Learning Disease State from Noisy Ordinal Disease Progression Labels
This work addresses a key challenge in medical imaging for nAMD diagnosis, but it is incremental as it builds on existing ordinal classification and noise-handling techniques.
The paper tackled the problem of learning disease state from noisy ordinal labels for neovascular age-related macular degeneration (nAMD), by modeling disease progression as an ordinal classification task with tailored methods like antisymmetric logit space equivariance and uncertainty-based loss re-weighting, resulting in strong few-shot performance for nAMD activity classification.
Learning from noisy ordinal labels is a key challenge in medical imaging. In this work, we ask whether ordinal disease progression labels (better, worse, or stable) can be used to learn a representation allowing to classify disease state. For neovascular age-related macular degeneration (nAMD), we cast the problem of modeling disease progression between medical visits as a classification task with ordinal ranks. To enhance generalization, we tailor our model to the problem setting by (1) independent image encoding, (2) antisymmetric logit space equivariance, and (3) ordinal scale awareness. In addition, we address label noise by learning an uncertainty estimate for loss re-weighting. Our approach learns an interpretable disease representation enabling strong few-shot performance for the related task of nAMD activity classification from single images, despite being trained only on image pairs with ordinal disease progression labels.