2 G E N D E R PA R I T Y The gender gap adversely affects AI Data holds a legacy of discrimination Report: Karoline Laarmann Technologies based on artificial intel- ligence (AI) are considered the epitome of progress. However, the data AI algorithms use to draw their conclusions is outdated. It ignores the existence of biological sex and socio-cultural gender and their effects on individual health and disease states. Thus, the ‘think- ing machines’ not only reproduce discriminating bias and prejudices but also produce suboptimal and even outright wrong results. German experts discussed the gender problem in healthcare AI at the Econ Forum by TK at virtual. MEDICA 2020. This was a women- only panel, underlining that it is mostly women who are interested in the gender issue. More so, the audience reflected the panel com- position, as Dr Christiane Gross, Chair of the Medical Committee on Telematics North Rhine Westphalia and President of female phy- association Deutscher sicians’ Ärztinnenbund, pointed out: ‘In training sessions on gender issues I often mainly see women. But the moment you call the session ‘men’s healthcare’ the audience is split evenly in 50 percent women and 50 percent men.’ The word ‘gender’ is still widely considered a women’s niche topic, although gender-specif- ic differences concern all humans. Men, however, have fewer reasons to feel excluded in medical sciences because they are over-represented g n i r p s t h g i L © at all levels. After all, the majority of participants in medical research projects are men, thus the training data fed into AI systems is either focused on men or neglects the gen- der dimension altogether. When machines think like men ‘The gender bias in medicine has been known for years but it took the research community quite a while to catch on,’ said Brigitte Strahwald, the Chair for Public Health and Health Services Research at Ludwig Maximilian University, Munich. During development of AI technologies, legacy data was used that carry the gender-related distor- tions from the analogue into the digital world. ‘This is due, among other reasons, to the gender gap in AI research and development,’ Strahwald pointed out. Algorithms are not only the product of the data sets on which they are based, they are also the product of the humans who program them – the majority of whom are men. Since most healthcare AI applications do not differentiate between sexes and genders, it can be assumed that the IT experts in charge either consider this aspect irrelevant or are not even aware of it. Gendering as a success factor Neglecting the gender dimension is not only a matter of discrimination against women – it undermines the usefulness of AI. Take the digital health applications that are being reimbursed by health insurers under the German Digital Care Act (DVG), which came into effect last October. Barbara Steffens, former health sec- retary of North Rhine Westphalia and current head of regional policy support of a large health insurer, Techniker Krankenkasse, points out that the ‘non-differentiating apps’ help neither men nor women, but only generate costs: ‘In the first certified apps we are aware of five health areas in which we know men and women are affected dif- ferently, not to mention different perceptions and courses of disease.’ More men than women, for exam- ple, suffer from chronic tinnitus, but women experience the disease more intensely. Thus it can be expected that the success of a tinnitus app also hinges on the question of Inclusion of women in studies and guidelines must increase Gender C F i d o D o c c i h C © medicine Women and men are different – nobody could argue against that. However, in medicine, gender plays a subordinate role. Neither research, prevention nor therapy adequately reflect this difference. ‘This is no longer acceptable,’ Professor Vera Regitz-Zagrosek emphasises. At the 127th Congress of the German Society for Internal Medicine, the internist and cardiologist emphatically demanded that females should be more included in studies and that guidelines should consider gender aspects. Medicines work differently in men and women. However, women are under-represented in most studies. Report: Sonja Buske The biggest problem is lack of data. ‘When studies are designed, frequent- ly no thought is given to the fact that women and men may have different reactions to medication, which is also why no research is carried out into potentially different dosage require- ments,’ explains Regitz-Zagrosek. ‘We then obviously cannot expect that doctors will treat men and women differently and in the best possible way, given that there are no studies and guidelines available.’ The cardiologist used a study on the low dose admin- istration of colchicine reduce cardio- to vascular events after myocardial infarction as an example to illustrate this neglect of women’s needs. This study of 5,000 patients included only 900 women. However, the effectiveness of colchi- cine was confirmed for the entire study group and the risk reduction was quoted as being 23%. ‘But only after reading the electronic appendix of the manuscript and taking a closer look did it become clear that the risk reduction in women is only 1%, and that it’s therefore not effective,’ Regitz- Zagrosek points out. According to this cardiologist the case is no exception. Only 11% of cardiovascular studies itemise side effects by gender, and only 30% list effectiveness by gender. Current Covid-19 studies are no different. The proportion of women in three quar- ters of treatment studies is less than 45%. In vaccination studies the per- centage is slightly higher at 45 – 55%, but the side effects are not itemised by gender. ‘This is a mistake,’ Regitz- Zagrosek stresses, because ‘ninety percent of allergic reactions after vac- cinations occur in women.’ Why the gender imbalance in stud- ies? The Berlin-based expert points to one reason: a lack of awareness in research within the pharmaceuti- cal industry. ‘It’s simply assumed that there are no major gender differences and, subsequently, there is also no interest or analysis. Allegedly, it is also too complicated for doctors to consider gender when determining dosage. This attitude is a catastrophe.’ She also believes the licensing authorities should be obliged to demand an appropriate proportion of women and gender-specific evalu- ation in studies. ‘There are guidelines for this, but they are not being imple- mented,’ Regitz-Zagrosek points out. She recommends that doctors should consistently ask companies to publish gender-specific dosage guidelines and urge professional medical associa- tions to reflect gender-specific differ- ences in their guidelines. The under-representation of women is a recurring theme in medicine as well as in cardiology training. Regitz- Zagrosek: ‘Women in cardiology earn less money for the same amount of fulltime work and receive less sup- port than their male colleagues, and thirty percent of female cardiologists also suffer sexual harassment – and Dr Christiane Gross is Chair of the Medical Committee on Telematics, North Rhine Westphalia, and President of the Deutsche Ärztinnenbund, the female physi- cians’ association. Former health secretary of North Rhine Westphalia, Barbara Steffens currently heads regional policy support of Techniker Krankenkasse. whether contents are prepared and communicated in a gender-sensitive manner. Currently, this is not the case. Gender was, and is, the big unknown, not only in digital health- care but also in teaching, training and research. There appears to be an easy way to strengthen aware- ness of gendered healthcare, said Dr Christiane Gross: ‘If we manage to get men on board we will have reached a major milestone.’ Vera Regitz-Zagrosek MD is an inter- nist, cardiologist and senior professor at the Charité Medical University Berlin, as well as a professor at the University of Zurich. She was a founding direc- tor of the Berlin Institute for Gender in Medicine (GiM) at the Charité and founding president of the German and the International Society for Gender Medicine. Regitz-Zagrosek researches molecular, clinical and sociocultural bases of gender differences in heart disease and their importance for treatment. In November 2018 she was awarded the Order of Merit of the Federal Republic of Germany, First Class, for her achieve- ments in Gender Medicine all this against a background of a persistent, highly competitive work environment and inflexible working hours. All this leads to women being pushed out of the competition for leading positions at university,’ she concludes. Whether or not there is a connection between the proportion of women in studies and the propor- tion of female medics in leading posi- tions remains to be proved. EUROPEAN HOSPITAL Vol 30 Issue 2/21

