Vol 33 ISSUE 1/24 • February 2024 THE EUROPEAN FORUM FOR THOSE IN THE BUSINESS OF MAKING HEALTHCARE WORK Technology-based therapy approach Self-administered brain stimulation as a novel tool for depression treatment Including EH@ECR able over the counter in all EU countries and some Asian countries have also shown interest in the product. User endorsements and future outlook Flow has garnered broad endorse- ment from users, although the company acknowledges that indi- vidual outcomes may vary. More research into the longer-term ef- fects of tDCS treatment are needed. The company is looking into ex- panding the duration of the treat- ment regime as well as its geo- graphical reach. Additionally, it explores cooperating with psychia- trists, general practitioners, clinical psychologists, and other healthcare professionals. Flow‘s broader ap- plication to conditions like bipolar disorder, post-traumatic stress dis- order (PTSD) and pain manage- ment shows promise, indicating the potential of stimulating neur- ons in addressing various brain- related disorders. ■ Report: Cornelia Wels-Maug in the The stark reality of depression is evident latest World Health Organization (WHO) stat- istics, indicating that around 280 million people worldwide suf- fered from depression in 2019. National estimates showed a doubling or tripling of depress- ion prevalence in many Euro- pean countries. Unfortunately, traditional treatments for de- pression, primarily medication, have shown limited efficacy. Ac- cording to the Economist Group, in Europe “Only 40% to 60% of patients achieve remission after initial antidepressant treatment, regardless of the chosen drug. Psychotherapy also faces chal- lenges such as limited success, restricted access and long wait- ing times. These factors con- tribute to a substantial number of patients classified as partial or non-responders, leading to a lower quality of life, increased comorbidity, and social and occu- pational disability. With the WHO predicting major depressive dis- order to become the leading cause of disability globally by 2030, the outlook is grim. Given the limited success of tradi- tional therapies, recent years have witnessed the emergence of tech- nology-based interventions for de- pression. One such intervention is transcranial electromagnetic stimu- lation (TMS), a noninvasive use of magnetic fields to stimulate the brain, approved by the Food and Drug Administration in 2008. However, it is considered costly and trials have concluded that TMS is associated with a small risk of seizures. (FDA) technique Another promising noninvasive brain stimulation is transcranial direct current stimu- lation (tDCS). tDCS, with its poten- tial cost-effectiveness, easy ap- plication, and safe profile, applies a weak direct current through scalp electrodes to modify neur- onal excitability. Clinical trials have shown tDCS to have a similar im- pact as antidepressants, with fewer and less severe side effects. To enhance tDCS‘s therapeutic effi- cacy, it has been integrated with www.healthcare-in-europe.com psychological interventions and is mainly used by licensed physicians in clinical settings. Bringing tDCS home with Flow Daniel Månsson, a clinical psychol- ogist with a BSc in Computer Science, recognized the need to make depression treatment more accessible: “I can treat five to eight people per day, but this is not to a large toolbox. We are here to provide something that augments current treatment options or acts as an alternative“, Månsson said. How Flow works Flow combines a medical-grade tDCS headset with an app-based behavioral therapy program, offer- ing a holistic approach to depress- ion treatment. The Flow headset, The Flow 2023 study conducted among 174 participants in the UK and the US showed that 56% of them experienced a complete re- moval of depression symptoms (re- mission) within 10 weeks, whilst an additional 6% were measured to have an improvement in symptoms of at least 50%. No major side ef- fects were reported from the treat- ment group. Overall, a greater than Woman wearing brain stimulation device on her head to treat depression © YouTube/Flow Neuroscience enough given the large number of people affected by it”. Teaming up with neuroscientist Erik Rehn, they founded the neurotech company Flow Neuroscience and developed Flow. Unlike other medical-grade tDCS devices, Flow is designed for home use, empowering individuals to self-manage their treatment. The solution aims to provide effective treatment without the drawbacks of side effects, long waiting times, high costs, or social stigma as- sociated with ap- proaches. While Flow does not aim to replace traditional treatments, it seeks to expand therapeutic op- tions com- plementary approach to existing treatments. “I think it’s extremely important that people have access and provide traditional a with two electrodes placed on the frontal part of the head, delivers a noninvasive treatment by sending a weak current into the left side of the brain, which activates neurons on the left, influencing feelings, thoughts, and inner mental pro- cesses. The overall effect is com- parable to antidepressants but without the associated side effects. Månsson explained: is unique about Flow is that it is a medical-grade, one size fits all tDCS treatment for depression that can be used at home. “ “What Flow’s track record Flow‘s effectiveness and safety are supported by real-world data and clinical evidence from a large ran- domized placebo-controlled trial. treatment compared threefold increase in the odds ratio of substantial improvement with active to placebo/sham stimulation were measured. Flow’s effectiveness sur- passes the average effect size wit- nessed in trials for the 21 best-sell- ing antidepressants in the UK and the US. Scaling up efforts Flow has received CE approval in the EU, Great Britain, and Brazil, with FDA approval expected by mid-2024. Currently, Flow has ap- proximately 17,000 users and its largest market is the UK. World- wide over 150 clinics deploy Flow, most of them based in the UK, some are in Switzerland, Germany, and Brazil. However, Flow is avail- CONTENTS CONTENTS HEALTH HEALTH IT EH@ECR EH@ECR LABORATORY/ LABORATORY/ PATHOLOGY PATHOLOGY 1-3 4-9 10-12 www.healthcare-in-europe.com

HEALTH IT Timely project Improved care after heart attack on AI-powered telemedicine platform 3 t c e j o r P - y l e m i T © offer positive reinforcements to keep motivation up. Artificial intel- ligence processes large amounts of data and detects arrhythmias at an early stage. ‘A change in sleep rhythm or an increased stress level indicate arrhythmias,’ says can t c e j o r P - y l e m i T © tracker provides medical examination, the partici- pants are equipped with various wearables for six months. In ad- dition to a portable, wireless 3-lead ECG, they received a blood press- ure monitoring device to measure resistance in the vessel and the speed of blood flow. Moreover, an activity in- formation about stress levels, ex- ercise and activity calories. The ‘heart’ of the study is the Timely app. It provides the patients with information on all aspects of rehab such as diet, stress management, weight loss or exercise and each participant can set their personal goals. Personal messages, push no- tifications and a chatbot regularly Schmitz and adds that in such cases the team ‘responds directly and contacts the patients who need immediate support.’ Digital dashboard Via a digital dashboard, physicians, therapists and case managers ac- cess all data and the individual risk profile of each patient, which is Boris Schmitz PD Dr Boris Schmitz is senior scien- tist at the Königsfeld Clinic, Center for Medical Rehabilitation. A biol- ogist by profession, he also holds a doctorate in sports science. He spent most of his professional career in Münster as a research assistant, pro- ject and laboratory manager as well as a private docent. Schmitz is a member of the European Society of Cardiology (ESC) and works as a re- viewer for various journals. In automatically evaluated and inter- preted. the event of ab- normalities – for example, when someone has not been exercising for a long time – the person is called on the phone to adjust the therapy if necessary or to arrange other kinds of help. The data is transmitted automatically via the mobile network and via secure sys- tems in Germany, Schmitz em- phasizes. After six and twelve months, the study participants return to the rehab facility for further extensive examinations. The last patient is expected to undergo this final exam by mid-2024. The primary endpoint of the EU-funded study is a biomarker score, which reflects the risk of mortality over the next ten years, and a test to assess physical performance. In addition, changes in diet and exercise be- havior are analyzed. Schmitz wants patients to continue using the app for the rest of their lives, whenever they need it. ‘Intensive use over six months is enough to internalize the content. However, there are always setbacks in life, perhaps career changes or other illnesses that make it necessary to return to work. My vision is that the app can be prescribed as a digital health application, a so-called diga, and be used not only for aftercare, but also for prevention.’ ■ Report: Sonja Buske Coronary artery disease (CAD) causes more than 1.8 million heart attacks each year in Eu- rope, 300,000 in Germany alone. In most cases, rehabilitation helps the patients to return to their everyday life. But what happens once the patients have left the rehab clinic PD Dr. Boris Schmitz the Timely project to find out. The project aims to develop a patient- centered platform that provides early cardiac risk prediction and supports prevention and inter- vention in post-rehab care for CAD patients using eHealth and artificial intelligence. launched In a rehabilitation clinic, CAD pa- tients learn how to change their life styles in order to be able to continue living with the disease for as long as possible and without re- infarction. A healthy diet, exercise, and stress reduction are the most important issues, but once back in their home many patients don’t fol- low through with the behavioral changes they learnt in rehab. ‘The first six months after rehab are cru- cial,’ says biologist and sports scientist Schmitz since ‘only during this period behavioral changes will become a habit.’ A senior scientist at the Center for Clinical Rehabilitation at the Kö- nigsfeld Clinic, Schmitz wanted to find out how CAD develops after treatment has been completed in order to be able to influence the patients’ state of health post rehab. But post-rehab care does basically not exist: the guidelines require only a single follow-up visit at a cardiologist’s after six months. Too little and far too late, says Schmitz. With his randomized, controlled clinical study, he wants to show that cardiovascular risk decreases with the help of the Timely plat- form. For this purpose, 360 adult patients were recruited who had undergone rev- ascularization after being diag- nosed with CAD as well as follow- up care as part of cardiac rehabilitation. The project is con- ducted in three study centers in Germany, Spain and the Nether- lands and 13 partners from indus- try, the outpatient sector and the clinical environment are participat- ing in the development of the plat- form. myocardial Equipped with wearables and app In order to design the platform with the patient front and center, Schmitz onboarded the potential participants right at the beginning of their rehab in the form of a ‘liv- ing lab’. ‘We interviewed the pa- tients’, explains Schmitz, ‘asked them about their wishes and needs, tried out several approaches and tested the user-friendliness of the app, among other things. It was important to us that all participants felt comfortable with the app – no matter how old they are and re- gardless of their previous technical knowledge.’ After a thorough www.healthcare-in-europe.com

EH@ECR 5 Imaging in characterisation and classification of tumour types Taking a closer look at breast cancer Breast cancer has no “one size fits all” therapy approach: sub- types differ significantly in ma- lignancy, progression, and treat- ment response. Therefore, the more is known about the type of carcinoma in a patient, the better the outcome. At the annual scien- tific EUSOBI meeting in Val- encia, Dr Ramona Woitek pointed out the potential of novel imaging techniques and computational image analysis for classification char- acterization of breast cancer as well as treatment response pre- diction. and Molecular imaging can provide valuable insights into cancer char- acteristics, Dr Woitek, a Professor of Radiology at the Research Center for Medical Image Analysis and AI (MIAAI) at Danube Private University in Krems, Austria, out- lined. It can help differentiate cancers via hallmarks such as in- creased self-proliferation, deregu- lated cellular metabolism and tu- mour vasculature. In a best-case scenario, imaging can help identify – potentially actionable – mu- tations, and select those patients who will benefit the most from undergoing tumour tissue sequenc- ing. as Beyond standard-of-care imaging such conventional mam- mography, ultrasound and MRI, functional imaging can provide more specific information, the ex- pert explained: 18F-FDG PET is mainly used in metastatic breast cancer, but can also help under- stand a tumour’s microenviron- ment and biology, while 18F-FLT PET and 18F-FMISO PET show cell proliferation and hypoxia in tu- mours, respectively. Another tool that can help make better treat- 89Zr- ment Trastuzumab, a tracer that ident- ifies heterogeneity between metastatic lesions. Waiting in the wings are advanced MRI techniques using hyperpolarised carbon-13 (13C) and sodium (23Na), which show promise in treatment response prediction and early re- sponse detection by distinguishing between actual progression and pseudoprogression. decisions is tumour Additional insights from semantic descriptors and heatmaps Semantic descriptors in MRI im- ages serve as an effective way to distinguish between breast cancer subtypes, Woitek pointed out. ‘It has been shown in several studies that some of these descriptors are more frequently found in certain subtypes.’ These distinctions can provide valuable and actionable in- formation, especially in settings where assessment of the KI-67 marker ‘For example, irregular borders and persistent enhancement are most frequently found in Luminal A cancer types, while Luminal B is characterised by more heterogen- eous enhancement. HR-/HER2+ breast cancers typically feature smooth borders, washout, and are more frequently associated with multifocality.’ These insights can be is not available. www.healthcare-in-europe.com shearwave (CNN), high-quality clinical features, Woitek reported. Using convolutional neural net- works in- formation can also be extracted from multimodal ultrasound im- aging (B-mode, colour Doppler and elastography). ‘These networks use deep learning to find the relevant features to clas- sify the tumours by themselves, and they are getting better and better at this. Usually, the more types of data from different sources and the more modalities are put together, the higher the chances of developing really good predictors.’ As with most AI algorithms, this approach is somewhat of a black box. However, this can be over- come using special heatmaps. These class activation maps show how much any pixel of a given diagnostic image contributes to the algorithms classification or pre- diction, leading to interpretable re- sults. For example, if the algorithm derives its findings mostly from pi- xels which are far away from the actual tumour site, it is likely that the interpretation is not trust- worthy. On the other hand, if the heatmap stays focused on the cancer, the result has high pre- dictive value. A more profound understanding of cancer types ‘Usually, oncologists’ decisions on whether or not to give targeted anti-HER2 treatment are based on single biopsies from single meta- static sites or even from the pri- mary tumour,’ Woitek pointed out. ‘But scans and tracers like this really allow us to get an idea about the heterogeneity of cancer, the re- Molecular imaging can help identify heterogeneity in HER2 expression in metasta- tic breast cancer: 18F-FDG (left) and 89Zr-trastuzumab PET scans (right) of two pa- tients © Bensch et al., EJNMMI 2018 (CC BY 4.0) further exploited via radiomics, where quantitative features such as shape, volume or texture of a tu- mour can be extracted. Using dy- namic contrast-enhanced MR im- aging (DCE-MRI), this provides high-accuracy es- pecially for distinguishing luminal B-type cancer from other forms, often outperforming semantic or information ceptor status and the expression of HER2 in different sites of the body, and that can actually have a huge effect on treatment decisions.’ image Computational analysis, combined with novel imaging tech- niques, holds great potential to help characterize cancers, the ex- pert summarized. ‘Especially with novel treatments, our focus on functional techniques will grow. It is our task to identify the key ques- tions we want to answer in the pre- diction, classification and char- acterization of cancer, and how this can have the biggest impact on our patients and improve their out- comes.’ ■ Report: Wolfgang Behrends Ramona Woitek Dr Ramona Woitek is Professor of Radiology at the Research Center for Medical Image Analysis and AI (MIAAI) at Danube Private University in Krems, Austria. Her work focuses on the analysis of medical imaging data for the development of quanti- tative biomarkers, and the use of AI to predict the presence of disease, its progression or treatment response. www.newtom.it NEWTOM 7G NEXT GENERATION CONE-BEAM CT NEW IMAGING HORIZONS. Featuring a large gantry aperture, the NewTom 7G is the most advanced CBCT device on the market. From in-depth detail to the big picture. ULTRA-DETAILED MULTI-DIAGNOSTICS Localised analysis over the entire body. Ray2D and 3D imaging up to 90 µm, also with radiocontrast agent. Artifact reduction and movement analysis using CineX and Cine-Scout. The eXtra Functions protocol extends the field of view longitudinally to analyse anatomical structures such as the spine and limbs. TECHNOLOGICAL EXCELLENCE High power generator (120 kV – 20kW). High-sensitivity 3D panel and innovative algorithms for volumetric reconstruction. 77 cm gantry aperture. ERGONOMICS AND PRACTICALITY Fully motor-powered table and 10” touch-screens, front and rear. Optimised examination flow thanks to certified NNT software with processing, sharing and RIS/PACS connectivity functions. MAXIMISING PATIENT CARE The patient lies comfortably on the table and X-ray doses are always proportioned to the patient’s build and the examination type thanks to SafeBeamtm technology. Stay tuned for groundbreaking updates on the technology at ECR 2024 | 314 - Expo X3 AND THERE’S MORE TO COME!

EH@ECR 7 Additional findings from the MIPA study Benefits of preoperative breast MRI Ordering preoperative breast MRI exams of diagnosed breast cancer patients used to be con- troversial: Did they aid surgical planning better than the com- bination of mammography and breast ultrasound Or did their findings cause overtreatment, specifically mastectomy, when breast-conservation surgery would have sufficed New re- search has now settled the issue. The real-world analysis of nearly 6,000 patients having surgery as the primary treatment for biopsy- proven breast cancer in 27 centres in Europe, Australia, the United Kingdom, and the United States showed that breast MRIs did not cause overtreatment. And as study data continues to be analysed, it confirms that when breast MRI is ordered, primarily by radiologists and/or breast surgeons, it posi- tively impacts the quality of staging and choice of surgical treatment. the MIPA Although data from (Multicenter International Pros- pective Analysis) observational study revealed that 9% more mas- tectomies were performed than in- itially planned by surgeons for pa- tients who had a bilateral contrast-enhanced breast MRI, this subgroup of women were at higher risk than those who just had con- ventional breast imaging. In fact, for a handful of women, breast MRI confirmed the feasibility of breast conservation surgery instead of planned mastectomy. Screening lead to increased mastectomy rate Initiated by the European Network for the Assessment of Imaging in Medicine, and funded with an un- conditional research grant from Bayer AG, the study recruited 7,245 patients ages 18 to 80. Of these, the data of outcomes for 5,896 were able to be analysed. Based on the recommendations of their phys- icians and the protocols of their treatment centres, 53.1% had a breast MRI supplementing con- ventional imaging. In addition to a larger number of women in the breast MRI cohort being younger and pre-menopausal, 17% more had dense breasts, 15% more had tumours larger than 20 cm, and 9% more had a higher rate of invasive lobular histology than the con- ventional imaging group. Mastectomy had already been planned for 547 (22.4%) prior to the breast MRI. Post MRI screen- ing, this number increased to 791 women (32.4%), and ultimately 773 were performed based on surgical recommendation. An additional 50 women self-opted to have a