
SM
Salman Mohammad, Furkan Kas{\i}m, Manuel G\"unther
· 1 min read
ResearcharXiv cs.CV
DAL: Dynamic Angular Loss for Imbalanced Medical Image Classification
arXiv:2611.24597v1 Announce Type: new
Abstract: Class imbalance remains a major obstacle in medical image classification, where rare but clinically important classes are often overshadowed by majority categories. Although current strategies for reducing bias, including angle-based softmax (ABS) losses, offer strong discriminative features, their behavior in settings with a small number of severely imbalanced classes remains insufficiently understood. Previous approaches that estimate the intrinsic parameters of ABS rely on the assumption of many classes, which does not hold in medical settings. We derive data-driven upper and lower bounds for the scale parameter and an upper bound for the angular margin using negative-angle statistics, and use these to drive a batch-adaptive angular margin (BAAM), which we combine with dynamically-weighted cross entropy (DWCE) to define our dynamic angular loss (DAL). Across four highly imbalanced medical image classification benchmarks spanning dermatology, ophthalmology, and gastrointestinal endoscopy, our method consistently achieves superior balanced accuracy and the lowest coefficient of variation of per-class recall. Although this entails a small performance reduction on majority classes, the gains on minority categories are substantial, yielding a reliable and equitable overall classifier. Our source code is available at: https://github.com/msalman95/DAL.
Original source
This story was published by arXiv cs.CV and written by Salman Mohammad, Furkan Kas{\i}m, Manuel G\"unther. SyncAI.news shows a preview; the complete article is on the publisher's site.
Read the full story on arxiv.org


