CIBM Annual Symposium 2022
The CIBM Annual Symposium will gather members of the biomedical imaging community in Switzerland and beyond allowing them to exchange during an exciting and intellectually stimulating day filled with insightful talks and poster presentations.
The CIBM annual symposium will be concluded with the inauguration of the new 7T MRI scanner at Campus Biotech, Geneva.
There are two different tickets, you can choose one or two depending on your attendance:
- 9:00am – 5:00pm: Scientific programme
- 5:00pm -9:00pm : Inauguration of the 7T MRI scanner & cocktail reception
PROGRAMME
(PDF available here)9:00 CIBM overview and 2022 achievements
CIBM
9:10 fMRI in mice: what we learned and what we should know
ABSTRACT
Genetically engineered mouse models of human neurological or psychiatric pathologies have evolved as important tools for elucidating pathophysiological mechanisms, identifying diagnostic and prognostic biomarkers, and evaluating novel therapeutic strategies. Structural and functional phenotyping, based e.g. on magnetic resonance methods have raised considerable interest given their inherent translatability to the clinics. In particular, fMRI has been used to study the functional architecture of the mouse brain and its adaptation under pathological conditions. Examples include the analysis of the functional reorganization imposed by spinal cord injuries or cerebral infarction, network alterations induced by pathological conditions such as chronic pain, anxiety/stress, amyloidosis or induced by pharmacological interventions. Optogenetic tools combined with fMRI complement these approaches by providing specific information on specific neurotransmitter systems.
Yet, fMRI in mice comes with some challenges. Focal sensory stimuli have be shown to elicit widespread responses not compatible with the mouse brain’s functional topology. Anesthetics were shown to significantly modulate location and shape of fMRI responses. Pathological or pharmacological involvement of the brain vasculature/microvasculature may dominate hemodynamics-based functional readouts, and there is compelling evidence that astrocytic contributions affect the BOLD response. Hence, proper interpretation of experimental data requires controlling or at least understanding these potentially confounding contributions.
ETHZ
BIO
Markus Rudin is retired professor for molecular imaging and functional pharmacology both at University of Zürich and ETH Zürich and currently director of The LOOP Zurich, a translational research center focusing in precision medicine founded by UZH and ETHZ and the four Zurich university hospitals. He concluded his studies of chemistry at ETHZ with a PhD in physical Chemistry at ETHZ. Thereafter he joined Sandoz AG, to become Novartis AG, where he served in different functions finally to become head of the Analytical and Imaging Science Unit and deputy head of the Core Technology Area. After 21 years in industry he returned to academia in 2005.
9:40 Molecular imaging: the gateway to precision medicine in neurodegeneration
ABSTRACT
Brain aging is associated with a higher risk of neurodegenerative disorders. Their detection before symptom occurrence, in the preclinical phase, is of the utmost importance to develop preventive programs and specific treatments. Neurodegenerative conditions are characterized by pathological aggregates in the brain and molecular imaging with Positron emission tomography (PET) represents the only biomarker currently able to assess the presence and the topography of pathological deposits in the brain in vivo.
The possibility to detect pathology before clinical signs has led to a paradigm change in our understanding of the disease onset and progression, from a description mainly based on symptoms in patients, to the assessment of disease markers in cognitively normal individuals to define risk profiles.
The talk will summarize the ongoing efforts of our group in Geneva, within a network of national and international collaborations, to implement these progresses into clinical practice, for prevention, diagnosis and treatment.
HUG UNIGE
BIO
Prof. Valentina Garibotto is Division Chair of Nuclear Medicine and Molecular Imaging at the University Hospitals of Geneva, Associate Professor at the Geneva University and Head of CIBM PET HUG UNIGE Molecular Imaging Section. After training in Italy and Germany she developed her clinical, research and teaching activity in Geneva, with the support of the Swiss National Research foundation, the Velux Stiftung, the Schmidheiny and Aetas foundation.. She is Congress Chair Elected of the European Association of Nuclear Medicine and Chair of the Neuroimaging group of the Swiss Society of Nuclear Medicine.