mastectomy. By com- parison, only 398 (14.4%) of women had planned mastectomies based on conventional imaging findings. This ultimately increased to 432 women, with 19 of this group electing to have a mastec- tomy instead of recommended breast conservation surgery. eration statistics. The MRI group had a significantly lower reoper- ation rate, at 8.5%, compared to 11.7% in the non-MRI group. By itself, this finding has major im- plications for the welfare of the pa- tients and medical cost savings as well. Identifying cancers that otherwise may be missed In a subsequent analysis published in European Radiology, Sardanelli and colleagues compared mastec- tomy rates and reoperation rates among three groups: ‘MRI did not alter the surgical plan- ning based on conventional im- aging in 84.5% of cases,’ wrote principal investigator Prof. Fran- cesco Sardanelli, MD, of the Uni- versità degli Studi di Milano, Italy, and co-researchers in European Radiology. The surgical planning value of breast MRI is also reflected in reop- • women for whom breast cancer had already been identified and breast MRI had been ordered for ipsilateral staging, contralat- eral screening, and surgical in- tent (79.7%), • women for whom a diagnostic exam had been ordered, pri- marily as follow-up to equivo- cal conventional imaging find- reinforce the fact that breast MRI identifies cancers that may not be seen -- or clearly seen – on mam- mography and breast ultrasound,’ they commented. Impact on DCIS in a patient matched analysis The most recent MIPA data-related study, also published in European Radiology, evaluates the influence of preoperative breast MRI on mas- tectomy and reoperation rates in patients with pure unilateral ductal carcinoma in situ (DCIS). Six hundred eighteen patients selected from a group of 1005 with biopsy- confirmed DCIS were equally matched 1:1 to eliminate bias. Three hundred nine had breast MRIs and 309 did not. The findings again confirmed breast MRI has a positive impact on surgical outcomes. Although the overall mastectomy rate was 23.3% in the group of patients who had breast MRI exams prior to surgery compared to 17.8% in the con- ventional imaging group, the reop- eration rate of the MRI group was dramatically lower. At 23.3%, the reoperation the con- ventional imaging group was more than double the 10% of the MRI group. ‘The increase in the overall mastectomy rate engendered by preoperative MRI is less than half the corresponding reduction in reoperation rates,’ the authors ad- vise. ■ rate of Report: Cynthia E Keen i e i a o r t e p A i i r e l a V - m o c . e b o d a . k c o t s © ings (16.6%), • and women for whom breast MRI was ordered for breast cancer screening (3.7%). screening MRI subgroup The proved to be women at the highest risk: younger, with complex cases, larger lesions, the highest rates of small and single-focus cancers, and the highest percentage of familial or personal genetically proven in- creased risk. The researchers were not surprised that this highest risk subgroup had the highest rate of mastectomies, at 39.5%. By com- parison, the rate of mastectomies for the diagnostic group was 24.1%. the Reoperation rates for close or posi- tive margins among three groups were not significantly dif- ferent, ranging from 8.2% to 10.5%. However, the difference between the MRI and conventional imaging only groups was. ‘These findings Using low-frequency electromagnetic fields Detecting breast cancer in two minutes The innovative technology aims to be more accurate as well as cheaper to provide than today’s most common diagnostic tools such as X-ray mammography, ultrasound and magnetic reson- ance imaging (MRI). Test runs have been completed in two minutes and used less energy than a smartphone. It would also be safer than X-rays, which expose patients to high-level radiation that can damage DNA and cause cancer. ‘We are coming very close to providing a method for breast cancer detection at an early stage that is inexpensive and harmless for women,’ said Dr Omar Ramahi, lead researcher and a professor in Waterloo’s Department of Electrical and Computer Engineering. ‘We’re trying to make a serious con- tribution to women’s health and create an alternative that is clini- cally and commercially viable.’ Ramahi began exploring new ways to detect early-stage breast cancer in 2001 and, for the past five years, has been studying the potential of low-frequency electromagnetic waves. Ramahi and his research team made what he described as “a big discovery,” namely that very low electro-magnetic frequencies “travel in straight lines.” The diagnostic device they created somewhat mimics X-ray mam- mography but without its draw- backs. In place of X-rays, low- frequency electromagnetic energy is emitted from an antenna, like the one found in a smartphone. Once the energy penetrates a patient’s breasts, it is picked up by a meta- surface, or circuit board, consisting of end-to-end pixels, each pixel acting as a receiver. The AI then in- terprets the pictures coming from the circuit board, removing the need for a human technician to re- view the results. The technology can locate the size and location of a tumour, even in breasts with dense tissue, something current diagnostic systems can miss. Ramahi has tested his system on breast phantoms. Next steps are to develop a system that can be tested on human subjects. ■ Source: University of Waterloo www.healthcare-in-europe.com Y u r g u i d e im a g i n g e c h n l g y a n d i n f rm a i c s i n E u r p e For more information, visit healthcare-in-europe.com

