natu- ral patterns of the cardiac tissue in B-Mode permits the user to quanti- tatively evaluate the movement and the thickening of the myocardium. i2DTT provides precise quantitative information such as longitudinal and radial strain, torsion rotation angle, displacement, wall thicken- ing and various other parameters to visualize, quantify and analyze myocardial mechanics. Applications include cardiac function analysis, resynchronization therapy, cardio- myopathy, stress echo and other global and regional studies. New analysis suggests workflow is key in remote monitoring It’s time to look again at IN-TIME As the world’s largest cardiology con- gress gets underway in Munich, it’s worth looking back to previous ESC sessions to see how scientific debates have evolved. At ESC 2016, held in Rome, REM-HF investigators presented data suggesting remote monitoring in implantable cardiac devices offered no added clinical benefit. Two years on, there are new reasons to re-examine that conclu- sion, with a recent analysis of the IN-TIME trial suggesting the key to remote monitoring benefits might be found in workflow processes. Published in The Lancet in 2014, the IN-TIME study is the only trial, to date, to have demonstrated a clear benefit of implant-based remote monitoring in heart failure (HF) patients – showing a more than 50% reduction in all-cause mortality – while eight other studies included in a 2015 meta-analysis, and three other recent trials, found no signifi- cant clinical benefit. However, IN-TIME was also the only implant-based remote monitor- ing trial using a transmission tech- nology that sent daily updates to a central monitoring unit. By contrast, REM-HF used technology that trans- mitted implant data on a weekly basis. So why does IN-TIME show positive results when other remote monitoring studies don’t? What is so fundamentally different in its methodology that might account for its results? Multiparametric data in This the recently published Remote is a central question Monitoring and Clinical Outcomes: Details on Information Flow and Workflow in the IN-TIME Study by Husser et al. The authors note that IN-TIME featured multiparamet- ric data that was transmitted daily. Crucially, a workflow process was set up such that study investigators could typically contact patients less than a day after receiving an event alert and arrange any necessary follow-ups for less than a week later. The study authors point out that, in the recent TRUECOIN meta- analy sis, the IN-TIME approach was shown to be beneficial for patients with heart failure, since it provides early enough warning to potentially prevent deterioration in the patient’s condition due to new onset atrial fibrillation, asymptomatic ventricu- lar tachycardia, or other adverse events. It is this early appraisal – facilitated by efficient workflow processes, including multiparamet- ric daily transmissions — that make the difference in the IN-TIME study, authors argue. As the European cardiology com- munity gathers for ESC 2018, it’s an excellent time to re-examine exist- ing evidence for clues we may have missed, alongside the latest break- ing research. That’s why it’s time to look again at IN-TIME. EUROPEAN HOSPITAL Vol 27 Issue 4/18 Home Monitoring Service Center Central Monitoring Unit Clinic Patient BIOTRONIK ICD/CRT-D Prompt in-office FU if interview is ‘positive’ (asap) Daily automatic transmission (Mo-Su) Prompt patient contact if IN-TIME event is clinically relevant (asap) Daily HM alert evaluation and classification as IN-TIME events (Mo-Fr) Device diagnostics (last 24h) HM alerts IN-TIME Events Structured Interview Immediate automatic data evaluation (Mo-Su) IN-TIME 365 messages 83% - 94% received after 1 - 3 day(s) n.a. 100% of HM alerts notified on the same day 4.0 In-Time events n.a. Until next working day 2.1 contacts n.a. Median 1 day [IQR 0 – 5] ACTIVITY WORKFLOW PERFORMANCE Events per patient year Success Delay IN-TIME Workflow Performance Less than one week

