Venue: Large Lecture Hall (2nd floor, Room 047, University Hospital)

Abstract

I will talk about the application of automated, quantitative image analysis in combination with machine learning and artificial intelligence in radiology. These techniques have the potential to revolutionize clinical routine. I will illustrate how machine learning and artificial intelligence can be used in radiology to solve typical problems, using examples from my research group.

However, there are also challenges and difficulties that can arise when implementing these technologies into clinical routines, such as data issues, and data and computing infrastructure. I will also highlight potential solutions and strategies to address these challenges.

I hope that after my talk, you will have a better understanding of how clinical data science can improve radiological diagnostics and thus have a positive impact on patient care.

Speaker: Prof. Dr. Michael Ingrisch (Clinical Data Science in Radiology, LMU Clinic Munich)

Short biography

I am leading the group for Clinical Data Science at the Department of Radiology. We employ advanced statistics, machine learning and computer vision techniques in the context of clinical radiology to enable fast and precise AI-supported diagnosis and prognostication. Open science and reproducible research in this field is highly relevant, especially with deep learning or machine learning. While it is easy to share analyses and code, the sensitive nature of medical images and associated clinical data poses challenges with respect of public data sharing. I believe the Open Science Center provides the ideal framework to address these challenges.

Current position: W2 Professor for Clinical Data Science in Radiology, Department of Radiology, University hospital, LMU Munich. A selection of scientific activities and memberships:

Since 2022: Fellow of the Konrad Zuse School of Excellence in Reliable AI (relAI).

Since 2021: PI in the Munich Center for Machine Learning (MCML); Member of the focus area “Next Generation AI”, Center of Advanced Studies; Coordinator of the „Clinical Open Research Engine“(CORE) established as shared, collaborative high performance computing environment at LMU Klinikum (Profs. Ingrisch, Hinske).

Venue: Lecture Hall N2045 (Faculty of Applied Computer Science)

Abstract

Speaker: Dr. Thomas Wendler Vidal (Technical University of Munich)

Short biography

Venue: Small Lecture Hall (2nd floor, Room 048, University Hospital)

Abstract

In this lecture the origin and development of the National Intensive Care Evaluation (NICE) registry, a Dutch quality registry including all ICU patients in the Netherlands will be presented. The lecture will explain which data is included, to what extent it is possible to use routinely collected data from the EHR to fill the NICE registry, and which measures are taken to optimize data quality and reduce administrative burden. The primary aim of the NICE registry is to support ICUs in monitoring and improving quality of care. Benchmarking or audit and feedback, i.e. the strategy that intends to encourage professionals to change their clinical practice by providing professional performance based on explicit criteria or standards back to professionals in a structured manner, is an important strategy of quality registries in realising quality improvement. The lecture will include examples on the effectiveness of audit and feedback, among which an RCT on actionable indicators and a toolbox for improvement activities in the domain of pain management. The secondary aim of the NICE registry is to provide an infrastructure to research medical and methodological medical informatics research questions. If time allows, some examples out of ~180 scientific journal papers and 15 PhD theses based on the NICE registry will be presented.

Speaker: Prof. PhD Dr. Nicolette F. de Keizer (Amsterdam Unviersity Medical Center)

Short biography

Nicolette de Keizer has a master and PhD in Medical Informatics of the University of Amsterdam. She has a special interest in reusing routinely collected data to evaluate quality of care and impact of health care information systems. She is one of the founders of the National Intensive Care Evaluation (NICE) quality registry for Dutch intensive care units and of the post-graduate Master Health Informatics. She is appointed Principle Investigator in AmsterdamUMC and full professor of the University of Amsterdam. She is chair of the department of Medical Informatics, one of the leading Medical Informatics departments in the Netherlands. With her department she provides Medical Informatics training and research at BSc, MSc and PhD level.