EH@ECR 9 Diagnostic imaging collaboration Canon Medical and Olympus team up for endoscopic ultrasound Two major players in diagnostic imaging combine their expertise in the field of endoscopic ultra- sound (EUS): At a press con- ference in Tokyo, Japan, Canon Medical Systems Corporation and Olympus Corporation an- nounced collaboration agreement, which includes intro- duction of a new, jointly devel- oped system takes ad- vantage of both companies’ experience in endoscopy and so- nography. their that As part of the collaboration, Canon Medical will develop and manu- facture diagnostic ultrasound sys- tems for EUS procedures, while Olympus will perform the sales and marketing efforts. With the traditional handshake, Frank Dre- walowski, Head of the Endoscopic Solutions Division at Olympus, and Toshio Takiguri, President and CEO of Canon Medical, affirmed the agreement. Both expressed their enthusiasm for the collabor- ation, highlighting the two com- tech- panies‘ nological the respective fields of endoscopy and diagnostic ultrasound. In this new constellation, both manufacturers aim to expand their business in diagnostic imaging markets, in- cluding Europe. rich history of in innovation Image quality to rival CT and MR imaging In a first demonstration of the co- operation, the companies unveiled a combined setup of Canon‘s Aplio i800 diagnostic ultrasound system Frank Drewalowski, Executive Officer and Endoscopic Solutions Division Head, Olympus Corporation (left) and Toshio Takigu- chi, President and CEO, Canon Medical Systems Corporation (right) at the global press conference in Tokyo. © Olympus for EUS and Olympus‘ Ultrasound Endoscope. Equipped with Canon’s proprietary imaging enhancement features such as differential tissue harmonic imaging (D-THI), superb (SMI), micro-vascular imaging shear-wave elastography (SWE) and attenuation imaging (ATI), the system can be used as a tool for a wide imaging ap- plications. range of Taking full advantage of the tech- nology, the combined EUS setup is designed to provide image resol- ution and quality compared to CT and MRI, Takiguri said. With the new device, the companies hope to enhance diagnostic efficiency, for example in the detection of smaller lesions in pancreatic cancer and bring benefits in other pan- creatobiliary diseases, which affect more than 60 million patients worldwide, added Drewalowski. Endoscopic ultrasound – which combines a sonography probe to the tip of an endoscope to assess diseases of the gastrointestinal tract and nearby organs, such as the liver, pancreas and gallbladder – has a range of diagnostic ap- plications, said Hironobu Kawano, Endoscopic Solutions Division Head at Olympus. However, the modality is often considered tricky, requiring a high level of expertise in sonography. The enhanced im- aging features in the new EUS sys- tem allow for more consistent image quality and improve access for difficult-to-reach areas of the body, he pointed out. Going into more detail on the tech- nology, Akihiro Sano, General Manager of the Ultrasound Systems Study examines diagnostic quality Special probes for obese patients Division at Canon Medical Systems, showcased the strengths of the em- bedded features: D-THI provides clearer images at greater depths, while SMI shows more and thinner vessels con- ventional Doppler ultrasound, en- abling a better visualisation of blood flow. Further, ATI is used to quantify and color-code changes in the tissue composition (e.g., fat content of the liver). compared with Coming to European markets this summer ‘We have a long history in diag- nostic ultrasound systems and have developed unique technologies that contribute to the early detec- tion of diseases in cooperation with doctors around the world,’ Takiguri said. ‘We are very pleased to collaborate with Olympus, a leading endoscope manufacturer, in the EUS field, which we have not been able to cover adequately until now, to deliver superior sol- utions to more patients.‘ Drewalowski added: ‘In the field of EUS, diagnostic ultrasound systems are very important, and we are pleased to have a cooperative rela- tionship with Canon Medical, which has advanced ultrasound imaging technology.’ Introduction of the EUS system to European and Japanese markets is planned for June 2024, Takiguchi said, while the timeline for US mar- kets is still pending. (WB) ■ Scan now for a Virtual Experience! A new study at the University of Leipzig Medical Center with sup- port from the Helmholtz In- stitute for Metabolic, Obesity and Vascular Research (HI-MAG) shows how obesity in patients leads to decreased image quality of ultrasound scans of the liver and kidneys. Ultrasound of the abdominal or- gans is a central diagnostic tool and is recommended as the first- line approach for many medical conditions. However, the accuracy of this method is usually limited in obese individuals because the im- aging quality of anatomical struc- tures is impaired. To date, the de- gree of obesity at which ultrasound diagnostics are no longer suffi- www.healthcare-in-europe.com ciently precise has not been suffi- ciently researched. the