1 8 L A B O R AT O RY Vital vitamin D testing LC-MS outperforms immunoassays In recent years, clinicians have increasingly focused on vitamin D deficiency. Studies show that previous reference values – particularly for Vitamin D3 – were most probably set too high. Liquid chromatography with mass spectrometry (LC-MS) can help achieve more precise measurements of vitamin D levels than previously established immunoassay procedures, explains Dr Torsten Binscheck-Domass, formerly a phar- macologist and toxicologist at the joint laboratory of the Charité and the Vivantes Group in Berlin, and an expert in clinical mass spectrometry at Thermo Fisher Scientific. However, in its current shape this com- plex technology is not suitable for all laboratories. Report: Daniela Zimmermann Although the relevance of the thresh- old values remains controversial, there is agreement on one point: Vitamin D deficiency is associated with a range of clinical pictures, such as osteomalacia or respectively, rickets in children, as well as with increased susceptibility to infections. In the aged, a lack of vitamin D can also have severe consequences, as Binscheck-Domass explains: ‘Studies are currently being carried out to examine a correlation between vita- min D deficiency in older patients and osteoporosis as well as a reduc- tion in muscle strength, which leads to an increased risk of falls. For older patients, falls are complications that can have severe consequences, such as fractures, traumatic brain inju- ries and long-term hospitalisation, etcetera.’ UV-radiation from the sun is a nat- ural remedy, but is often not enough. ‘We recommend vitamin D screen- ing for people with little exposure to sunlight, particularly for vitamin D3,’ he adds. Basically, everyone whose face, hands, and ideally also lower arms, are exposed to the sun for at least 30 minutes twice a week should absorb sufficient vitamin D3 to also last them through longer periods without sunlight. However, deficiencies are documented for hos- pitalised patients, people in nursing homes and sometimes even for chil- dren and adolescents who spend lit- tle time outdoors. Vitamin D2, which is mainly absorbed via nutrition – such as edible mushrooms – plays a slightly less important role. LC-MS separates the wheat from the chaff One problem with measuring vita- min D levels in the body is that the different intermediate products are often very similar and cannot be dif- ferentiated correctly with antibody- based tests such as immunoassays. Together with the comparatively low concentration of calcitriol – the bio- logically active form – this often leads to measurement values that have little significance, the expert points out. LC-MS has a big advan- tage here as it can separate the bio- logically inactive epimers from the molecule of the vitamin. Therefore, LC-MS-based diagnosis achieves sig- nificantly higher specificity. Currently there are still hurdles which laboratories need to over- come when switching from immuno- assays to LC-MS, Binscheck-Domass explains; this centres less around the associated costs – the acquisi- tion requiring investments in the mid-six figure range, depending on the device – but rather the changes to the normal work flows: ‘These are complicated devices and you need trained staff who can not only oper- ate the equipment but also carry Torsten Binscheck-Domass MD is a Systems Lead Scientist in Clinical Mass Spectrometry at Thermo Fisher Scientific. Qualified in medicine, he is a medical and forensic expert consultant in analytical pharmacology and toxicology. He has more than 25 years of experience in the fields of therapeutic drug monitoring, drugs of abuse testing and systematic toxicological analysis. Dr Binscheck-Domass is a strong advocate of clinically applied LC-MS/MS and has been exploring the technology’s unique benefits to clinical labs for more than 10 years. out evaluations.’ Furthermore, some laboratories are deterred by regula- tory requirements because the quali- fication of the systems requires a lot of effort, with numerous calibration and test runs. Selection criteria for a laboratory information system Marry (a LIS) in haste, repent at leisure standard operating procedures (SOPs) they can visualise the spe- cific workflow of core processes to ensure relevant standards are met. Laboratorians should have easy access to quick visual overviews Buying a laboratory information system (LIS) means entering a long- term relationship with a software vendor. The selection criteria are many, but which, ask Markus Neumann, Harald Maier and Gabriele Egert, are just fashionable and which might be underestimated? donations, or for own blood dona- tions, are less well known. Generally, the trend is towards process monitoring. If a lab has The decision to buy a LIS – i.e. to form a relationship with one or more software vendors – is based on a slew of criteria, and excitement occurs over a long-term relationship. However, years pass before quality can be assessed. Hard economic, technical and functional criteria can easily be expressed in figures, whilst others are ‘soft’ and difficult to quan- tify; however, they should provide sound information to shape qual- ity in vendor/user cooperation. A purchasing decision results