4 M A NAG E M E N T & R E S E A RC H Longer term impact of Covid-19 revealed A new insight into the longer-term impacts and recovery time for patients hospitalised with Covid-19 has been revealed in a UK-based study. The PHOSP-Covid study found that seven in 10 patients have still not fully recovered five months after discharge and many face continu- ing impacts on their physical and mental health, and their ability to get back to work. It also highlighted that those who experience the most severe pro- longed symptoms tend to be white women aged 40-60 with at least two long-term health conditions such as diabetes, lung or heart disease. Led by the National Institute for Health Research (NIHR) Leicester Biomedical Research Centre (a partnership between Leicester’s Hospitals, the University of Leicester and Loughborough University), the research also found that 20% of participants had a new disability, while cognitive impairment was also a predominant symptom in some patients. Burden of symptoms Dr Rachael Evans, an associ- ate professor at the University of Leicester and respiratory consult- ant at Leicester’s Hospitals, said: ‘Our results show a large burden of symptoms, mental and physical health problems and evidence of organ damage five months after dis- charge with Covid-19. It is also clear that those who required mechani- cal ventilation and were admitted to intensive care take longer to recover.’ The study analysed 1,077 Covid- 19 patients who were discharged from hospital between March and November 2020. The most common symptoms were muscle pain, fatigue, physi- cal slowing down, impaired sleep quality, joint pain or swelling, limb weakness, breathlessness, pain, short-term memory loss, and slowed thinking but many patients had sev- eral symptoms at the same time. The study also found a quarter of participants had anxiety and depres- sion and 12% had symptoms of post-traumatic stress disorder after five months. In addition, of the 67.5% of par- ticipants who were working before Covid, 17.8% were no longer work- ing, and nearly a fifth experienced a health-related change in their occupational status. The study team underlined the need for employers to understand this and acknowledge Covid patients may take a long time to fully recover. Proactive approach Dr Evans, the study’s lead co-inves- tigator, said that much of the wide variety of persistent problems was not explained by the severity of the acute illness, indicating other under- lying mechanisms ‘One of the most striking findings of the study is just how few peo- ple are fully recovered five months down the line,’ she said. ‘This dem- onstrates how healthcare services will have to re-think strategies to deal with this post-Covid burden and how they will offer support to this group of patients. We have highlighted the need for a proactive approach. ‘With both mental and physical impairments at five months, wider access to mental health and physi- Image source: Raul lucus from Pixabay (Background); Centers for Disease Control and Prevention (CDC)/Alissa Eckert, MS; Dan Higgins, MAM (Virus); Mashup: HiE/Behrends Professor Louise Wain is the GSK/British Lung Foundation Chair in Respiratory Research in the Department of Health Sciences at the University of Leicester. As a genetic epidemiologist, her research is geared towards understanding the genetic factors that contribute to risk of develop- ing respiratory disease. more prevalent in this group, but further investigation is needed to fully understand the processes.’ Researchers classified the types of recovery into four different groups or ‘clusters’ based on the partici- pants’ mental and physical health impairments. ‘The evidence for different recov- ery clusters, and ongoing inflamma- tion, really is important in guiding how we conduct further research into the underlying biological mechanisms that drive Long-Covid,’ Professor Wain added. Largest study The PHOSP-Covid study also meas- ured the impact of medicines given during hospitalisation to see if they affect patients’ recovery. Early indi- cators show that while giving cor- ticosteroids is a factor in reducing mortality in hospital, it does not appear to have an impact on longer term recovery. There are more than 300,000 post- hospitalisation survivors that have been discharged from UK hospital’s following Covid-19 and while the study only represents a small sam- ple of these patients, it is the largest study to report in detail on pro- spectively assessed outcomes across multiple UK centres to describe the impact of Covid-19 on the medium- term health of survivors. The study will see participants followed-up at 12 months, as well as blood samples analysed, and ongoing effects on mental health and their immune systems assessed. (MN) Dr Rachael Evans is an NIHR Clinical Scientist at the University of Leicester and an Honorary Consultant Respiratory Physician at Leicester’s Hospitals and a co-investigator of the study PHOSP-Covid. Her areas of expertise include exercise physiology, training and rehabilitation for people with long-term cardiorespiratory conditions. cal health interventions is required as well as a co-ordinated approach so patients are not seeing several different specialists.’ Dr Evans said health systems face a long-term legacy from Covid-19 and will need a range of strategies and interventions to cope with this. Recovery clusters The research has also uncovered a potential biological factor behind some post-Covid symptoms. A biological marker associated with inflammation, C-Reactive Protein (CRP), is elevated in all but the mild- est of post-hospital cases. Professor Louise Wain from the University of Leicester, and also a co-investigator for the PHOSP-Covid study, said: ‘From previous studies, it is known that systemic inflamma- tion is associated with poor recov- ery from illnesses across the disease spectrum. ‘We also know that autoimmun- ity, where the body has an immune response to its own healthy cells and organs, is more common in middle- aged women. This may explain why post-Covid syndrome seems to be Skull & bones? Female specialists fear not Family or career? Hättich has seen many female col- leagues throw in the towel: ‘Very often it is during the specialist phy- sician training that a decision has to be made: do I want to start a family or do I want to stay on the career track towards senior physician or even head of department?’ Female physicians who opt for the family tend to abandon their specialist phy- sician training or transfer to an off- hospital physician’s practice where work schedules are more compat- ible with being a mother. The Hamburg-based orthopaedist blames – among others – the lack of support for and by women: ‘Men support each other more, be it in mentoring programmes or in net- works. While indeed there are a few universities that offer mentoring programmes for female surgeons – albeit not specifically in ortho- paedics and trauma surgery – we women are such a minority that we have the feeling we have to fend for ourselves.’ This is where the Junges Forum O&U – a forum for junior physicians in orthopaedics and trauma surgery – comes in. The group was founded about 20 years ago to represent the interests of the young generation of orthopaedists and trauma surgeons. Hättich is heading the chapter deal- ing with support for junior physi- cians and she is involved in issues surrounding training, family and career, science and interdisciplinary cooperation. While both men and women are targeted, the majority of members as well as leaders are women. A topic which is close to Hättich’s heart is surgery during pregnancy – which has long been a taboo. ‘Today, studies confirm that under certain conditions working in the OR is not associated with par- ticular risks for pregnant women,’ Hättich explains and ads that ‘when no gas and x-rays are involved there is no reason to exclude pregnant surgeons. Many of my pregnant col- leagues are indeed grateful for the opportunity to perform surgeries Continued from page 3 because thus they lose less time dur- ing their further training.’ In early 2021, another organisation was founded to provide a platform for female surgeons to exchange knowledge and experiences. ‘Die Chirurginnen’ – the female surgeons – arose from the women‘s desire to network, to learn from each other and to grow together. Working as a team – on equal terms Hättich wants the women to organ- ise, to find ways to be more vis- ible and to further their careers. Nevertheless: ‘Our aim is to work in a team and on equal terms, no matter which time model. We want the team to function as a unit rather than being split into differ- ent groups with differing interests.’ Annika Hättich herself has success- fully taken the next step: as of 1 July she will be senior physician at her Hamburg hospital. r a m A l a G © We Can Do It! EUROPEAN HOSPITAL Vol 30 Issue 2/21

6 M A NAG E M E N T & R E S E A RC H Implantation of cartilage or stem cells Remarkable cartilage regeneration Report: Michael Krassnitzer Joint cartilage usually regenerates very poorly. The matrix of cartilage (scaffolding) contains very few cells in deep layers. Moreover, joint carti- lage is highly isolated: there are nei- ther regenerative cells in the imme- diate vicinity that could migrate into the site of the defect and trigger repair, nor vessels that could trans- port regenerative cells to the carti- lage. Unsurprisingly, researchers all over the world are experimenting with different approaches to trans- porting biomaterials to the joint to make cartilage repair more robust. In pre-clinical trials a research team* at the Medical University Vienna (Department of Orthopaedics and Trauma Surgery, Division of Trauma Surgery) could achieve remarkably extensive regenera- tion of joint cartilage from cartilage matrix as biomaterial. The matrix used was harvested from surgery patients who had consented to have their tissue examined for this study. All cartilage cells were removed from the donor matrix leaving native cartilage tissue consisting of collagen type II fibres und glycosa- minoglycans. This material, called CartiScaff, is transported to the defective sites in the joint, to create an ideal environment for cartilage cells or stem cells. ‘The challenge here is to insert native cells into the tissue,’ explains Private Docent Dr Sylvia Nürnberger, the project leader. ‘Cartilage cells are indispensable for cartilage repair,’ she pointed out, because cartilage cells build collagen type II and have to respond to different types of stress. Cartilage matrix without Editor-in-Chief: Brenda Marsh (BM) Editorial team: Wolfgang Behrends (WB), Sonja Buske (SB) Senior Writer: Mark Nicholls (MN), Great Britain Mélisande Rouger (MR), Spain Publishing Director: Mareike Scholze (MaS) Executive Director: Daniela Zimmermann (DZ) Founded by Heinz-Jürgen Witzke ISSN 0942-9085 Correspondents Austria: Michael Krassnitzer (MK), Christian Pruszinsky (CP) France: Jane MacDougall (JMD) Germany: Cornelia Wels-Maug (CWM), Karoline Laarmann (KL), Copyright: MedUni Wien / LBI-Trauma µCT scan showing a 3-D view of the laser incisions. The material is implanted tip-down, i.e. perpendicular to the defect. The (implanted) biomaterial covers the defect to the joint cavity cells would suffer from wear and tear over time. ‘Cartilage cells keep the cartilage intact,’ she explained. ‘They are the maintenance crew of the cartilage.’ Laser opens the door for helper cells Laser beams have long been a surgi- cal tool, e.g. to correct pathologies in the ear or eye. Nürnberger’s team perforate the cartilage matrix with a laser beam to create an opening for the cells. Tiny grooves were engraved in the matrix, with high precision and all equal in size and distance. These served as access for the regenerative cells, which could thus migrate even in deeper tissue layers. There are three ways to transport the cells to the matrix: either a hole is drilled to the bone marrow and the stem cells can enter the site of the defect, or the matrix and bone marrow harvested from the iliac crest are inserted into the joint in one go. The stem cells then develop into cartilage cells. The third option: the matrix is seeded with stem cells or cartilage cells prior to implanta- tion. The laser incisions offer another advantage: they provide the carti- lage tissue with a high degree of flexibility so it can be more eas- ily fitted onto the rounded surface of the joint bone – comparable to the flexibility of a concertina, Nürnberger observed. ‘The cartilage tissue that grows in the grooves is of rather good quality considering that it’s regenerated tissue.’ Collagen II fibres are orientated perpendicular to the bone and, since the surface is created by the implanted bio- material, it has the same structure as native cartilage. ‘With the bio- Katrin Schreiter (KS) Spain: Eduardo de la Sota (EdS) The Netherlands: Madeleine van de Wouw (MvW) USA: Cynthia E. Keen (CEK), Lisa Chamoff (LC) Subscriptions Dorothea Fleischer, Theodor-Althoff- Str. 45, 45133 Essen, Germany Subscription rate 4 issues: 32 Euro, Single copy: 8 Euro. Send order and cheque to: European Hospital Subscription Dept Printed by: WVD, Möhrfelden, Germany Publication frequency: quarterly Representatives China & Hongkong: Gavin Hua, Sun China Media Co, Ltd. Phone: +86-0755-81 324 036 E-Mail: 627416876@qq.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: jund@european-hospital.com GB, Scandinavia, BeNeLux: Simon Kramer Phone: +31 180 6200 20 E-Mail: kramer@european-hospital. com Israel: Hannah Wizer, International Media Dep. of El-Ron Adv. & PR Co., Ltd., Phone: +972-3-6 955 367 E-Mail: ronin@netvision.net.il 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 Maria Kaiser Phone: +1 250 726 4007 E-Mail: mkads@mac.com Roxsy Mangiante Phone: +1 21 469 48542 E-Mail: mangiante@european-hospital.com All company, brand and product names in this publication are the property of their respective holders. Users must obtain permission from those holders before copying or using the owner’s trademarks, prod- uct and company names or logos. material being so similar to native cartilage, it can stay inside the defect as a part of the regenerated tissues. Alternatively, it can degrade and be replaced with new matrix,’ Nürnberger explained, adding that follow-up studies will show how the implants develop over time. * ‘Repopulation of decellularised articular cartilage by laser based matrix engrav- ing’. S. Nürnberger, C. Schneider, C. Keibl, B. Schädl, P. Heimel, X. Monforte, Dr Sylvia Nürnberger is a researcher in the Department of Orthopaedics and Trauma Surgery, in the Trauma Surgery Division, at the Medical University of Vienna, Austria, where she heads the Tissue Regeneration group. Her research focus is biomaterials, chondrocytes and alternative regenerative cells (e.g. fat stem cells) in biological therapeutic substances. A.H. Teuschl, M. Nalbach, P.J. Thurner, J. Grillari, H. Redl, S. Wolbank. doi:10.1016/j. ebiom.2020.103196 a m u a r T - I B L / n e i W i n U d e M : t h g i r y p o C Histochemical section of the implanted biomaterial. The tips are shown in blue; the gaps are filled evenly with neo-tissue (brown). Low-cost technique for missed genetic mutations Report: Mark Nicholls at A new low-cost method targeting genetic mutations often missed by existing diagnostic approaches has been developed. Researchers Virginia Commonwealth University (VCU) in the United States noted that most rearrangement mutations implicated in cancer and neurological diseases fall between what can be detected by DNA sequence reads and optical microscopy methods. The new technique combines digital polymerase chain reaction (dPCR) and high-speed atomic force microscopy (HSAFM) to create an image at a nanoscale resolution, which allows technicians to meas- ure differences in the lengths of genes and identify these missing mutations. Nanoscale resolution Dr Jason Reed, an associate profes- sor in the department of physics at the VCU College of Humanities and Sciences, said: ‘Our approach is the first to combine dPCR with HSAFM to create an image with such nanoscale resolution that users can EUROPEAN HOSPITAL Vol 30 Issue 2/21

