Genomics and Digital Health

Medical Information Sciences

Data, algorithms and knowledge for human’s health. Digitalization is a major change in our society and applies to the whole life science ecosystem. The “Medical Information Science” group is working on the phenotypic side. Working with very heterogeneous sources of multimodal data - personal health records, such as sensors, captors and activities, patients-related data, such as computerized patient records, behavior and lifestyle - environment, exposition factors - the living ecosystems - regulatory frameworks - knowledge sources in order to build actionable data pipelines that can be used to connect with *omics. The activities of the group focus on semantics, data interpretability tools such as natural language processing. People work with symbolic, rule-based and probabilistic instrument

Visit our website

Artificial intelligence in multimodality medical imaging

The focus of our research lies in the application of artificial intelligence in multimodality medical imaging. Our group has assumed a leading role in Switzerland and became internationally recognized for excellence in medical imaging research with multimodality imaging being a focus for its activities. The group gained international recognition for contributions to cutting-edge interdisciplinary biomedical research and clinical diagnosis including the development and validation of new image correction and reconstruction techniques to realize the full potential of quantitative imaging in preclinical and clinical hybrid imaging (PET/CT and PET/MRI) as well as the development of computational modelling and radiation dosimetry algorithms.

Visit our website

Social and lifecourse epidemiology

Pre Silvia Stringhini is Head of the Population Epidemiology Unit of the Department of Primary Care Medicine at the Geneva University Hospital as well as Assistant Professor at Department of Community Health and Medicine of the University of Geneva. She holds a Master's Degree in Global Health and a PhD in Public Health and Epidemiology. Her main areas of research are social inequalities in chronic diseases and aging, the role of health behaviors in the genesis of social inequalities in health, the biological consequences of social inequalities, the role of environmental factors in social inequalities in health. During the COVID-19 crises, the activities of the Unit have been mostly devoted to the population surveillance of SARS-COV-2, and of the health consequences of the pandemic.

Visit our website

Health Informatics for Innovation, Integration, Implementation and Impact

At the core of the development of digital health, biomedical informatics plays a key role in the implementation, integration and evaluation of innovative digital methods and tools. The HI5lab (Health Informatics for Innovation, Integration, Implementation and Impact) aims at connecting these multiple dimensions with the ultimate ambition to demonstrate the impact of eHealth on the health of individuals and populations. The HI5lab is located at Campus Biotech, connected to the various expertise domains such as global health, medical information science, citizen cyberscience, bioinformatics, affective and cognitive sciences. It is also connected to global actors such as WHO, ITU, various UN agencies and NGOs from the International Geneva.

Visit our website

Computational pathology

The Computational Pathology Laboratory within the division of clinical pathology is dedicated to advancing the diagnostic and research capabilities of pathology through cutting-edge image analysis algorithms and data integration techniques. Our laboratory combines expertise in pathology, computer science, and data analytics to provide innovative solutions for both clinical applications and research projects by developing and implementing machine learning and AI algorithms to analyze digital pathology images, supporting diagnostic tasks, and conducting research to discover novel biomarkers and predictive models by integrating pathology data with other data modalities.

Visit our website

Multimodal & explainable machine learning

Our group works on approaches that combine machine learning on medical image data with text analysis approaches to build clinical decision support systems in radiology, histopathology and ophthalmology. A particularly focus is put on the interpretability of the machine learning models.

Visit our website

Epigenetic Control of Developmental Processes

Our aim is to understand how epigenetic information, in particular the 3D organization of human and mouse genomes instruct the development of organs and structures during embryogenesis. To map and functionally disrupt these chromatin states, we employ a variety of state-of-the-art technologies in pluripotent stem cells and in vivo during embryogenesis. The understanding of these basic concepts is critical to unravel the molecular mechanisms that underlie pathological gene misregulation in congenital malformations and cancer as well the mechanism that enable evolutionary novelties and diversity of life.

Visit our website

Augmented Magnetic Resonance Imaging

Magnetic resonance imaging (MRI) is currently one of the most powerful non-invasive tools to detect and diagnose diseases as well to monitor effect of treatment. Due to the absence of a significant hazard for the human body and to the progress of the technology, the number of images per MRI exam has grown exponentially in the last year. MRI has moved from a purely morphological assessment toward a functional evaluation. This evolution presents a significant challenge related to the large information to process. Located in the Radiology division at the interface of the physics, IT and medicine, our group is specialized in the improvement of the MRI workflow from the image acquisition to the image analysis and clinical validation with a special focus on cardiac and renal diseases.

Visit our website

Pharmaco-omics and precision medicine

Our research group's work focuses on investigating genetic determinants of variability in therapeutic response (pharmacogenomics). We study gene-environment-pathology interactions, especially at the pharmacokinetic level, including drug metabolism and transport. Phenoconversion, the discordance between predicted and actual phenotypes, is assessed using the Geneva cocktail. We evaluate genetic vulnerability to drug interactions and develop in vitro, in vivo, and in silico tools to measure and predict drug response. A key focus is optimizing PBPK modeling for personalized prescriptions and progressively integrating phenotypic and pharmacodynamic measures, with future use of machine learning algorithms.

Visit our website

Data Science for Digital Health

The Data Science for Digital Health (DS4DH) group aims to design and implement novel algorithms and computational methods for the management and analysis of complex and large-scale datasets in the health and life sciences domain in order to foster innovative digital health solutions. The group is particularly interested in developing research around machine learning and natural language processing models that can blend and exploit (semantically) rich and (often) non-Euclidean datasets to create actionable insights. DS4DH is located at Campus Biotech and works closely with other digital health players, such as SIB, HUG and the SIMED, BiTeM and HI5lab groups. It is also well connected with private actors working at the forefront of innovation in the health and life sciences sectors.

GIRAPH (geographic information for research and analysis in public health)

We are a diverse group of scientists of different experience and backgrounds, with a shared focus on understanding how spatial, environmental and health data can be combined to provide insights into disease mechanisms and etiologies.

Visit our website

Genomics and genetics in evolutionary perspective

Genomics and genetics is transforming with advances of sequencing technologies. It also transforms the healthcare, from assessing human genetics, to tracking microbial outbreak or antibiotic resistance, and associating microbiomes with diseases. These big data challenge our computational techniques as well as our evolutionary models to interpret the genetic diversity.

Visit our website

Precision oncology

Incorporating clinical biomarkers into oncology treatment strategies allows healthcare professionals to select the most appropriate therapies for individual patients, taking into account their unique genetic and molecular profiles, and thus significantly improving treatment outcomes and reducing the likelihood of unnecessary or ineffective interventions. In our group, we leverage various data sources including pathology images, radiology, clinical and molecular data, to develop actionable biomarkers for oncological treatment. Our goal is to utilize large scale multimodal data and provide machine learning driven approaches that can better guide clinical decisions and benefit oncology patients.