from balancing hard and soft factors. Specifications The basis of a specifications docu- ment to be provided to bidders for the tender is what’s required. If there is no in-house IT specialist, use of an IT consulting firm might be advisable. The checklist offers an initial overview of topics to be covered. Vendors invited to submit a quotation need careful selec- tion. Vendor criteria e.g. finan- cial position, years of continuous market activity and responsibility/ accountability. Visits to reference labs and infor- mation exchange with users can offer important insights. Beware: although criteria are important, they are no guarantee that soft- ware development will continue. Technical criteria Hard data concern software, hard- ware and database structure. Which operating systems, programming languages and web technologies will be used? How does the data- base perform under heavy work- load? An important question is user autonomy. Can you customise parameters, e.g. database queries? What training is included; what cost? Finally, physical interfaces support must be ascertained: Are popular interface protocols, e.g. HL7/IHE or ASTM, used and how are they practically integrated within the lab instruments? For vendors, programming customised device interfaces is a core busi- ness; over years this can amount to several thousand euros per ana- lyser – a significant position in the budget. Interfaces are also relevant in background processes, such as billing and controlling, or external communication with physicians to exchange order data and results. Functional criteria Every vendor offers many mod- ules listed in the table under ‘Functionality’. However, it makes sense to look at procedures and review their control by master data and parameters. Nowadays, many lab workflows must meet certain standards. Quality management standards, or ISO 15189, 17025, or even ISO 22870 for POCT, are pretty obvi- ous; others, such as GMP for blood Company Product Functionality Integration Results Feature Size, financial position Continuity, accountability/responsibility Reference customers Customer testimonials Soft skills Implementation concept Service Training Architecture Hardware Software Response time Maintenance contracts License policy Modules Data protection/privacy concept Tracking Compliance Master data management Order entry and capturing results Quality control Technical and medical validation Reading Billing Archiving Statistics Configurability Internal communication External communication hardware Existing data Costs Own contributions Costs over five-year period Continued on page 20 Checklist for a LIS selection Explanation Number of permanent and freelance staff, revenue, core business Type of company, year of incorporation, guarantees, ISO certification Number, size, scope, reputation Adherence to deadlines and delivery schedules, keeping promises, staff commitment, cooperation with instrument manufacturers Personality and competence of vendor’s staff during conversation Is it comprehensible? Volume, solution visualisation Hotline, response time, specialist availability Documents, online support, on-site training, seminars Client-server architecture, periphery, central master data server Central computer, PC network, terminals, printers, storage space, operating system Data base, data protection/privacy concept and confirmations, use of standard software, customisation of parameters (by user) Depending on database volume, number of web connections and online users (particularly in multi-site facilities) Type and scope, hardware, software, remote access Online devices, database, multiple installations (campus license), pay-per-use (instead of purchase) Clinical chemistry, microbiology, blood bank, pathology, billing, order entry, automation and control Permission-based, controlling access to patient data (e.g. lab values) Search and sort functions, order status and sample tracking RiLi-BÄK (in Germany), ISO 15189, ISO 17025, ISO 22870, GMP, etc. Data maintenance, parameters, copying, use of legacy data, across modules Keyboard, card reader, order entry, material ID, different bar codes, device interfaces RiLi-BÄK (Germany), Westgard, other quality systems, visualisation Auto-validation, rules, alarms, automated repeat measurements, visualisation Sort criteria for orders and lab, rules systems, ICD codes, footnotes in identical texts Multi-client capability, service portfolio, accrued accounts Procedure, duration, access, reactivation, archiving of image data Saved and spontaneous queries, SQL assistant, export to standard software Adaptation of software and user screens to the on-site situation Link to HIS and departments/wards, POCT devices, de-central printers, standard interfaces can be configured (HL7, Link to other hospitals, specialist physician labs, lab groups Use of existing printers, scanners, PC, receipt readers Use of legacy data, retaining identification logic Acquisition costs incl. commissioning, database, operating costs Which contributions are required (configuration, AP computer, preparation of hardware and server), These indicate whether a Europe-wide tender is required EUROPEAN HOSPITAL Vol 27 Issue 4/18