8 L A B O R AT O RY Specimen diversion devices lo Preventing blood cultur Blood cultures are among the most important laboratory tests a patient can have, especially to diagnose serious infections. Contamination is a constant concern – a patient could be misdiagnosed due to a false-posi- tive test. Even with the most diligent quality control measures to limit contamination, including rigorous skin disinfection, contamination may occur from skin fragments colonised with bacteria that are dislodged with venipuncture. However, when initial specimen diversion (ISD) techniques are used, potential rates of contami- nation from skin fragments can drop to near zero. The effectiveness of using an ISD device to reduce blood culture con- tamination (BCC) rates and a new national movement in the USA to reduce the BCC hospital bench- mark to 1%, or less, were topics of an April 2021 Dark Daily webinar featuring two experts in infection control. They discussed the impact that an ISD device that separates the initial blood draw from the laboratory samples can have in significantly reducing BCC rates in hospitals, referencing findings of numerous clinical trials. Unacceptable: The USA’s 3% BCC benchmark In 2007, the Clinical and Laboratory Standards Institute (CLSI) recommended that BCC rates should not exceed 3%, a threshold that is now consid- ered to be acceptable in U.S. hospitals. By 2011, articles in peer-review journals began cit- ing this as a standard. ‘This rate is woefully inad- equate. It was never published as a standard,’ said Dennis J Ernst, founding director of the Centre for Phlebotomy Education in Mio, Michigan, USA. ‘Emergency department tests for sepsis, the number one cause of death and readmission in U.S. hospitals, average about 8% positive. If you factor a 3% contam- i n a t i o n rate, 40% of these are false positives,’ said Barbara DeBaun, RN, of Cynosure Health in San Francisco. ‘As a nurse,’ she added, ‘I cannot think of any other test performed in a hospital that is incorrect 40% of patient is given unnecessary antibiotics, these may cause kidney damage, higher risk of staph infections, and resistance in the future. A false-positive blood culture can increase an in-patient’s hos- pital stay by two or more days, which is costly, a waste of hos- pital resources, and a risk to the patient of developing an infection. DeBaun stated that the average cost per false-positive incident is an estimated $4,300 to $4,700 for an in-patient. In the USA, a national move- ment is underway among health- care professionals to establish a new standard benchmark of less than 1% for BCC rates. This is attainable, according to the presenters. They believe that the U.S. Centers for Medicare Steripath Gen2 Initial Specimen Diversion Device Copyright: Magnolia Medical Technologies, Inc. Seattle, USA. the time.’ When a patient has a false-posi- tive test for sepsis the implications are huge, she pointed out. If a and Medicaid (CMS) should make this a quality measure and report- able. It is imperative for hospital health- care providers to rigorously practice and enforce proven techniques and Mass spectrometry quality control measures associated with BCC prevention. These include skin disinfection, strict aseptic prac- tices and quality control, the use of a pre-packaged blood culture kit, and ample education and training. ‘But, when a hospital is doing everything right, contamination may still be caused by skin fragments. Active diversion of the initial 1.5 to 2.0 mL of blood using a closed system has been repeatedly clinically proven to reduce BCC,’ Ernst said. ‘Manual diversion to a waste tube will not solve the problem as effectively. In the best case-controlled clinical study scenario, the lowest published contamination rate achieved was 2%.’ uality measure and reportable. Reducing BCC rates Use of a pre-assembled, sterile blood culture system (Steripath Initial Specimen Diversion Device, Magnolia Medical Technologies) designed to divert and separate the initial 1.5 to 2.0 mL of blood from the rest of the sample, combined with rigidly enforced quality control measures, can reduce BCC rates to 1% and less. This device, which is U.S. FDA approved, is preassembled and ster- Skin swabs can detect Covid-19 Skin swab samples analysed using mass spectrometry could be used to detect Covid-19 in patients, accord- ing to research conducted at the University of Surrey in the UK. Current Covid-19 testing is via a polymerase chain reaction (PCR) test, which involves taking a swab of the back of the throat and inside the nose, but the team from Surrey – working with Frimley NHS Trust and the Universities of Manchester and Leicester – have used sebum samples from the outer layers of skin (stratum corneum) to identify the presence of Covid-19. They collected samples from 67 hospitalised patients – 30 who had tested positive for Covid-19 and 37 who had tested negative – by swab- bing skin between the shoulder blades. This area of the body is rich in sebum, the oily, waxy substance produced by sebaceous glands. Samples were then analysed using liquid chromatography (to separate A laboratory scientist prepares samples for high-performance liquid chromatograph-mass spectrometry v a l s y d a l V l e d o S © Covid-19 on the host metabolism, by looking at what the virus does to patients, rather than looking for the virus itself. A challenge faced by researchers was that comorbidities also dysregulate the skin, making it difficult to differentiate between Covid effects and conditions such as Type 2 Diabetes. However, this was addressed with medication and additional health conditions fac- tored in.gative results.’ the lipids) and mass spectrometry (to identify the lipids). The Partial Least Squares Discriminant Analysis (PLSDA) statistical modelling tech- nique was used to differentiate between the Covid-19 positive and negative samples. Patients with a positive Covid-19 test had lower lipid levels – or dyslipidemia – than those with a negative test. Study co-author Matt Spick said: ‘The test differentiates between pos- itive and negative Covid-19 results by looking at sebum lipids in the outer layer of skin. ‘We found that participants infected with Covid- 19 had markedly reduced levels of many triglycerides and ceramides in particular. Although no single lipid gave a definitive signal, by looking at the overall lipid profile through multivariate machine learning, we could differentiate between positive and negative results.’ Sebum sampling The team believe sebum sampling has the potential to support diag- nosis and prognosis, as well as for investigation into the impact of Differentiating lipids is a challenge ‘We controlled for this by match- ing comorbidities,’ added Mr Spick, ‘for example, looking at the differ- ence between Covid-19 positive and negative within the subgroup of participants with T2DM. We found that when controlled for comorbidi- ties our model was 82% accurate.’ In addition, with thousands of dif- ferent lipids present in the human body, differentiating between them proved a further challenge. However, co-author Dr Melanie Bailey explained: ‘Mass spectrom- etry lets us identify the individual lipids, both by accurate molecular mass and also by breaking the mol- ecule apart and identifying charac- teristic fragmentation patterns.’ PCR remains the gold standard in Covid testing in terms of accu- racy, and there is no suggestion that sebum-based tests can replace them, but certain researchers believe sebum swabs may provide an alter- native, although they acknowledge the test is not yet suited to clini- cal settings. At present they think the work is more likely to inform research into skin health and how this may be influenced by infection. Dr Bailey said: ‘The most obvi- ous implication of our findings is that skin can be added to the list of organs that are dysregulated by Covid. Understanding skin health Another feature of Covid-19 is the range of symptoms that those infect- ed may present with. We do not yet fully understand how this variation might translate to skin, but it does appear that the reduction in sebum triglyceride and ceramide levels is a consistent feature across the popula- tion that we sampled. ‘More broadly, understanding how diseases interact Researcher Mike Wilde is processing an extracted gauze sample ready for analysis. with other comorbidities in influ- encing the metabolism of the stra- tum corneum can only help us understand skin health.’ The work is one strand of research into skin health as a diagnostic tool, with work at the University of Manchester, for example, looking at how sebum sampling followed by mass spectrometry can be used to diagnose Parkinson’s disease. Because the skin is so rich in lipids, EUROPEAN HOSPITAL Vol 30 Issue 2/21