increased, In the clinical study, researchers found that the quality of ultra- sound scans of the liver and kid- neys was significantly lower in obese patients. ‘The results showed that as the body mass index of the patients image the ultrasound de- quality of creased. It also became clear that modern probes with matrix tech- nology improve the quality of ultrasound imaging,’ says Professor Thomas Karlas, head of the study. The matrix probes have an im- proved transmitting and receiving performance and therefore achieve greater penetration depth into the tissue. This allows for better diag- nostic accuracy in high-risk pa- tients. The performance of matrix probes in obese individuals has not yet been scientifically validated. The current data show that these relatively expensive special probes for ultrasound equipment can pro- vide significant added value. The 40 participants in the study under- went a highly standardised ultra- sound examination of the ab- domen with three different probes. The liver and right kidney were examined in people with varying degrees of obesity, and the quality of the ultrasound imaging was scored. Source: Leipzig University ACCUTRON® CT-D VISION. SIMPLY MORE. Accutron® CT-D Vision. The diagnostics specialist that can do more. More comfort, more mobility, more operating safety. More integration through Injection Data Sharing with RIS/PACS connection. View now at medtron.com Hauptstrasse 255 · 66128 Saarbruecken www.medtron.com

LABORATORY/PATHOLOGY 11 Improving quality and efficiency Preparing for artificial intelligence in clinical laboratories Some year in this decade, AI tools will become ubiquitous within clinical laboratories. AI has the potential to increase the accuracy of laboratory testing and improve the quality and effi- ciency of operations and service of testing labs. Clinical laboratorians must prepare to help lead this initiative, for their knowledge will be the key to suc- cessful implementation. They need to learn how AI algorithms are de- veloped and validated, how to jus- tify and analyze impact from the perspective of clinical laboratory medicine, and how to implement them to best benefit the patient and the hospital. Several scientific sessions at the 2023 Association for Diagnostics and Laboratory Medi- cine (formerly the AACC) Annual Scientific Meeting focused on this topic. AI is in its naissance in medical lab development, and this was very ap- parent in the technical exhibition of the ADLM/AACC meeting. David McClintock, MD, Chair of the Divi- sion of Computational Pathology and AI in the Department of Lab- oratory Medicine and Pathology of Mayo Clinic, pointed out that only 30 out of the 941 exhibitors at the meeting included the terms “ar- tificial intelligence“ (AI) and/or “machine learning“ (ML) in their product/company descriptions on the AACC exhibitor website. Ten companies included ‚analytics‘ in their description, but only four were separate analytics-based com- panies selling clinical lab AI/ML software. ‚This is an emerging space, just as radiology PACS was 30 years ago,‘ McClintock said. ‚Now is the time to learn about it or perhaps even start to developing models that can benefit your lab.’ Uses of AI/ML in the lab There are numerous ways that both simple and complex AI tools can aid a clinical laboratory. These include: • Automated spectroscopic data analysis and disease detection; multivariate analysis of disease conditions; test interpretation; • Digital image analysis for microbiology, haematopathol- ogy, immunology, and foren- sics; • Data entry automation for spe- cific tasks and processes; • Creating standardised reports for lab test results and auto- mated entry into Laboratory In- formation System (LIS); test results • Minimising laboratory testing for inappropriate test orders, predicting from other available data on patient chart, and reducing redundancy and duplication of lab tests, based on prior type and date of tests already performed; www.healthcare-in-europe.com concurs. ‚Will AI be replacing staff in labs ‘ he asked rhetorically. ‚Probably not. The practice of lab- oratory medicine has been con- stantly evolving, often due to in- creasing automation. I think that AI will be another tool in our tool bag, to aid in efficiency and quality control. With a steady increase in an aging population in the United States, we are going to need all the help we can get.‘ ■ Report: Cynthia E Keen Mayo Clinic clinical kidney stone workflow with machine learning. © Dr. David McClintock, Mayo Clinic • Data analytics for laboratory operations planning, such as predicting volume workflow, employee require- ments, etc.; staffing • Identifying and alerting for ab- normal test results; • Auto-verification of test results for quality control. Automating spectral analysis for kid- ney stones and fecal analysis Mayo Clinic spent six years devel- oping an AI model to automate the spectral analysis of stones passed by patients. It is based on the clas- sification of 708 unique kidney stone types. In the first 90 days of implementation, commencing April 2023, 20,000 stones have been re- viewed, 40% of which were newly able to be bulk auto-released to pa- tients’ medical records (EHR). the model was Before im- plemented, the conventional work- flow began with cleaning and dry- ing the stone, after which it was