2 0 L A B O R AT O RY Progress despite challenges and delays En route towards digital pathology Dr David Clark is a consultant pathologist for Nottingham University Hospitals and Path Links (a pathology network across all five district general hospitals in Lincolnshire). A histopathologist, he is part of the team implementing the integrated Haematological Malignancy Diagnostic Service for East Midlands Cancer Network, and is also on the Path Links Digital Pathology steering group. to report cases digitally directly into the Path Links LIS. Takeover aids implementation ‘Digital pathology works for routine diagnostic use and remote reporting but we realised that workflow rede- sign and integration with LIS report- ing systems is critical in realising the full benefits,’ Clark pointed out. While he acknowledges there have been teething problems with the implementation process – delay- ing plans for a significant roll out – he remained confident of Path Links reinvigorating its digital pathology project and aiming to move enthusi- astically forward towards full imple- mentation. That has been boosted by the take- over of digital pathology company Omnyx by Inspirata on January 31, 2018, and in the joint presentation at FiLM 2018, Mark Lloyd, execu- tive vice president and founder of Inspirata, outlined the acquisition of Omnyx and the new opportunities it would give to what is traditionally a cancer diagnostics company. ‘We are a nimble and quick com- pany,’ he said, ‘There’s a great deal of opportunity to move quickly in this field to provide value for pathologists, but we know that requires a deep understanding of what the pathologist needs.’ Path Links – which was formed in 2001 by the amalgamation of NHS pathology services in Boston, Grantham, Grimsby, Lincoln, and Scunthorpe and provides a wide range of diagnostic investigations and clinical services to hospitals and 186 GP practices – is now hoping to press ahead with full digital pathol- ogy implementation. Report: Mark Nicholls The benefits of digital pathology have been well-documented along with the need for significant capital investment. Yet, as some pathology networks are discovering, there can be ongoing challenges to bringing an integrated system into play. Path Links – a single managed clinical pathology network oper- ating across Lincolnshire in the east of England – and Nottingham University Hospitals in the UK sought to go down the digital route from 2013 and drew up a business plan which reflected where the department was at that point in time and future considerations. David Clark, consultant patholo- gist with Path Links and Nottingham University Hospitals, explained to delegates at the Frontiers in Laboratory Medicine (FiLM) Congress 2018 in Birmingham (in January) how his organisation is working through the process towards implementation of digital pathology. significant progress will be made throughout 2018. In his presenta- tion ‘Digital pathology – from busi- ness case to full implementation’, he outlined the vision and the reasons why Path Links and Nottingham University Hospitals decided that the digital route was the correct way to go. Tackling a backlog of cases The backdrop, he said, was of a mature pathology department with highly standardised work condi- tions, good IT, and bar coding in place. ‘We were in a good position for digital pathology,’ he said. ‘So why was digital pathology good for us? In 2013 we could see the writing on the wall; the workload had been going up rapidly, and cases had gone up, but they were also gener- ally more complex.’ Each whole time equivalent con- sultant pathologist was handling about 4,500 cases a year, but it was also an aging workforce with 62% of the pathologists over 55, with several nearing retirement. Although there have been delays in the process, Clark told dele- gates that he remains optimistic that The situation took a downturn in 2017 due to losing staff through maternity leave and long-term ill- © Shutterstock/Pakpoom Nunjui ness with a backlog of cases as the seven-day turnaround time of 80% dipped to a 10-day turnaround time of about 35%. When the department began look- ing at producing the business case for digital pathology, he explained, there was insufficient consultant capacity, despite a generally lean workflow. However, the department engaged with the project and began looking to increase existing capacity through improved workflow. Targets were to increase capacity by 10-15% and reduce waste, use digital pathology as being attractive to enhance recruitment and reten- tion, shift to a digital focus, and improve quality. Implementation started in April 2016 with the April-November peri- od designated for installation, LIS integration and validation. Diagnostic reporting began in November 2016 with two patholo- gists reporting 5,000 cases digitally with remote reporting coming into play in December 2017 with digi- tal workstations being installed in the Cellular Pathology department at Nottingham University Hospital allowing Pathologists in Nottingham i a r k a t n O n u r a J / k c o t s r e t t u h S © Precision cardiovascular medicine New biomark for CVD A range of new biomarkers and diag- nostics for precision cardiovascular medicine were outlined in a session at the British Cardiovascular Society annual conference held recently in Manchester. Speakers from King’s British Heart Foundation Centre looked at how mass spectrometry allows clinicians to measure large numbers of proteins simultaneously, discussed a new biomarker for cardi- ac ischaemia that was discovered by a proteomics approach, highlighted a diagnostic structure-function atlas for titin mutations in inherited heart disease, and with first phase ejection fraction outlined a marker for LV diastolic dysfunction. In his presentation ‘A multi-omics approach to cardiovascular risk in the era of precision medicine,’ Professor Manuel Mayr highlighted the advan- tages of mass spectrometry in meas- uring proteins. Measuring a wide range of proteins and lipids ‘Mass spectrometry allows us to meas- ure a large number of proteins simul- taneously and to apply this technique to longitudinal cohorts so that we can associate biomarker measurements with onset of disease or disease progression,’ Mayr said. ‘Basic meas- urements to evaluate cardiovascular risk have not changed that much but, by using mass spectrometry, we can now measure a wide range of pro- teins and lipid species.’ He suggested that the time is right to take advantage of mass spectrom- Marry (a LIS) in haste, repent at leisure Continued from page 19 of processes, e.g. on a process monitor. Data protection is a major chal- lenge. Many people with widely differing training and task levels can be involved in processes, so access to sensitive data in a cen- tral lab and at POCT sites must be clearly defined. Thus a detailed data security concept is an integral component of any LIS offer. One aspect to address is the use of legacy data, since a new LIS con- cept usually involves replacing an old system. One stop shopping is no longer the guiding principle. For some functions, is better sourced from a specialist, e.g. for software document management and docu- ment control programs that sim- plify tasks for quality managers. Economic criteria Cost is among the easier factors to define. Even when the exact LIS implementation and operating costs cannot be specifically deter- mined pre-installation, an estimate is possible and necessary since tenders with a €209,000 amount, spread over five-year period, must be published Europe-wide. Costs to consider include licenc- ing fees, e.g. for database or exe- cution environment; for software implementation and customisation and for a support/maintenance contract; plus, obviously, for hard- ware, e.g. computers/printers. In this context, decide which ser- vices can be provided in-house, either by the lab or hospital, and which existing ones can be used. New purchasing can bring the opportunity to part with obsolete technology. Software and hardware installa- tion could be done by an in-house IT team; so could training during the implementation phase, again reducing costs. All staff needs training when an electronic order entry system is implemented; how- ever, training could take thousands of hours, so only key users should be trained by the software firm. Conclusion Purchasing a new LIS is a huge challenge, but is also an oppor- tunity to examine all processes, weed out inefficiencies and update tried and tested workflows. A good LIS vendor is a partner for life – in prosperity and adversity. Contact: Markus Neumann: markus. neumann@labcore.de The article was first published in Trillium Diagnostik 2018. EUROPEAN HOSPITAL Vol 27 Issue 4/18

