TL;DR

Aim and Scope

Research in Fairness, Equity, and Accountability in Machine Learning has highlighted biases in learned systems between demographic groups, potentially adding risk to outcomes for specific sub-populations. Recent work has uncovered these risks in both Medical Image Computing (MIC) and Computed Assisted Intervention (CAI) application domains. Our workshop aims to host a vibrant discussion of recent and ongoing work in Fairness/+ in Medical Imaging in a half-day fully virtual workshop.

Schedule

All times GMT

Time Speaker Title
1200 Commitee Welcome + Introduction to Fairness of AI in Medical Imaging (Recording)
1230 Karim Lekadir FUTURE-AI: Guiding Principles and Consensus Recommendations for Trustworthy Artificial Intelligence in Medical Imaging (Recording)
1300 Haoran Zhang Group Fairness in Chest X-ray Diagnosis: Helpful or Harmful? (Recording)
1330 Coffee Break/Chat
1350 Esther Puyol Fairness in Cardiac Magnetic Resonance Imaging: Assessing Sex and Racial Bias in Deep Learning-Based Segmentation (Recording)
1420 Qingyu Zhao Confounder-aware Deep Learning Models for Neuroimaging Applications (Recording)
1450 Ira Ktena Exploring fairness in the transductive setting for neuroimaging studies (Recording)
1520 Panel Discussion (Recording)

Speaker Bios

Karim Lekadir (Universitat Pompeu Fabra)

Karim Lekadir Dr. Karim Lekadir is a Ramon y Cajal senior researcher at the BCN-MedTech Centre of the Universitat Pompeu Fabra, Barcelona. He received a PhD in Computing from Imperial College London and was a postdoctoral researcher at Stanford University. His current research focuses on the development of data science and machine learning approaches for the analysis of large-scale biomedical data. He is the Project Coordinator of the recently funded euCanSHare H2020 project (2018-2022), leading a consortium of 16 institutions to address data sharing and big data approaches in cardiovascular personalised medicine.

Haoran Zhang (CSAIL, MIT)

Haoran Zhang Haoran Zhang is a Ph.D. student at the Computer Science and Artificial Intelligence Laboratory at MIT, advised by Prof. Marzyeh Ghassemi. His research focuses on methods to construct robust and fair machine learning models which maintain their performance across real-world distribution shifts. He is also interested in the application of such models in the healthcare domain. Before joining MIT, Haoran received his M.Sc. at the University of Toronto and his B.Eng. from McMaster University.

Esther Puyol (King’s College London)

Esther Puyol Dr. Esther Puyol completed her PhD in the Biomedical Engineering department at King’s College London in 2018. From 2018 to 2022 did a postdoc at King’s College London, where she developed novel deep learning models for automated analysis of cardiac imaging and investigated the bias effect of using imbalanced medical imaging databases for training segmentation models. Currently, she is working in HeartFlow as a research scientist. 

Qingyu Zhao (Stanford University)

Qingyu Zhao Dr. Qingyu Zhao is an instructor in the Department of Psychiatry and Behavioral Sciences at Stanford University. He obtained his Ph.D. in computer science in 2017 from the University of North Carolina at Chapel Hill and was a postdoc and research scientist in the Stanford Psychiatry department. His research has been focusing on identifying biomedical phenotypes associated with neuropsychiatric disorders by statistical and machine-learning-based computational analysis of neuroimaging and neuropsychological data. Dr. Zhao is a recipient of the K99/R00 Pathway to Independence Award from the National Institute on Alcohol Abuse and Alcoholism.

Ira Ktena (Deepmind)

Ira Ktena Dr. Ira Ktena is a Senior Researcher at DeepMind working on Deep Learning research for Life Sciences. Previously, she was a Machine Learning Researcher at Twitter, where she carried out research at the intersection of recommender systems and algorithmic transparency. Her exploration on algorithmic amplification of political content on Twitter published at PNAS was featured by the Economist and the BBC, among others. She completed a PhD in Medical Image Computing at Imperial College London with Professor Daniel Rueckert and her research focused on developing methods for modelling and analysing graph-structured neuroimaging data at an individual or population level using traditional graph theoretical approaches and geometric deep learning. During her PhD, Ira visited the Stroke Group in Massachusetts General Hospital, Harvard Medical School where she worked with Professor Natalia Rost supported by an EMBO Short-Term Fellowship.

Organizers

Aasa Feragen, DTU Compute, Technical University of Denmark
Andrew King, King’s College London
Ben Glocker, Imperial College London
Daniel Moyer, Vanderbilt University
Enzo Ferrante, CONICET, Universidad Nacional del Litoral
Eike Petersen, DTU Compute, Technical University of Denmark
Esther Puyol, King’s College London
Melanie Ganz-Benjaminsen, Dept. of Computer Science, University of Copenhagen & Neurobiology Research Unit, Rigshospitalet
Veronika Cheplygina, IT University Copenhagen

Contacts

Please contact Daniel (dmoyer@csail.mit.edu) with any questions, or reply in our website’s github issues which may serve as a forum.