We actively develop new methodology and explore learning theory relevant to applications in medical imaging. Areas of research include causality, domain shift, robustness & reliability, bias & fairness, generative modelling, semi-supervised & self-supervised learning, deep representation learning, and geometric deep learning.
Causal Machine Learning
Mitigating attribute amplification in counterfactual image generation, International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2024
High Fidelity Image Counterfactuals with Probabilistic Causal Models, International Conference on Machine Learning (ICML), 2023
Measuring axiomatic soundness of counterfactual image models, International Conference on Learning Representations (ICLR), 2023
Deep Structural Causal Models for Tractable Counterfactual Inference, Advances in Neural Information Processing Systems (NeurIPS), 2020
Causality matters in medical imaging, Nature Communications, 2020
Robustness & Reliability
Counterfactual contrastive learning: robust representations via causal image synthesis, MICCAI Workshop on Data Engineering in Medical Imaging (DEMI), 2024
A geometric approach to robust medical image segmentation, Medical Image Analysis, 2024
Automatic correction of performance drift under acquisition shift in medical image classification, Nature Communications, 2023
Robustness Stress Testing in Medical Image Classification, MICCAI Workshop on Uncertainty for Safe Utilization of Machine Learning in Medical Imaging (UNSURE), 2023
Failure Detection in Medical Image Classification: A Reality Check and Benchmarking Testbed, Transactions on Machine Learning Research, 2022
Analyzing Overfitting under Class Imbalance in Neural Networks for Image Segmentation, IEEE Transactions on Medical Imaging, 2021
Representation Learning
Identifiable Object-Centric Representation Learning via Probabilistic Slot Attention, Advances in Neural Information Processing Systems (NeurIPS), 2024
Grounded Object-Centric Learning, International Conference on Learning Representations (ICLR), 2024
Domain Generalization via Model-Agnostic Learning of Semantic Features, Advances in Neural Information Processing Systems (NeurIPS), 2019
Semi-supervised learning via compact latent space clustering, International Conference on Machine Learning (ICML), 2018
Fairness & Bias
A causal perspective on dataset bias in machine learning for medical imaging, Nature Machine Intelligence, 2024
Risk of Bias in Chest Radiography Deep Learning Foundation Models, Radiology: Artificial Intelligence, 2023
The Role of Subgroup Separability in Group-Fair Medical Image Classification, International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2023
Algorithmic encoding of protected characteristics in chest X-ray disease detection models, eBioMedicine, 2023
Potential sources of dataset bias complicate investigation of underdiagnosis by machine learning algorithms, Nature Medicine, 2022
Machine Learning on Graphs
The Importance of Model Inspection for Better Understanding Performance Characteristics of Graph Neural Networks, IEEE International Symposium on Biomedical Imaging (ISBI), 2024
A Comparative Study of Graph Neural Networks for Shape Classification in Neuroimaging, Proceedings of Machine Learning Research, GeoMedIA Workshop, 2022
Graph Convolutional Gaussian Processes, International Conference on Machine Learning (ICML), 2019
Disease Prediction using Graph Convolutional Networks: Application to Autism Spectrum Disorder and Alzheimer’s Disease, Medical Image Analysis, 2018
Metric learning with spectral graph convolutions on brain connectivity networks, NeuroImage, 2018
