Learning Large Margin Sparse Embeddings for Open Set Medical Diagnosis

TL;DR

This paper introduces a novel deep learning approach for open set recognition in medical diagnosis, effectively identifying known and unknown diseases by leveraging large margin sparse embeddings and a new loss function.

Contribution

It proposes the Open Margin Cosine Loss (OMCL) combining margin loss with adaptive scaling and open-space suppression, advancing open set recognition in medical AI.

Findings

MLAS outperforms existing methods on benchmark datasets.

The approach improves accuracy, AUROC, and OSCR in recognizing unknowns.

Extensive ablation studies validate each component's effectiveness.

Abstract

Fueled by deep learning, computer-aided diagnosis achieves huge advances. However, out of controlled lab environments, algorithms could face multiple challenges. Open set recognition (OSR), as an important one, states that categories unseen in training could appear in testing. In medical fields, it could derive from incompletely collected training datasets and the constantly emerging new or rare diseases. OSR requires an algorithm to not only correctly classify known classes, but also recognize unknown classes and forward them to experts for further diagnosis. To tackle OSR, we assume that known classes could densely occupy small parts of the embedding space and the remaining sparse regions could be recognized as unknowns. Following it, we propose Open Margin Cosine Loss (OMCL) unifying two mechanisms. The former, called Margin Loss with Adaptive Scale (MLAS), introduces angular margin for reinforcing intra-class compactness and inter-class separability, together with an adaptive scaling factor to strengthen the generalization capacity. The latter, called Open-Space Suppression (OSS), opens the classifier by recognizing sparse embedding space as unknowns using proposed feature space descriptors. Besides, since medical OSR is still a nascent field, two publicly available benchmark datasets are proposed for comparison. Extensive ablation studies and feature visualization demonstrate the effectiveness of each design. Compared with state-of-the-art methods, MLAS achieves superior performances, measured by ACC, AUROC, and OSCR.