1 0 L A B O R AT O RY AI algorithm reads patterns of Covid-19 infection in blood ‘Barcode’ brings quicker test results Report: Madeleine van de Wouw When patients are admitted to a hospital emergency room (ER) it is immediately vital to deter- mine whether s/he has Covid-19. However, with a regular PCR test a result can take up to a few hours. Thus, initially, the patient must be isolated. During the height of the corona pandemic last year, research- er Ruben Deneer from Eindhoven University of Technology (TU/e) and clinical chemist Arjen-Kars Boer from the Catharina Hospital were asked whether they could find a quicker way to rule out Covid-19. The answer was ‘yes’. ‘During the first period of Covid- 19 no quicker PCR tests were avail- able; results sometimes took up to four hours,’ Boer explained. ‘Now there are faster tests on the market, but they are expensive – in the Netherlands about €110 per test. We knew that patients admitted to the emergency room have their blood taken as standard for a so-called quick scan. The blood is tested for as many as 30 different values. So we wondered if an infection might be visible in the blood, appearing as a sort of recognisable barcode that is characteristic of Covid-19.’ The researchers worked on that idea and their studies showed that, with the help of a calculation algo- rithm, it was indeed possible to show these changes in the blood. This led to the CoLab score. Unlike a PCR test, which indicates the pres- ence of virus particles, the CoLab score analyses the changes the virus causes in the blood. Those changes are present even before a patient notices or knows that he or she is infected. The CoLab analysis rules out the signs of an infection so a PCR or LAMP test is only necessary when infection cannot be ruled out. CoLab screening does not replace a PCR test but is an additional tool for quick initial Covid-19 screening and is – with an accuracy of almost a hundred percent on the absence of the ‘barcode’ in the blood – extremely suitable for application in the emergency room. There are several advantages: expensive PRC- tests are needed less often and, because the results are back within an hour, patients with a negative result can leave isolation quickly. Also, the medical staff will know in a short time frame whether they can approach the patient without exten- sive personal protective equipment. How the test score works The CoLab score is software based on an artificial intelligence (AI) algorithm. With this algorithm, pat- terns – the ‘barcode’ that indicates a Covid-19 infection – are recog- nised in the blood. If no patterns are present, an infection is virtually excluded. ‘During the search for that bar- code’, Boer explained, ‘the data of more than 10,000 patients entering the emergency room were digitised to develop the algorithm. The data were both from infected and non- infected people. To analyse this data manually would be an almost impossible, painstaking and labori- ous process, so we used AI to draw the right conclusions from those data. ‘We looked for the smallest set of results that makes a good predic- tion and we managed to identify ten laboratory tests that together form the best predictors of Covid-19.’ Results are expressed as a cut-off value from 0 to 5 – a value used to determine whether a result is abnormal. The number indicates the values found. With 0, 1 and 2, the chance of corona is almost zero. At 3, 4 and 5, a subsequent PCR test is needed to demonstrate infection. Because a PRC test is not a rou- tine procedure if people do not show symptoms, missing a very small portion of patients has more consequences when the prevalence is high rather than low. The cut-off value, the value used when a result Arjen-Kars Boer PhD, who has worked in the Catharina Hospital, in Eindhoven since 2006, is listed in the European reg- ister of specialists in laboratory medicine, as well as an endocrinologist. Besides this specialty he is known for using big data solutions in clinical chemistry, bari- atric surgery, cardiology, and emergency medicine. He also trains various laboratory specialists and co-promotes several PhD students. is called abnormal, is therefore prev- alence dependent. A €150,000 grant Recently these Dutch researchers received a grant of €150,000 to accelerate the introduction of the CoLab score in European hospi- tals. ‘That’s quite a challenge,’ Boer observed. ‘Every hospital carries out different tests in the emergency room. We are therefore going to make our CoLab score ‘ smarter’ so that it can be used in every hospital in Europe. We’re working hard on that along with the company Gaston Medical. Every hospital in Europe should be able to use it very sim- ply: plug and play. It’s easy if they already use Gaston Medical’s soft- ware, but a connection must also be sought for hospitals that use other software.’ from The grant comes the COVID-X programme which aims to beat Covid-19 with data solutions. There were 112 applications sub- mitted and only 15 were honoured, including one from the Netherlands. Funded by the EU within the Horizon 2020 programme, the Covid-X program aims to make data solutions ready for the European market in nine months. Speeding up diagnostics to detect antibiotic r New goals w continuous inno Infectious disease diagnostics are noto- riously slow. The gold standard for laboratory diagnosis of bacterial and fungal infection involves growing the pathogen from a clinical specimen – an overnight event, or even longer. The healthcare focus is on improv- ing the use of antibiotics for better patient outcomes and reducing the environmental pressures that drive antibiotic resistance. To impact anti- biotic use, diagnostic tools need to be faster without compromising the performance characteristics neces- sary for definitive therapy decisions – a lot to ask, but this is a new era of innovation and companies are driven to fill this need. Antimicrobial susceptibility test results enable definitive treatment decisions. This test traditionally takes three or more days from specimen collection to result, but technology is chipping away at the timeline. Molecular tests can identify the path- ogen and detect resistance mecha- nisms from a positive blood culture in less than two hours. Knowing the infectious aetiology is powerful. This information alone can significantly reduce the time to de-escalate empir- ic therapy decisions to more targeted therapies *1. Detecting resistance mechanisms is most helpful for Gram-positive bacteria where a limited number of acquired resistance mecha- nisms are responsible for the most important phenotypes, like mecA in Staphylococcus aureus and van genes in Enterococcus. Assays to detect resistance mechanisms in Gram-negative bacteria can provide early evidence of what drugs will not work, but failure to detect a resistance mechanism is insufficient evidence to de-escalate antibiotics because many acquired resistance EUROPEAN HOSPITAL Vol 30 Issue 2/21 © Shutterstock / Designua