10:00 Treating brain hypometabolism
ABSTRACT
GliaPharm is a biopharma company, focusing on the development of therapeutic solutions that target glial cells to treat neurological diseases. The role of glial cells in brain function and in neurological diseases is being increasingly recognized. Glial cells, which represent half of the cells of the brain, are fundamental for the proper function of neurons by providing metabolic, signaling and energy support. Therefore, therapeutic interventions aimed at re-establishing neuronal-glial cross talk will be fundamental to treating neurological diseases.
GliaPharm has developed know-how and a platform to characterize glial cells, to develop a pipeline of active molecules and to evaluate the potential of several therapeutic solutions. The company has a project in an orphan monogenic disease with marked brain hypometabolism, as well as other larger indications with brain hypometabolism, including Alzheimer's disease.
Gliapharm lead compound GP-119, is an orally active small molecule that penetrates the blood brain barrier. It is a positive allosteric modulator of an astrocytic enzyme. Gliapharm is currently establishing translational biomarkers, such as FDG-PET, that will guide the proof of mechanism in human clinical trials.
Gliapharm
BIO
Dr. Sylvain Lengacher is a molecular biologist with over 20 years of experience in the field of neuroscience research, biotechnology management, and technology transfer. He is co-founder of GliaPharm where he serves as co-CEO. Dr. Lengacher was a research director at UNIL, then at Ecole Polytechnique Fédérale de Lausanne for 11 years. He was also the Knowledge and Technology Transfer officer responsible for the management of Intellectual Property of the Synapsy-NCCR program of the Swiss National Science Foundation. Dr. Lengacher holds a PhD in molecular biology and graduated from EPFL in the Management of Biotech.
10:20 Break: Posters and Demos (CIBM Research Staff Scientists + PhD students + Collaborators + Exhibitors)
11:00 Making neural networks for image reconstruction more trustworthy
ABSTRACT
Deep neural networks (DNN) have a remarkable ability to improve the quality of reconstruction of biomedical images. While the results reported so far are extremely encouraging, serious reservations have been voiced pertaining to the stability of these tools and the extent to which we can trust their output. After providing an overview of the current state of the field, we describe an iterative reconstruction scheme that relies on stable Lipschitz-constrained networks with learnable activation functions. Unlike the first generation of DNN-based methods, it has a feedback mechanism that guarantees consistency. Our algorithm outperforms the state-of-the-art in compressed sensing in terms of image quality.
EPFL
BIO
Michael Unser is Full Professor at the EPFL, the Head of CIBM SP EPFL Mathematical Imaging Section, and the academic director of EPFL's Center for Imaging, Lausanne, Switzerland. His primary areas of investigation are biomedical imaging and applied functional analysis. He is internationally recognized for his research contributions to sampling theory, wavelets, the use of splines for image processing, and computational bioimaging. He has published over 400 journal papers on those topics.
Prof. Unser is a fellow of the IEEE (1999), an EURASIP fellow (2009), and a member of the Swiss Academy of Engineering Sciences. He is the recipient of several international prizes including five IEEE-SPS Best Paper Awards, two Technical Achievement Awards from the IEEE (2008 SPS and EMBS 2010), the Technical Achievement Award from EURASIP (2018), and a recent Career Achievement Award (IEEE EMBS 2020). He was awarded three ERC AdG grants: FUNSP (2011-2016), GlobalBioIm (2016-2021), and FunLearn (2021-2026).
11:20 Uncertainty estimation in medical image analysis
ABSTRACT
Associating uncertainties to predictions is one of the requirements of medical applications for integrating machine learning systems in practice. In this context, uncertainty estimation in machine learning models have attracted attention of the research community. In this talk, I would like to present the problem, describe current solutions, and discuss open issues."
ETHZ
BIO
Ender Konukoglu is an Associate Professor in the Department of Information Technology and Electrical Engineering at ETH Zurich. His research focuses on developing algorithms for interpreting visual data, and in particular medical images. Before joining ETH in 2016, he worked at Athinoula A. Martinos Center for Biomedical Imaging and Harvard Medical School as an Instructor in Radiology and Assistant in Neuroscience, Microsoft Research Cambridge as a post-doctoral fellow, and did his PhD at Inria Sophia Antipolis and Universite de Nice. He is from Istanbul, Turkey.
