Improving Out-of-Distribution Detection in Echocardiographic View Classication through Enhancing Semantic Features

TL;DR

This paper introduces a label smoothing technique to enhance semantic features, significantly improving near-out-of-distribution detection in echocardiographic view classification, surpassing traditional Mahalanobis distance methods.

Contribution

The study proposes a novel use of label smoothing to enrich semantic features, boosting the effectiveness of OOD detection in echocardiography.

Findings

Enhanced semantic features improve near-OOD detection accuracy.

Combining label smoothing with Mahalanobis distance sets a new benchmark.

Method outperforms conventional OOD detection approaches.

Abstract

In echocardiographic view classification, accurately detecting out-of-distribution (OOD) data is essential but challenging, especially given the subtle differences between in-distribution and OOD data. While conventional OOD detection methods, such as Mahalanobis distance (MD) are effective in far-OOD scenarios with clear distinctions between distributions, they struggle to discern the less obvious variations characteristic of echocardiographic data. In this study, we introduce a novel use of label smoothing to enhance semantic feature representation in echocardiographic images, demonstrating that these enriched semantic features are key for significantly improving near-OOD instance detection. By combining label smoothing with MD-based OOD detection, we establish a new benchmark for accuracy in echocardiographic OOD detection.