1 2 A I & M A M M O G R A P H Y Identifying cancers in symptomatic younger women DBT shows superiority Report: Cynthia E. Keen Digital breast tomosynthesis (DBT) increases detection of breast can- cer in symptomatic women under the age of 60, especially in dense breasts. A large, multi-institution- al study conducted in the United Kingdom to compare the sensitivity of full-field digital mammography (FFDM), DBT, and FFDM plus DBT supports findings of two similar published studies, both conduct- ed in China within the same time frame. Much research has been carried out to confirm the increased cancer detection capabilities of DBT tech- nology as a breast cancer screening tool, but very little has been under- taken to evaluate and quantify the accuracy of DBT in symptomatic patients. The rigorous double-blind ret- rospective study, led by research- ers at the University of Dundee in Scotland, utilised images acquired from a different mammography sys- tem than that used by the Chinese researchers. Because DBT technol- ogy differs significantly, findings of a single-vendor study do not necessarily apply when other equip- ment is used. This new study pub- lished online in the British Journal of Radiology complements the research in China with similar data for a European population. The participants included 300 women with symptoms or signs of breast cancer representing a 20% or greater likelihood of malignancy fol- lowing a clinical examination. They had both FFDM and DBT at five UK hospitals’ specialist breast centres within a five-year period, starting in 2011. All sites used MAMMOMAT Inspiration units from Siemens Healthineers. Images were acquired at all sites. The study cohort includ- ed all 152 women recruited who were diagnosed with cancer, plus a randomised selection of half of the 209 recruited women with benign ma in situ (DCIS) represented only 1% of the 157 cancerous breasts. Twelve breast imagers experi- enced with Siemens DBT indepen- dently analysed a batch of 50 cases for each category, for 150 read- ings, two readers per batch. Each batch had similar distributions of different aged patients, and can- cerous, benign, and normal cases. To replicate real-life practice condi- tions, they were provided with the patient’s age and clinical records. The MAMMOMAT Inspiration from Siemens Healthineers is equipped with 50˚ Wide-Angle Tomosynthesis and offers high-quality results using up to 30% less dose in FFDM, with MoodLight and OpComp personalised compression to improve patient experience. lesions and half of the 76 women with normal breasts. The mean age was 47 years, but some participants were as young as 24 years. None of the patients had breast implants, as DBT had not yet been approved for use. Masses, at 89%, were the dominant radiologi- cal feature in malignant breasts, and 85% were unifocal tumours. The majority of cancers were Grade 3 or 2 invasive ductal tumours (42% and 33% respectively) and invasive lobu- lar carcinoma (12%). Ductal carcino- Case 1. A symptomatic woman was screened with both FFDM and wide-angle digital breast tomosynthesis. Imaging revealed an 8 mm spiculated mass that was more visible on wide-angle digital breast tomosynthesis as compared to FFDM. Biopsy samples from this lesion demonstrated an invasive ductal carcinoma on histopathology. The imagers made their interpre- tations using FFDM images only, DBT images only, and FFDM+DBT images, marking regions of interest, measuring lesions they identified, and describing abnormalities. They scored suspicious lesions on a 1-to- 5 scale (normal to malignant). For FFDM images, the readers catego- rised breast density using a BI-RADS density 1-4 category score, and used an on-screen 0-100 mm visual ana- logue scale to assign an area-based percentage mammographic density. The percentage of volumetric breast density was also assessed using Volpara Data Manager software (Volpara Solutions, Wellington, NZ). Andy Evans MD, Professor of Breast Imaging at the University of Dundee and honorary consultant radiologist in NHS Tayside, and co- authors, compared the 1,800 case findings with the diagnosis of each patient, based on clinical exam, medical imaging, and histopathol- ogy. Readers identified 260 unifocal malignant lesions on DBT images compared to 214 on FFDM. Highest sensitivity The FFDM+DBT combined exams had the highest sensitivity in detect- ing breast cancer, at 97%, followed by 89.1% for DBT alone, and 86.6% for FFDM alone. In the densest breast category, FFDM+DBT identi- fied 10.3% more cancers than FFDM alone, but DBT alone was most accurate for breasts in the third BI-RADS breast density category. Overall specificity was highest for DBT alone (84.6%), followed by FFDM (81.4%), and then FFDM+DBT (79.6%). The researchers attribute the lower FFDM+DBT reader results as a function of needing to inter- pret two exams instead of just one. Tumour measurement accuracy was comparable for all three groups. Based on their findings, the authors recommend that DBT be used for symptomatic younger women with known very dense After leading the mammography train- ing service at King’s College Hospital in London, since 2009 Patsy Whelehan has researched clinical breast imaging in Dundee alongside Professor Andy Evans. As well as radiology research, she has a strong interest in patients’ mammogra- phy experience. Clinically, she exemplifies the extended role opportunities available to radiographers in the UK, whereby with rigorous additional training radiographers can undertake advanced practice, such as reporting mammograms and performing breast biopsies. ‘With breasts, or as a second-line exam following a negative FFDM. And, in view of increasingly available contrast-enhanced digital mammog- raphy (CEDM), they also recom- mend that the performance of DBT be evaluated in clinical studies with this breast exam. respect to additional research, the Dundee team is cur- rently conducting a UK multicentric study called CONTEST, comparing the diagnostic accuracy of CEDM with FFDM, DBT and MRI,’ lead author Patsy Whelehan, a research- er at the University of Dundee School of Medicine and a consultant radiographer at NHS Tayside, told European Hospital. ‘The study start- ed as single centre in 2018, but we now have several centres on board and recruitment is ongoing. Another major current UK-wide project from the Dundee team is MEDICI, which aims to assess whether a reduc- tion in mammographic density in women taking adjuvant endocrine therapy following breast cancer sur- gery is associated with a lower risk of recurrence and/or a lower risk of dying.’ Malignancy detection as good as radiologists AI potential in breast imaging efficiency The detection and classification of breast cancer as well as all aspects of the breast imaging workflow chain could be much easier with AI models. K s a m o T © The contribution of Artificial intelli- gence (AI) has great potential in breast imaging efficiency, Professor Linda Moy MD told attendees at the 2021 Society of Breast Imaging/American College of Radiology (SBI/ACR) Breast Imaging Symposium this April. AI models for breast imaging have focused mainly on the diagnostic classification and detec- tion of breast cancer. However, AI appli- cations for workflow optimisation can provide support for interpretation tasks and increase overall operational efficien- cies in all aspects of the breast imaging workflow chain. ‘There are currently numerous AI applications in development that have the potential to benefit breast imaging, from test ordering to report communication,’ said Moy, who is professor of radiol- ogy at New York University’s Grossman School of Medicine. ‘At a time when there’s a shortage of breast imagers, there’s increasing patient volume, and with increased utilisation of digital breast tomosynthesis (DBT) and breast MRI, an increasing number of images in an exam. Radiologists can benefit from intelligent, automated, time-efficient tools to perform or expedite repetitive tasks. These can help increase efficiency for breast imagers.’ The workflow chain in imaging includes multiple segments, starting with an order for the patient. This needs to be reviewed for imaging appropriateness and the protocol to be used, sched- uled, performed, managed, reported with results communicated, and billed. A staff member is needed at each step to inte- grate information and accurately pass the information along to the next step. AI can potentially address every aspect of the workflow, Moy believes, ‘enhancing practice workflow efficiency, reducing variability, and improving quality.’ ‘Most attention for AI in breast imaging goes to pixel-based image analysis, with a focus on narrow AI tasks including detection, classification, segmentation, prepopulat- ing a report, prediction of breast can- cer risk, and treatment response predic- tion,’ she explained. ‘Less attention is being paid to radiology workflow, but I think the greatest potential for AI lies in making these back-end processes more efficient. The current limitation and dif- More than just MRI accessories www.allmri.com EUROPEAN HOSPITAL Vol 30 Issue 2/21