11:40 Modelling tissue structure
ABSTRACT
Computational Pathology, the computerised version of histopathology is a growing field of application of deep-learning, especially convolutional neural networks. A characteristic of the field is the high reliance on domain specific knowledge. Many works have looked at end-to-end methodologies and multiple-instance learning to limit the dependence to the scarce and expensive pixel-level annotations. Using classification as a pretext task, holistic tissue models provide representations of the histological structures of organs that can be used to increase the speed or the accuracy of computational pathology tasks.
Roche
BIO
Pierre Moulin MD, PhD, is physician-scientist, computational pathologist, and a teacher. He has experience in molecular localisation, gene expression profiling, and image processing for digital pathology. He has developed deep-learning models of histopathology for toxicology and is the architect and industry leader of the IMI BigPicture consortium, a collaborative effort bringing together Universities, small businesses and major pharmaceutical companies for a 6-years endeavour aiming at establishing a central repository for digital pathology and catalyse the development of AI in pathology. Beyond his role in the Early Biomarker Development Oncology at Roche, Pierre is attending pathologist at the CHUV, and lecturer at the University of Basel.
12:00 Lunch: Posters and demos (CIBM Research Staff Scientists + PhD students + Collaborators + Exhibitors)
13:30 Safer machine learning models for healthcare with out-of-distribution detection
ABSTRACT
Optical Coherence Tomography imaging has now become the most common medical imaging modality in routine clinical use. Its impact has been immense in diagnosing and monitoring chronic eye conditions that affect over half a billion people worldwide. Given this volume of patients and monitoring frequency, it is unsurprising that machine learning methods play a critical part in the global strategy to ease routine care as the burden on healthcare systems is mounting with the aging world population. Yet developing safe and resilient machine learning models remains a challenge and is at the heart of the work we present here. Specifically, we consider the problem of Out-of-Distribution detection, whereby test scans which differ from examples in the training set, must be identified to avoid machine learning models to behave unexpectedly or potentially dangerously. We study the case where such divergent cases are unknown to the machine learning model and show how a simple approach can provide impressive abilities in reducing the risk of machine learning models from making unforeseen mistakes.
University of Bern
BIO
Prof. Raphael Sznitman is the director of the ARTORG Center for Biomedical Engineering Research and the director for the Center of Artificial Intelligence in Medicine at the University of Bern. After receiving a BSc in Cognitive Science from the University of British Columbia (Canada) and a PhD in Computer Science from Johns Hopkins University (USA), he was a postdoctoral fellow at the EPFL (Switzerland) and then joined the medical faculty at the University of Bern in 2015. He has also co-founded two key startups in the domain of AI and ophthalmology (www.retinai.com and www.perivision.io, Switzerland).
14:00 Multimodal neuroimaging to better understand stroke recovery
ABSTRACT
Stroke is the main reason for long-term impairment with patients varying considerably in terms of outcome and course of recovery. Multimodal neuroimaging offers the opportunity to better understand underlying mechanisms, predict course and outcome and the degree of recovery after a stroke, especially taking into acocunt that stroke has to be considered as a network disorder. Here, I will present multimodal insights in structural and functional connectomics/disconnectomics and how they impact on the recovery process, its underlying mechanisms and the prediction of outcome.
EPFL
BIO
"Prof. Dr. med. Friedhelm Hummel is a translational systems neuroscientist and trained neurologist, with more than 15 years of experience in clinical neurology. Since September 2016, he is appointed Full-Professor at the École Polytechnique Fédérale de Lausanne (EPFL), Director of the Defitech Chair of Clinical Neuroengineering and Adjunct Professor at the University Medical Hospital Geneva. His scientific interests are in the understanding of stroke recovery, cognitive impairment and healthy aging using multimodal systems neuroscience methods and in neuro-technologies to develop innovative interventional strategies to enhance recovery, reduce impairment by means of non-invasive brain stimulation techniques and the combination of neuro-technologies.