ground into a fine powder and manually analyzed with FTIR spec- troscopy. A technician manually entered the results into a LIS, fol- lowed by a second technician re- viewing the interpretation. Only then are results uploaded to a pa- tient’s electronic medical record (EHR). AI tool to automate multiple processes After evaluating how AI could im- prove workflow, reduce costs, and increase efficiency, the Kidney Stone lab at Mayo Clinic Rochester, in conjunction with an AI team from Mayo Clinic Florida, created an AI tool to automate multiple processes following FTIR spectra analysis. The AI model was trained on 70,000 kidney stone spectra and validated with 16,491 kidney stone spectra. Quality assurance required 81,517 kidney stone spectra. ‚This is a lot of data, which took a lot of then the it and work, and a lot of computing time,‘ commented McClintock. ‚But now the process is automated and is achieving our expectations. If a stone is not complex, the AI system classifies in- formation is automatically entered into the LIS and subsequently re- leased to the patient’s EHR. When it identifies a stone as complex, the results produced are manually flagged for review by a technician. The lab is now saving a lot of time, which equates to tangible cost sav- ings and opportunities for labora- tory staff work reduction/redi- rection.’ Other applications of AI in the clinical labs also exist, such as commercially developed AI tools in clinical microbiology to detect fecal ova and parasites (O&P). For most labs, up to 95% of O&P cases can be negative, and thus the pro- cess of reading slides can be mon- otonous. The investigational AI- assisted screening tool (Techcyte, Orem, UT) uses a convoluti onal neural network to identify and count parasite cysts and tropho- zoites, yeast, and red and white blood cells and groups them by class. Techcyte claims the tool is five times more sensitive than manual examination, with a sensitivity of 98.9%. It produces findings within 30 seconds instead of the average five minutes, automatically upload- ing negative findings to a LIS. Posi- tive samples are flagged for tech- nologist review and assessment. At Mayo Clinic, this test has just been implemented, with initial impres- sions positive by laboratory staff who can now remotely review slides, improving employee experi- ence. What clinical lab managers need to think about ‚Don’t get enamoured with AI for your lab,‘ McClintock cautioned. ‚Always remember a clinical lab’s primary objective: to deliver the right information to the right per- son at the right place and right time in the right way. No system today can integrate all potential outputs of AI tools,‘ he em- phasized. ‚LIS, EHR, and middle- ware solutions take considerable effort to integrate with any AI tool without encumbering the pathol- ogist, laboratorian, or clinician. For starters, you need to think about data governance, data pipelines, regulatory guidelines, ethical re- view, custom programming and coding, computational computing power, either locally or in the cloud, cybersecurity, and risk man- agement. ‘Don’t forget the cost and availabil- ity of AI maintenance, support and quality control, which all require new IT support skills. There are also AI specific tools, such as al- gorithm drift. And then, in the end, will the AI tool save you enough money or at some point be re- imbursable so that it can pay for itself ’ to Understanding the barriers and challenges Experts recommend that lab man- agers need to focus on generating clinical evidence for AI benefits and understand the barriers and challenges implementation when they select a tool. New train- ing on AI is essential. In general, experts recommend implementing the new tool while still keeping the existing system functioning, to give practitioners time to learn and get comfortable with AI while main- taining the status quo as back-up. McClintock also described a new framework for clinical AI life cycle implementation, from idea gener- ation to final validation, go-live, and system maintenance. By adopt- ing a similar AI lifecycle, he en- couraged attendees to embrace the potential of AI in their labs. Co-presenter Christopher Lee Wil- liams, MD, Assistant Professor of Pathology at the University of Ok- lahoma Health Sciences Center, Christopher Lee Williams Christopher Lee Williams, MD, is the Director of Informatics in the De- partment of Pathology at the Univer- sity of Oklahoma Health Sciences Center in Oklahoma City, where he also serves as an Assistant Professor Dr Williams’ current research inter- ests include how to operationalise AI tools for analysis and reporting in the laboratory setting, and in opti- mizing UI/UX design for laboratory workflows. David McClintock David McClintock, MD, is the Chair of the Division of Computational Pa- thology and Artificial Intelligence within the Department of Laboratory Medicine and Pathology at Mayo Clinic in Rochester. His primary clini- cal interests include clinical in- formatics, laboratory workflow optimisation, digital pathology im- plementation, analytics, and clinical ML/AI model deployment. His re- search interests include understand- ing the role and effects of digital pathology within the clinical labora- tories and the use of AI and ML for improved diagnostics, more efficient workflows, and better patient out- comes.