Nicolette is internationally recognized as an expert in quality assessment, audit and feedback and standards for data reuse and she acted for years as an expert for the Dutch National ICT institute in health care and the international SNOMED CT quality committee for terminology development and maintenance. She was co-chair of several international medical informatics conferences and workshops. She is chair of the Dutch cooperation of Quality Registries and member of the Dutch data governance committee for quality registries. She supervised 25 graduated and 13 ongoing PhD students and over 40 Bachelor and Master students during scientific research projects. She published over 300 scientific research papers and book chapters, and was co-editor of the book “Applied interdisciplinary theory in health informatics: a knowledge base for practitioners”. Her H-index is 63 (Google Scholar)

Venue: Lecture Hall N2045 (Faculty of Applied Computer Science)

Abstract

This talk gives a short overview on the literature on prognostic and predictive models for patients with Multiple Sclerosis. Besides a methodological overview, I discuss issues on the performance quality of such models and how how they reflect patient interests. Own experience will be reported gained from model development on data available within the DIFUTURE consortium as well as data from the french national MS Registry (OFSEP). At the end, we have to discuss how to cope with a very unsatisfying overall perspective:

Do we need better and more data? Do we need more advanced methods?

Do we need a deeper understanding of the disease?

Speaker: Univ. Prof. Dr. Ulrich Mansmann (Director of the IBE, LMU Medical Faculty Munich)

Short biography

Prof. Dr. Ulrich Mansmann is head of the Department of Medical Information Sciences, Biostatistics, and Epidemiology at the Ludwig-Maximilian’s University Munich. His background is a PhD in Mathematics (1990, Technical University Berlin). In 1991 he started to work as assistant professor at the Medical Center of the Free University in Berlin working in Clinical Epidemiology with main focus on clinical trials and prognostic studies. In 2000 he moved to the University of Heidelberg and cooperated with colleagues from the German Cancer Center (also located in Heidelberg). He was member of the German National Genome Research Network (NGFN). In January 2005, Prof. Mansmann was invited as head of Department at the Institute of Medical Information Science, Biometrics, and Epidemiology (IBE) at the University of Munich (LMU). The IBE represents methodological research on a wide spectrum of Medical Sciences: Public Health, Epidemiology, Clinical Epidemiology, as well as Molecular Medicine. This position allowed deepening and extending activities in the fields of research which are at the heart of his interest. Especially, his experiences in prognostic oncological research combined with activities in the field of Bioinformatics as well as clinical trials were helpful to enter the fields of translational medicine as well as personalized medicine. Prof. Mansmann is also spokesman of the LMU’s program in Public Health and Epidemiology.  From 2012 to 2020 he served as a DFG collegiate. Since 2017, Professor Mansmann joins the federal Medical Informatics Initiative (MII).

Venue: Large Lecture Hall (2nd floor, Room 047, University Hospital)

Abstract

Digital technologies are changing the field of medicine and health. Ubiquitous medical devices can be used as point-of-care tools to measure and timely deliver personalized medical treatments across the whole continuum of care. However, this comes with a number of technical and medical challenges that will guide the research and development of digital health technologies in the coming years. In this talk I will highlight how medical automation and artificial intelligence can open new avenues to enable easy access to medical technology outside specialized clinical centers, with an example in sleep research. Intelligent user-machine interaction,  automation, and machine learning approaches will have a huge impact on future medical technologies and will find applications in many medical domains, from prevention to treatment.

Speaker: Prof. Dr. Walter Karlen (Institute for Biomedical Engineering, University of Ulm)

Short biography

Prof. Walter Karlen is professor for Biomedical Engineering at Ulm University since May 2021 where he specializes in the research on design and algorithms for medical wearables and their applications.

He was a Swiss National Science Foundation professor at the Eidgenössische Technische Hochschule Zürich (ETH Zürich) from 2014 to 2020 where he founded and directed the Mobile Health Systems Lab. Between 2005 and 2014, he held research positions at the University of Stellenbosch, South Africa, BC Children's Hospital and Child and Family Research Institute (CFRI), Vancouver, Canada; the University of British Columbia (UBC) in Vancouver, Canada;  and Ecole Polytechnique Fédérale de Lausanne (EPFL), Switzerland. Walter Karlen holds a Master degree in micro-engineering from EPFL and a Docteur ès sciences (PhD) in Computer, Communication and Information Sciences (also EPFL).