"
14:20 Making sense of our senses
ABSTRACT
This talk provides an overview of our efforts to understand the multisensory nature of human cortices. First, a summary of how human brain imaging, mapping and stimulation have been innovated to document multisensory processes in human cortices is provided, followed by evidence of the behavioural and perceptual relevance of such processes. Next, the persistence of multisensory processes across time is demonstrated by documenting their contribution to memory functions. This is followed by demonstrations across infancy, childhood, adulthood, and ageing of how low-level multisensory processes scaffold and even predict the integrity of higher-level cognitive functions. Finally, the contributions of multisensory processes to the accessible and democratized rehabilitation of sensory loss is presented. Collectively, this work outlays a roadmap for the central role of multisensory processes across the lifespan in health and disease.
CHUV UNIL
BIO
Micah is an associate professor in the Radiology Department CHUV-UNIL, Section Head of the CIBM EEG CHUV-UNIL, and adjunct professor at Vanderbilt University. In 2021, he was named Founding Scientific and Academic Director of The Sense Innovation and Research Center; a joint venture of CHUV-UNIL and HES-SO Valais. He earned a double bachelor’s degree from The Johns Hopkins University and received his PhD in neuroscience with honours from the Albert Einstein College of Medicine. Among others, he received awards from the Leenaards Foundation, Swiss Brain League, and Swiss Society for Biological Psychiatry. Micah’s work has characterised new schemas of brain functional organization, spurring new diagnostic and intervention strategies for cognitive dysfunction using sensory processes as an access point.
14:40 Multimodal brain imaging to probe and restore cognitive functions
ABSTRACT
Neuroimaging techniques such as functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) allow not only measuring brain activity with increasing precision but now also manipulating it through real-time neurofeedback protocols. In the latter conditions, healthy volunteers or patients can train to regulate neural activity in one or more areas of their own brain in order to modify particular sensorimotor, cognitive, or affective functions. This talk will present a few examples of such applications from recent or ongoing work in our laboratory, to illustrate how neurofeedback may be leveraged to enhance or attenuate reactivity in specific brain regions, uncover their role within wider networks, and remediate functional abnormities in brain diseases such as stroke.
UNIGE
BIO
Patrik Vuilleumier is a neurologist who is using brain imaging techniques (such as functional resonance magnetic imaging, fMRI) to study the cerebral mechanisms of cognition, emotion, and consciousness. After training in clinical neurology and neuropsychology in Geneva, Lausanne, and Paris, he pursued his research in cognitive neurosciences at the University of California in Davis, and then at University College London. He is now heading the Laboratory for Neurology and Imaging of Cognition at Geneva University Medical Center and Hospital, and a research group at Campus Biotech Geneva. He was director of the Geneva University Neuroscience Center from 2007 to 2017, and co-founded the Brain & Behaviour Laboratory (BBL) in 2008 with the Swiss Center for Affective Sciences (CISA) where is also a steering committee member. He joined the CIBM in 2022 to head the Cognitive and Affective Neuroimaging Section. His past and current research focuses on neural mechanisms of perception and memory and their regulation by attentional control and emotion processes, in the healthy brain as well as in neurological and psychiatric diseases.
15:00 Break: Posters and demos (CIBM Research Staff Scientists + PhD students + Collaborators + Exhibitors)
15:40 Chronic kidney disease & the CIBM
ABSTRACT
A new approach to measure renal diffusion using MRI based on the Readout Segmentation of Long Variable Echo Train (RESOLVE) MR sequence and the cortico-medullary apparent coefficient diffusion difference ΔADC has been developed at the CIBM. ΔADC from this RESOLVE sequence was highly correlated to fibrosis observed on the biopsy specimens of 29 homogenous kidney allograft patients. This result was further validated in a monocentric cross-sectional study including164 patients. More recently, our group demonstrated that the evolution of ΔADC outperformed the eGFR to assess the evolution of IF and that ΔADC was a prognostic marker of renal function decline independent of baseline e-GFR and proteinuria. The story of this research as well as its clinical implication will be presented.