1 4 I N T E RV E N T I O NA L R A D I O LO G Y ECIO 2021 AI in interventional oncology: promise & reality Report: Cynthia E. Keen Is artificial intelligence (AI) technology ready to be utilised as a clinical tool by interventional oncologists? Not yet, but when it is, AI technology’s clinical impact may be as profound as advanced imaging is today. This is the consensus of two leading researchers developing AI for inter- ventional oncology use, who pre- sented back-to-back scientific ses- sions at ECIO 2021 on both the promise of AI and current concerns and limitations about AI use today. The potential impact of AI in clini- AI could help standardize and automate image interpretation and cancer detection stratification and cost-effective clas- sification, and auto-segmentation and registration for interventions and followup. Dr Wood cited an investigation of the impact AI could have on clinical trial eligibility. Researchers Genetech and Stanford University’s Department of Electrical Engineering evaluated AI software called Trial Pathfinder to emulate completed clinical trials of advanced non-small- cell lung cancer using data from electronic health records of over 61,000 patients. AI-defined inclu- sion criteria doubled the number of eligible patients and improved the hazard ratio, suggesting that many patients who were ineligible to par- ticipate in the clinical trials could predictive instruments, and imaging techniques for the diagnosis, charac- terisation and image-guided therapy of hepatocellular carcinoma (HCC). Tumour treatment management today is based on inadvertent biased interpretation of evidence. Decisions are made based on evidence accrued from experience and knowledge reflective of an individual physi- cians’ perceptions. There are more than 25 unclear, varying, and incon- sistent diagnostic imaging guide- lines. No unified staging system for HCC exists, and the wide variety of loco-regional therapy options are all supported by different data. ‘The hypothesis of utilising AI is that it is exclusively data driven and therefore neutral to the personal a l l e n u m a s © cal oncology is immense, according to Brad Wood MD, director of the National Institutes of Health Center for Interventional Oncology and chief of Interventional Radiology at NIH Clinical Center and the National Cancer Institute in Bethesda, MD. He leads a large multi-disciplinary team of NIH researchers and academic and industry partners that develops devices, software, and navigation approaches for cancer patients via novel local and regional minimally invasive image-guided therapies. Potential capabilities AI will help create accurate predic- tive models by integrating deep learning to genomic molecular, clini- cal, histology, imaging and radiomic data. It will be used for staging and response criteria and risk and out- come prediction. AI biomarkers will be used in drug discovery, with the ability to select drugs and pre- dict side effects. This capability will shorten time to drug development, help determine where to biopsy and then correlate image to pathology, explained Dr Wood. In imaging, AI will help stand- ardise and automate image inter- pretation and cancer detection in face of widely variable radiology and pathology human practices. Examples of this include tumor have potentially benefitted from the treatments. AI has the ability to navigate through a huge number of images that is infeasible for a human researcher and has the potential to make clinical assessments. These will include detection of abnormali- ties and characterisation of them, such as segmentation defining the boundary extent of abnormality for subsequent diagnosis and treatment. AI tools will be able to evaluate and classify abnormalities as malignant or benign, and to stage them into multiple predefined categories. AI will aid change analysis, by tracking object characteristics across multiple temporal scans and across multiple modalities for diagnosis as well as evaluating treatment response. ‘All this will happen, but it is easier said than done, and there are a lot of pitfalls,’ Dr Wood said. AI: A ‘data wrangler’ of the future Julius Chapiro MD PhD., co-director of the Yale Interventional Oncology Research Lab of the Department of Radiology and Biomedical Imaging at the Yale University School of Medicine in New Haven, CT, agrees. He leads an interdisciplinary team of basic, translational, and clinical scientists who are developing new quantitative imaging biomarkers, cognitive bias of a physician. But is this really true? Can AI help us break out of the vicious unintended circle of bias? The problem is that any AI decision support tool will only be as good as the data that we use to train it,’ Dr Chapiro explained. Large amounts of high-quality, annotated curated data are needed from multiple sources. Selecting the data to use is very complex; the quality of data is key. A large data volume alone does not guarantee success. The question of explaining what decisions are made, why the AI tool does what it does, and why it is needed also must be addressed. No algorithm will be clinically appli- cable unless it can be independently verified. AI cannot solve every problem ‘Is the problem that an AI tool resolves worth solving?’ queried Dr Chapiro. ‘AI cannot solve every problem in medicine, and certainly not in IO. AI tool developers need to start with problems that have sufficient data to begin with, of frequently encountered pathologies with high incidence rates rather than rare findings. Other questions to ask include when evaluating an AI tool for IO: Was standardised reporting used? Is the study methodology respecified? What are the sources of data and what were the measurement bench- marks for quality? How is data split between AI tool development and validation? Was the dataset used in model training reflective of clinical reality and intended use case? Was the output of the model interpret- able? Is the performance reproduc- ible and generalisable? Dr Wood discussed similar issues. ‘To realise the promise of AI in IO, we must define the clinical task, which is easier said than done. We must determine where AI is most likely to have a positive impact. We need multi-disciplinary teams to cre- ate AI tools. And most importantly, uniformly standardised data needs to be shared throughout the world, through registries, research data- bases, and reporting.’ Predicting future oxygen requirements Federated learning is one method to train AI models that are gen- eralisable and reproducible with data from multiple sources while maintaining anonymity of the data. Dr Wood cites a collaboration by 20 institutions to create an AI tool that predicts future oxygen require- ments of patients infected with Covid-19. He also explained how a CT AI Covid-19 detection and classification model was trained to be generalisable across nations. Simple AI models can be building blocks that feed larger blocks to create more complex AI models, he emphasised. This is happening now. Like diagnostic imaging, AI tools offer immense promise to improve IO. This is starting to happen now, and probably will steeply accelerate Brad Wood, M.D., is director of the National Institutes of Health Center for Interventional Oncology and chief of Interventional Radiology at NIH Clinical Center and the National Cancer Institute in Bethesda, MD. Julius Chapiro MD PhD, is co-director of the Yale Interventional Oncology Research Lab of the Department of Radiology and Biomedical Imaging at the Yale University School of Medicine in New Haven, CT. in ensuing years. Right now, howev- er, it’s important to keep knowledge about AI utilisation in perspective. The battle of f Low-field MRI: quality images with patient empath e p o r u E s m e t s y S l a c i d e M i h c a t i H : e c r u o S EUROPEAN HOSPITAL Vol 30 Issue 2/21

