Multi-source ultrasound reveals current limits of anomaly detection
A Multi-Source Ultrasound Benchmark Revealing the Limits of Contemporary Self-Supervised Anomaly Detection Methods
Computer Vision and Pattern RecognitionMachine Learning
Summary
Detecting medical problems in ultrasound images without labeled examples is promising but hard. The authors created a large dataset called SADUSI that combines many types of ultrasound images to test how well computers find unusual areas. They found existing computer methods struggle to separate normal from abnormal features across different body parts and imaging styles. This shows detecting anomalies in diverse ultrasound images remains an open challenge that needs better methods.
What this means in practice
- •For medical imaging developers: Evaluate and improve anomaly detection algorithms using diverse ultrasound data spanning multiple anatomies and acquisition types.
- •For machine learning engineers: Train and benchmark self-supervised models for detecting abnormalities in complex ultrasound images beyond anatomy-specific training.
Authors
Marco Riedenauer, Daniel Kienzle, Pratik Mayekar, Rainer Lienhart
Abstract
Self-supervised anomaly detection is a promising paradigm for medical ultrasound, as normal images are often easier to obtain than exhaustive annotations of all possible pathologies. However, most existing evaluations are limited to a single anatomy or task, making it unclear whether models learn a robust notion of normal ultrasound appearance or only a source-specific representation. We introduce the SADUSI benchmark, a multi-source ultrasound dataset designed to train and evaluate anomaly detection methods across a broad range of anatomical regions, views, and acquisition protocols. The goal of SADUSI is to provide a diverse normal ultrasound distribution and a benchmark for visible structural anomalies that can be assessed from single images. We evaluate representative self-supervised anomaly detection methods and find that current approaches struggle in this setting. In particular, reconstruction-based diffusion methods such as AnoDDPM and DeCo-Diff achieve pixel-level AUROC values of 0.56-0.72 and maximum F1 scores of 0.10-0.26, indicating limited separation of pathology from normal image regions. Feature-based PatchCore variants perform better, reaching pixel-level AUROC values of 0.76-0.83, but remain limited with maximum F1 scores of 0.14-0.40. These findings suggest that broad multi-source ultrasound anomaly detection remains an open challenge and that SADUSI can serve as a resource for developing methods that generalize beyond anatomy-specific settings.