HUG UNIGE
BIO
Jean-Paul Vallée is a certified radiologist specialised in cardiovascular and nephro-urological radiology with a PhD thesis in Physics. He is currently Full Professor at the Geneva Faculty of Medicine and since 2015, head of the cardiovascular radiological unit in the Diagnostic department of the Geneva University Hospital. His main research interest is focused around clinical validation and methodological development using MRI with clinical translation of numerous experimental models in the cardiovascular and urogenital fields.
16:00 Ketone in brain, heart and kidney metabolism – application of MRS and PET imaging
ABSTRACT
There is growing interest in the metabolism of ketones owing to their reported benefits in neurological and more recently in cardiovascular and renal diseases. As an alternative to a very high fat ketogenic diet, ketones precursors for oral intake are being developed to achieve ketosis without the need for dietary carbohydrate restriction, Method to detected non-invasively ketones in human allow for measurement of their metabolic effect, and include 1H-MRS and 11C-AcAc-PET. To illustrate the effect of ketones in human, the results of recent clinical studies will be presented.
Nestlé
BIO
Bernard Cuenoud is the Head of Global Research and Clinical Development at Nestle Health Science in Lausanne, focusing on developing innovative products in medical nutrition and rare diseases. He is also an Adjunct Professor at the Faculty of Medicine and Health Sciences - Sherbrooke University. Bernard Cuenoud steered the discovery and development of many drugs and nutrition products. He obtained is PhD from Yale University and did his post-doc at Harvard Medical School.
16:20 Implanted systems to restore neurological functions after spinal cord injury
ABSTRACT
NeuroRestore combines cutting-edge basic, translational and clinical research platforms that enable synergia between discoveries, innovations and real-world applications. During this talk, I will present a few examples of these unconventional developments, including the design and validation of neuroprosthetic systems that reversed paralysis in people with spinal cord injury.
CHUV UNIL
BIO
Jocelyne Bloch is a Professor of Neurosurgery at the University Hospital Lausanne (CHUV) where she leads the functional neurosurgery unit. Together with Grégoire Courtine, Professor of neuroscience at the Swiss Federal Institute of Technology Lausanne (EPFL), they cofounded .NeuroRestore, a center dedicated to the development of neurotherapies involving neurosurgical interventions. They are known for the conception of neuroprosthetic implants that restored walking in people with chronic paralysis. They also cofounded ONWARD medical; a start-up that aim to translate their discoveries into medical products.
16:40 Thinking inside the voxel, a new window into myelin water imaging
ABSTRACT
This lecture will cover some of the recent advances on myelin water imaging that have been enabled by thinking of our voxels as a mixture of compartmentalized MR signals.
Using realistic biophysical modelling of the interaction between this compartments with the strong magnetic field used in MRI allows the extraction of microstructural information such as Myelin Water Fraction, g-ratio as wells as various compartment specific relaxation times.
Radboud University
BIO
José P. Marques is an Associate PI at the Donders Centre for Cognitive Neuroimaging of the Radboud University, where he leads a group on Quantitative Structural Imaging.
He is a Portuguese physicist that got his PhD at the University of Nottingham, and got into quantitative imaging while wondering how to get decent MR images while working with one of the first human 7Ts at the EPFL.
Having worked both on the more fundamental/basic physics and clinical ends of MRI, has strengthened his conviction that the most relevant contribution a physicist can make to neuroscience is not just to offer good quality images but, more importantly, create quantitative and understandable information that relates to tissue properties.
17:00 Inauguration of the new 7T MRI scanner at Campus Biotech, Geneva
17:15 Genesis of the project “Ambition Neurosciences”
Fondation Campus Biotech Genève
17:30 The importance of diagnostics in tomorrow’s medicine
University Hospital of Geneva
18:00 Cocktail reception and networking
Date
- 30 Nov 2022
- Expired!
Time
- 9:00 am - 9:00 pm
More Info
Location
- Campus Biotech
- Geneva