AI for Life
At Idiap, we aim to gain a deeper understanding of complex diseases and develop novel therapeutic approaches by leveraging advanced learning and inference paradigms that allow us to integrate diverse, longitudinal, interventional data and prior scientific and expert knowledge.
We embrace the complexity of individuals’ health and personal care and a broad range of multimodal health and biological data.
We are dedicated to transparently and intelligibly enhancing health outcomes while ensuring democratic access to solutions.
Expertise domains
#Bioinformatics&HealthInformatics
#DataScience&SocialComputing
#HumanComputerInteraction
#Imaging&ComputerVision
#MachineLearning
#NaturalLanguageProcessing
#Robotics&AutonomousSystems
#Security&Privacy
#SignalProcessing
#Speech&AudioProcessing
This program contributes to the following UN SDG
People
There was an error while rendering this tile
Publication highlights
Deep Learning Architectures for Estimating Breathing Signal and Respiratory Parameters from Speech Recordings. S. Nallanthighal, Z. Mostaani, A. Härma, H. Strik, and M. Magimai-Doss. Neural Networks, 2021.
This work is a result of a collaboration between Idiap and Philips Research Eindhoven on deep learning techniques for sensing breathing signal and breathing parameters from speech for healthcare. This work has potential implications for the development of novel speech-based methods for healthcare, such as breathing monitoring in telehealth and assessment for early recognition of abnormal breathing syndromes.
Spatially-Variant CNN-Based Point Spread Function Estimation for Blind Deconvolution and Depth Estimation in Optical Microscopy. A. Shajkofci and M. Liebling. IEEE Trans. on Image Processing, 29, 5848–5861, 2020.
Optical microscopy is a central tool for biomedical research and diagnostics, which calls for instruments performance to be continuously improved. We leverage a learning approach to locally determine image distortion parameters in the form of a parametric point spread function without requiring instrument- or object-specific calibration. This approach is robust to photon noise and is a key element for downstream applications, such as spatially variant deconvolution, depth localization, and flow estimation. This work is an outcome of the SNSF project Computational biomicroscopy: advanced image processing methods to quantify live biological systems (2018-2022).
Image-based Deep Learning Reveals the Responses of Human Motor Neurons to Stress and VCP-related ALS. Verzat, J. Harley, R. Patani, R. Luisier. Neuropathology and Applied Neurobiology, 48 (2): e12770, February 2022.
Amyotrophic lateral sclerosis (ALS) is a rapidly progressive and incurable neurodegenerative disease, for which the early cellular and molecular events remain poorly understood. Morphological attributes of cells and their substructures have been relatively understudied in ALS research. Transfer learning was used to leverage the power of imaging fluorescent data to enable unbiased, robust and efficient testing of biological hypotheses, and to resolve the extent of aberrant cellular morphological indices during earlier phases of ALS pathogenesis. This work, performed in collaboration with the Francis Crick Institute, is transformational to the information that can be gleaned from image analysis.
Project highlights
AI4Autism, 2021-2025, SNSF, Sinergia project, Odobez
ChaSpeePro, 2021-2025, SNSF Sinergia, Kodrasi
NeuroCIRT, 2022-2026, SNSF, Luisier
Full list of related projects
AI4Autism, 2021-2025, SNSF, Sinergia project, Odobez
Design automatic children behavior and interaction analysis digital tools for the screening and automated profiling of autism phenotype
ePartner4ALL, 2021-2024, Eureka NL-CH call, Odobez
A (personalized and) blended care solution with virtual buddy for child health
BipedAI, 2023-2024, Innosuisse, Odobez
Develop perception tools for helping blind people navigate outdoors
WeNet, 2019-2023, H2020, Gatica-Perez
Smartphone sensing to support well-being of university students
AI4Media, 2020-2024, H2020, Gatica-Perez
AI-based analysis of European health news
SECure, 2022-2024, The Ark foundation, Anjos, Freitas
Safe & Explainable Clinical AI for Orthopaedic Surgical Assessment with Med4CAST
CAD4IED, 2022-2023, Idiap & Luzerner Kantonsspital, Anjos
Computer assisted detection and grading of inflammatory eye diseases via fluorescein angiograms
ABRoad, 2022-2024, The Ark, Freitas
AI-augmented scientific discovery of new antibiotics
LunarBase, 2023-2024, The Ark, Freitas
LunarBase: industrial-scale inference
NCCR Evolving Language, 2020-2024, SNSF, Garner
Low-level mechanisms of language evolution
NCCR Evolving Language, 2020-2024, SNSF, Magimai Doss
Discovering the parallel between human and animal communications, as part of biological discovery
TIPS, 2019-2023, SNSF, Magimai Doss
Exploring relation between speech activity, breathing activity and cardiac activity
EMIL, 2021-2025, Bridge Discovery, Magimai Doss
Prediction and monitoring of Parkinson’s Disease symptoms
SWITCH, 2021-2024, SNSF, Calinon
Learning by switching roles in physical human-robot collaboration for physical assistance for sitting/standing
INTELLIMAN, 2022-2026, Horizon Europe, Calinon
AI-powered manipulation system for advanced robotic service, manufacturing and prosthetics for prosthetic hands
Melas, 2023-2025, Idiap-NIBR, Luisier
Towards the development of predictive biomarkers for patient stratification and immunotherapy response in cancer
NeuroCIRT, 2022-2026, SNSF, Luisier
Investigating the role of cytoplasmic intronic sequences in ALS pathogenesis
MelAS, 2022-2023, Novartis Foundation for Medical-Biological Research, Luisier
Alternative splicing and polyadenylation from single-cell RNA sequencing towards tumor subpopulation identification in melanoma
ChaSpeePro, 2021-2025, SNSF Sinergia, Kodrasi
Characterizing speech disorders and processes