2 2 I N F E C T I O N C O N T R O L Myths and truths about antibiotics, antiseptics and vaccination Differentiate and select well established down to the general hospital level. However, this does not solve the problem in the non- hospital sector where, indeed, 85% of all antibiotic prescriptions are gener- ated, mostly with neither lab results nor X-rays (www.rai-projekt.de). There are two possibilities to increase the safety of decisions for or against antibiotics: ‘Either rapid diagnostic tests for biomarkers in the blood, to find out whether we are dealing with a viral or a bacte- rial infection. A study conducted in Hanover showed that, with such a rapid test, we would be able to reduce antibiotic prescriptions by general practitioners (GPs) by up to 60 percent. The second option is the direct detection of pathogens in a rapid test. Last winter we benefited immensely from an influenza rapid test done with a throat sample. This test detects viruses within 20 minutes and we used it right in the emer- gency room.’ However, as Pletz points out, many of these rapid tests are not reim- bursed. No antiseptics without indication Another problem, which is almost as pervasive as antibiotic resistances, are hospital-acquired (HA) – called nosocomial – infections. The number of the six most important HA infec- tions is twice the number of the 32 most relevant contagious diseases. Europe-wide, this translates into 2.6 million infections and approximately 90,000 deaths per year. Infections caused by exogenous pathogens are easier to avoid than those caused by endogenous patho- gens. ‘In the past few years research has focused on antiseptics, since they can reduce endogenous bac- teria. Since antiseptics don’t kill microorganisms selectively, their use needs to be plausible and it needs to be part of hospital-wide hygiene plans and policies. Non- targeted use does carry certain risks and is to be avoided,’ said Professor Simone Scheithauer, Vice President of the Congress and Head of Central Hospital Hygiene and Infectiology at University Hospital Göttingen, Germany. The case of the central venous catheter (CVC) illustrates the ben- efits and limitations of antiseptics. ‘Rather than using a fast alcoholic substance or a mixture of alcohol and PVP iodine to disinfect the entry on the skin, over the past few years a combination of alcohol and octe- nidine respectively chlorhexidine with remanent efficacy has become popular. In 2015, Mimoz et al. published their results in The Lancet showing that the use of chlorhexidine sig- nificantly reduced catheter-induced sepsis,’ Scheithauer explains. The German Commission for Hospital Hygiene and Infection Prevention (KRINKO) thus recommends the use of such combination products. Antiseptic occlusive dressing With particularly vulnerable patients, or when high infection rates are recorded, the commission at the Robert Koch Institute recommends antiseptic occlusive dressing for vas- cular access. ‘When local antiseptic administration is effective it makes sense to apply the substance right at EUROPEAN HOSPITAL Vol 27 Issue 4/18 Professor Simone Scheithauer MD graduated in 2000. Seven years later she became a consultant in medical microbiology, virology and infectious disease epidemiology. She was appointed consultant for hygiene and environmental medicine in 2010 and, in 2012, became a consultant for infectious disease. From 2014 she has headed the division for infection control and infectious diseases at the University Hospital Göttingen, where she became a professor in 2015. Very recently a professorship has also been offered in Göttingen, Germany. development,’ Pletz said, addressing another widespread misunderstand- ing. It is equally misleading to describe MDROs as a mere hospital problem. Whilst, in Germany, MRSA is found mainly in hospitals and numbers are decreasing, other types of pathogens are gaining ground, such as multi- resistant gut bacteria, so-called ESBL- producing bacteria, which enter the body via food. In Germany, 10-13 per cent of patients admitted to a hospital are found to carry this pathogen; in India, 40-60 per cent of the general populace are colonised. They are not eradicated by antibiotic therapy – quite the contrary, further resistances are created. The renowned infectiologist pro- poses a three-pronged approach to fight antibiotic resistance: ‘The alpha and omega is a rational approach to antibiotics in line inter alia with the Antibiotic Stewardship Programme (ABS), which is tested and demand- ed in hospitals. Another important issue is proper hospital hygiene. It is the only way to effectively prevent MDRO transmission from one patient to another. Last and by no means least, there is vaccination, e.g. influ- enza and pneumococci vaccinations are effective measures to reduce the number of antibiotics prescrip- tions in the out-patient sector and thus to reduce the development of resistances.’ Today, quality manage- ment of antibiotics prescriptions is Sixty-two percent of Germans fear antibiotic resistance, according to a survey recently conducted by the German Federal Institute for Risk Assessment. ‘Patients colonised by multi-resistant pathogens are particu- larly scared. But many of these fears are rooted in misunderstandings,’ explained Professor Mathias Pletz, Director of the Institute of Infection Medicine and Hospital Hygiene at University Hospital Jena, speaking at the Congress for Infectious Diseases and Tropical Medicine (KIT), held in Cologne, Germany, this June. Many people believe the develop- ment of resistances can be avoided by the correct use of antibiotics. However, this is wrong: the develop- ment of resistance is part of bacte- rial evolution. Resistance can neither be predicted, nor avoided,’ the KIT Congress President explained, add- ing: ‘in permafrost soil, in bacteria that are more than 30,000 years old, the same resistance genes were found that we see today in vancomy- cin-resistant enterococci (VRE). It’s true, however, that the spread of resistance genes and of multidrug- resistant organisms (MDROs), does correlate with the use of antibiotics: the higher the use of antibiotics, the faster the development of resistance.’ MDROs divide more slowly than sensitive bacteria Another myth is the claim that MDR bacteria are in general more aggres- sive than sensitive bacteria. It is cor- rect that MDROs, as a rule, divide more slowly than sensitive bacteria. Resistance genes offer a benefit only when antibiotics are present. If not, they are nothing but a genetic bur- den. In people who carry MDROs, this reduced biological fitness of the MDROs leads to the human flora being able to overwhelm the MDROs if no antibiotics are administered over a lengthy period of time. ‘Increased mortality in the case of severe MDRO infections is mostly due to the fact that the initially administered antibiotics did not help because, in the early stages of an infection, the physician does not know which type of patho- Professor Mathias Pletz MD specialises in internal medicine and pulmonology and is also certified in infectiology and hospital hygiene. He gained his medical degree in Leipzig, Germany, and then became a research fellow at Emory University’s Rollins School of Public Health and at the Centers for Disease Control and Prevention in Atlanta, USA, for two years. In 2008, he received his habilitation and in 2011, he became Professor of Clinical Infectiology at Friedrich Schiller University Jena. gen caused the infection and thus treats “on suspicion”. Today, detec- tion takes one to two days, which is exactly why we need faster diagno- ses,’ Pletz emphasised. The fact that no new antibiotics are being developed is, above all, an eco- nomic issue, since their development Antiseptics for external use with different degrees of efficacy: the range covers purely alcoholic (far right) and remnant solutions as well as impregnated occlusive dressings and skin washes (far left). Photo: B. Dinkloh Having studied medicine in Münster and Mainz, Professor Markus Knuf MD, in 1997 completed his 5-year training as paediatrician at University Hospital Mainz, with a focus on infectiology, paediatric immunology and neuropaediatrics. From 2001-09 he was a consultant at the Mainz centre; he then became director of the paediatric clinic in Wiesbaden and was appointed chair of paediatric infectiology and interdisciplinary paediatric intensive medicine at the University of Mainz. can take up to ten years and can cost up to one billion US dollars. To make this investment volume economi- cally feasible, Pletz and colleagues demand an extension of patent pro- tection as in the USA. ‘There is reason to hope, because recent literature describes many substances with anti- bacterial effects that have the poten- tial for clinical y a r G t i l i L / k c o t s r e t t u h S ©

WORLD FORUM FOR MEDICINE Leading Internatisnal Trade Fair DÜSSELDORF, GERMANY 12Œ15 NOVEMBER 2018 www.medica.de Member sf • The medical wsrld sn its way ints the digital future • Wsrldwide sverview: Only at MEDICA 2018 BE PART OF THE NO.1! Messe Düsseldorf GmbH P.O. Box 10 10 06 _ 40001 Düsseldorf _ Germany Tel. +49 211 4560-01 _ Fax +49 211 4560-668 www.messe-duesseldorf.de