1 6 R A D I O LO G Y 0.05T to 0.5T scanners expand MRI potential Low-field MRI: Less is more Report: Michael Krassnitzer Currently, two opposing trends can be observed in MRI: on the one hand 1.5T scanners are increasingly replaced by 3T scanners for stand- ard clinical MRI applications. On the other hand scanners with lower and even significantly lower field strength have commercially avail- able in the past two years. A session at this year’s European Congress of Radiology (ECR 2021) took a closer look at low-field MRI scanners with about 0.5T and ultra-low-field MRI scanners with a field strength of 0.05T. ‘The key challenge of those low and ultra-low-field MRI systems is the poorer signal-to-noise ratio,’ said Professor Andrew G Webb, Director of the Centre for High-Field MRI at the University Hospital Leiden (Netherlands). The signal-to-noise ratio (SNR) of a 1.5T scanner is six times higher than in a 0.55T system and even 300 times higher than in a 0.055T system. Each MRI system aims to deliver a good ratio of true signal (i.e. reflecting actual anat- omy) to unavoidable background noise. While low-field and ultra- low-field MRI scanners struggle with SNR, they do offer certain advan- tages over standard MRI scanners. Webb pointed out the areas where low-field MRI scanners receive top scores: firstly, the low purchasing and site costs. Secondly, low-field MRI is less susceptible to arte- facts caused by metal implants and unwanted effects caused by air-filled cavities (lungs, nose). The specific absorption rate (SAR) is about one tenth lower than in 1.5 T scanners, the wave lengths of the magnetic waves is above the critical value for implants. ‘Therefore, we don’t need to be worried about counter- indications,’ Webb underlined. Since gradient acoustic noise is weaker, noise stress for the patient is sig- nificantly lower. Moreover, low-field MRI offers advantages with regard to relaxation times: T1 relaxation time is significantly shorter than in 1.5T scanners while T2* relaxation time is significantly longer. its Magnetom Free. Max Siemens Healthineers set a With n e g n a l r E l a t i p s o H y t i s r e v i n U / s r e e n i h t l a e H s n e m e i S : e c r u o S Low-field MRI scores points in lung imaging new standard. The 0.55T system, launched last year, is suited to examine lung, abdomen, brain and cardiovascular structures. In addi- tion, image-guided interventions can be performed. ‘This is basically a standard MRI scanner,’ explained Professor Matthieu Sarracanie from the Department of Biological Engineering at the University of Basle. The system’s benefits are light weight (3,200 kg) and a com- pact design with the large bore and closed helium-free cooling system. The drawbacks are the poorer reso- lution compared to 1.5T systems and that the system requires the same space and infrastructure as a standard scanner. In the ultra-low-field MRI segment the Hyperfine scanner, launched in 2019, aims to be the new bench- mark with a field strength of 0.054T. Weighing just 630 kg and being the size of a nightstand this mobile system can be used at the bedside, primarily in emergency settings. No site costs, no specific power supply, so specific cooling, even fewer arte- facts caused by medical implants, no undesirable side effects caused by air – all are advantages, accord- ing to Webb. ‘The magnetic waves are so long that the critical length for implants is 75 metres, which is irrelevant for clinical purposes,’ the physi- cist pointed out. ‘This system has enormous potential,’ Sarracanie, a co-founder of Hyperfine. He also pointed at the low costs of approx. $30,000 plus a monthly fee. A dedi- cated site is no longer needed; the carbon footprint is comparatively small and the system can be pow- ered by a standard wall outlet. However, he also points out draw- backs: the system is tailored for specific exams. Currently only brain and extremities can be scanned. Moreover, the Hyperfine is a pro- The battle of fields in MRI Continued from page 15 50 mT human MRI system developed in the Department of Radiology at the Leiden University Medical Center. prietary system which means, for example, that images can only be stored in the company’s own cloud. Open-source projects make MRI available for everyone However, there are also non-com- mercial ultra-low-field systems. With his University Hospital Leiden team, Webb built a 0.05T portable system which is used for neuroimaging. ‘Rather than using several large and heavy magnets we use many small ones – each about the size of a one-cent coin,’ he explained. The dice-shaped 12 mm magnets are mounted on acrylic glass rings – 23 altogether with 2848 magnets. Costs: sand €4,000 plus a few thou- euros more for gradient and RF coils and amplifiers. ‘This is an open-source project. Anybody who is interested can build such a scanner based on our construction plan.’ These sys- tems are also designed for one spe- cific exam, for example paediatric hydrocephalus. The system was designed primar- ily for use in developing countries. ‘Seventy percent of the world popu- lation don’t have access to MRI,’ Webb pointed out. ‘Due to low costs, mobility and standard power supply ultra-low-field MRI scanners can now make MRI available all over the globe.’ ESR/EFOMP – High-field vs low-field MRI: Time for a re-think?, ECR 2021, Vienna, 4/3/2021 patients. His strategy includes the pleasant design of the scanner, sophisticated lighting and ceiling windows that open up to the sky. Additionally, the staff is trained to support patients before, during and after the exam. ‘We aim to make their stay as pleasant as possible – from the first to the last sec- ond,’ Figge underlined. This is more important to him than the technical specifications of the low-field scan- ner. He believes ‘number crunch- ing is irrelevant.’ Nevertheless, he compared his low-field 1.2T with his 1.5T and reported that the 1.5T scanner yields slightly better results than the 1.2T scanner for prostate exams. However, for musculoskeletal imaging and neuro-imaging he did not see relevant differences. In breast imaging the open scanner offers the advantage that women can lie flat on the chest. Dr Mihaly Aradi, director of a private radiology practice in Pécs (Hungary), uses 0.4T as well as 3T systems. ‘With low-field MRI you need to schedule more exam time,’ he explained, adding that ‘the new digital signal processing tech- nologies compensate weaker signals well.’ In his experience, 0.4T scan- ners are as good for routine exams as the 3T system. ‘For MRI scans of the spine both systems deliver pretty much identical quality,’ the Hungarian radiologist said. He sug- gested that indeed ‘every clinic that specialises in spine surgery should Image comparison: MRI scans with Lucent Plus’ IP-RAPID functionality ON/OFF have a 0.4T scanner as well as a 3T scanner.’ Aradi also pointed out that the low-field scanner’s advantage is the reduced number of artefacts: ‘Distortions are so minute that struc- tures are clearly visible even when they are next to metal implants.’ No liquid helium cooling Last, but not the least, Dr Liliana Rechmeier who, with her husband, co-heads a radiology practice in Bad Neuenahr, Germany, stressed the low operating costs of a low-field MRI scanner. Their private practice has to keep the costs under control, she pointed out. ‘Due to the low costs for servicing and electricity and, since no liquid helium cool- ing is required, the operating costs are only about 60 percent of those for a conventional scanner. Over the course of eight to ten years of use, that’s a substantial amount of money.’ In view of the good image quality she also urges colleagues to rethink current prejudices: ‘Low- field MRI is no amateur scanning; it delivers really good radiological images.’ EUROPEAN HOSPITAL Vol 30 Issue 2/21