Venue: Lecture Hall N2045 (Faculty of Applied Computer Science)

Abstract

The research project ‚GenoPerspektiv‘ of the UMG (University Medicine Göttingen) focuses on the handling of genomic high-throughput data, taking into account perspectives from the fields of clinical practice, ethics, law, and biomedical information technology. The project aims to address the various challenges and opportunities associated with the use of genomic data in healthcare. Within the realm of medical informatics, Professor Sax is specifically engaged in a subproject that delves into the application of information and health technology in medicine. This entails studying the integration of digital systems and tools to enhance healthcare delivery, data management, and decision support. Therefore, he conducts research on data privacy and security in telematics, with a particular focus on healthcare settings, where he explores the implementation and utilization of electronic patient records and personal health records in translational research. His work aims to optimize the functionality and interoperability of these digital platforms, facilitating seamless information exchange and enhancing patient care.

Another area of interest for Professor Sax is the handling of genomic data in both research and healthcare, within the broader field of biomedical informatics. He examines strategies for securely storing, analyzing, and utilizing genomic information to advance precision medicine and improve patient outcomes. Furthermore, he investigates innovative approaches for the analysis, management, and presentation of complex data in translational research. This research focus aims to develop efficient methodologies to extract meaningful insights from large datasets and effectively communicate research findings, ultimately facilitating evidence-based decision-making and the translation of research into clinical practice.

Speaker: Prof. Dr. Ulrich Sax (Institute for Medical Informatics, University Medicine Göttingen)

Short biography

Venue: Large Lecture Hall (2nd floor, Room 047, University Hospital)

Abstract

As doctors in Critical Care Medicine in the Age of Digitalization, we find ourselves confronted with a vastly increasing amount of patient data available. Processing all these data quickly to find the relevant information bits under time pressure becomes more and more demanding. Computerized Clinical Decision Support Systems can potentially alleviate these challenges by assisting with targeted presentation of patient information and clinical knowledge. In this talk, we discuss potential benefits and risks that might come with computerized clinical decision support and why such systems are still not widespread in clinical care. We also explore current research that provides some insight on how we might be able to develop systems that ultimately lead to better clinical care.

Speaker: Prof. Dr. med. Falk von Dincklage (Clinic for anesthesia, intensive care, emergency and pain medicine, University Medicine Greifswald)

Short biography

Prof. Falk von Dincklage is deputy director of the Department of Anesthesia, Intensive Care, Emergency and Pain Medicine and head of the research group “Medical Informatics” at University Medicine Greifswald. His research in medical informatics is focused on all aspects around clinical decision support, including (1) data standardization and interoperability, (2) development of machine learning based prediction models for clinical events, (3) development of knowledge-based and machine learning based decision support systems and (4) usability of clinical decision support systems.

Venue: Lecture Hall N2045 (Faculty of Applied Computer Science)

Abstract

In the last two decades evidence has gradually accumulated suggesting that the eye may be a unique window for cardiovascular risk stratification based on the assessment of subclinical damage of retinal microvascular structure and function. This can be facilitated by non-invasive analysis of static retinal vessel diameters and dynamic recording of flicker light-induced and endothelial function-related dilation of both retinal arterioles and venules. Recent new findings have made retinal microvascular biomarkers strong candidates for clinical implementation as reliable risk predictors. Beyond a review of the current evidence and state of research, the article aims to discuss the methodological benefits and pitfalls and to identify research gaps and future directions. Above all, the potential use for screening and treatment monitoring of cardiovascular disease risk are highlighted. The article provides fundamental comprehension of retinal vessel imaging by explaining anatomical and physiological essentials of the retinal microcirculation leading to a detailed description of the methodological approach. This allows for better understanding of the underlying retinal microvascular pathology associated with the prevalence and development of cardiovascular disease. A body of new evidence is presented on the clinical validity and predictive value of retinal vessel diameters and function for incidence cardiovascular disease and outcome. Findings in children indicate the potential for utility in childhood cardiovascular disease prevention, and the efficacy of exercise interventions highlight the treatment sensitivity of retinal microvascular biomarkers. Finally, coming from the availability of normative data, solutions for diagnostic challenges are discussed and conceptual steps towards clinical implementation are put into perspective.

Speaker: Prof. Dr. Henner Hanssen (Department of Sport, Exercise and Health, University of Basel)

Kurzbiographie

Prof. Henner Hanssen is a board member and associated professor of the Department of Preventive Sports Medicine, Sport, Exercise and Health (DSBG) at the University of Basel. In 2013, he obtained his habilitation on the topic "Endurance exercise and the cardiovasculature: the retinal microcirculation as a vascular biomarker". His latest clinical and scientific qualifications include: Hypertension Specialist (2018) certified by the German Hypertension League (DHL) and Fellow of the European Society of Cardiology (FESC) since 2014.

His research projects and interests revolve around the effects of exercise on chronic eye disease, exercise and vascular health in children and individuals with chronic cardiometabolic and inflammatory diseases, exercise and vascular aging in the older population, retinal vessel imaging as a microvascular biomarker of cardiovascular risk, and exercise in the prevention and treatment of hypertension.

Venue: Small Lecture Hall (2nd floor, Room 048, University Hospital)

Abstract

Speaker: Prof. Dr. med. Peter Krawitz (Institute for Genomic Statistics and Bioinformatics, University Hospital Bonn)

Short biography

Peter Krawitz studied Medicine and Physics in Munich. He continued his specialization in Medical Genetics at Charité Berlin and did a postdoc in Bioinformatics. In 2017 he was appointed a full professor at university Bonn and established the Institute for Genomic Statistics and Bioinformatics.

Venue: Lecture Hall N2045 (Faculty of Applied Computer Science)

Abstract

Leiden University Medical Center (LUMC) is dedicated to getting AI from code to clinic. To achieve this goal, the Clinical AI Implementation and Research Lab (CAIRELab) was launched almost four years ago. During this talk, Marieke van Buchem will discuss CAIRELab’s journey from code to clinic at LUMC, using real-world examples from the past few years to illustrate the process. She will address the essential conditions for AI implementation, highlighting how they are often lacking in hospital settings and the strategies employed by CAIRELab to overcome these challenges.

Speaker: Marieke van Buchem (Leiden University Medical Center)

Short biography

Marieke van Buchem is an innovation manager at CAIRELab LUMC, where she is dedicated to identifying opportunities, launching new AI projects, and cultivating partnerships with external organizations. Furthermore, she is finishing her PhD in natural language processing (NLP) applications in healthcare. With a background in medicine and medical informatics, her work focuses on developing, validating, and implementing NLP models, both at LUMC and Stanford University.

Venue: Lecture Hall N2045 (Faculty of Applied Computer Science)

Abstract

The molecular tumor board brings together experts from various disciplines such as oncology, pathology, genetics, and bioinformatics. Together, they analyze the molecular characteristics of a tumor, captured through advanced technologies like next-generation sequencing. By identifying specific gene mutations, chromosomal aberrations, and other molecular alterations, physicians gain valuable insights into the biological properties of the tumor and develop personalized treatment recommendations for cancer patients. Thus, a crucial process for MTB decision-making is the analysis, compilation, and presentation of high-dimensional sequencing data, which are used for both preparation of and case presentation to all stakeholders. Processes and tools in use will be illustrated in the context of the molecular tumor board Freiburg.

Speaker: Prof. Dr. Dr. Melanie Börries (Institute for Medical Bioinformatics and Systems Medicine, University Hospital Freiburg)

Short biography

Prof. Dr. Melanie Börries pursued an MD/PhD program as a scholarship recipient from 2001 to 2005 at the University of Basel and at the Medical University of Lübeck. She worked as a postdoctoral researcher at the Institute of Experimental and Clinical Pharmacology and Toxicology (University of Freiburg) from 2005 to 2009 and was project leader at FRIAS from 2009 to 2012. Since 2013, she is group leader at the German Consortium for Translational Cancer Research (DKTK). Since 2019, she is W3 Professor of Medical Bioinformatics and leads the Institute for Medical Bioinformatics and Systems Medicine at the University Medical Center Freiburg. Prof. Dr. Börries' research focuses on several key areas. She specializes in analyzing big data and multi-omics data to identify disease biomarkers and to explore potential therapeutic options using machine learning techniques. She has developed high-throughput data analysis pipelines in both clinical research and application. Prof. Dr. Börries also conducts analysis of sequencing data for molecular tumor boards and actively contributes her expertise to the MIRACUM consortium. Furthermore, she explores innovative strategies for innovative health care to benefit patients.