cs.AINov 6, 2026

Generative Active Learning for Label-Efficient Pan-Organ Ultrasound Diagnosis

Authors: Yaofei Duan, Yuhao Huang, Yuan Gao, Ruixiang Qi, Tianyu Zhang, Yanlin Geng, Yue Sun, Sio-Kei Im, 

Organizations: Macao Polytechnic University, Faculty of Applied Sciences, Macau, China · The Netherlands Cancer Institute, Department of Radiology, Amsterdam, 1066 CX, The Netherlands · Radboud University Medical Center, Department of Medical Imaging, Nijmegen, 6525 GA, The Netherlands · Shenzhen University, Medical Ultrasound Image Computing (MUSIC) Lab, Shenzhen, 518060, China · 

Abstract

Medical ultrasound (US) enables real-time, non-invasive diagnosis, yet building US-based AI systems is limited by data scarcity, costly expert annotation, and limited cross-center generalizability. Privacy constraints and instance-level reporting hinder the collection of diverse abnormal cases and frame-level labels, while repeated retraining for new centers increases computational cost. We introduce EvoDiff, a generative active learning framework that integrates diffusion modelling with information-theoretic objectives, evolutionary search, and curriculum-based sampling to select informative samples across heterogeneous US datasets. EvoDiff formulates sample selection as a generate-to-retrieve process, in which diffusion-based optimization in feature space guided by information gain produces synthetic probes, and only the real unlabeled samples retrieved by these probes are annotated and used for downstream training. Across six benchmarks covering fifteen datasets and 16,696 images, we evaluate EvoDiff on convolutional network and transformer-based foundation model. EvoDiff approaches fully supervised performance with substantially fewer labels, outperforms active learning baselines, and improves foundation model generalization under cross-domain external validation on unseen cohorts. A reader study with six radiologists indicates the potential to support clinical annotation workflows, with a model-based estimated 24% to 84% reduction in annotation workload at matched downstream performance.

Figures & tables

Explore similar work

CardsList
  1. Robust Cross-Domain Generalization Using Unlabeled Target Data with Source-Domain Supervision

    May 27, 2026Yuyue Zhou, Shrimanti Ghosh, Michael +7Ultrasound Image SegmentationUnsupervised Domain Adaptation

  2. AWDiff: An a trous wavelet diffusion model for lung ultrasound image synthesis

    Mar 3, 2026Maryam Heidari, Nantheera Anantrasirichai, Steven Walker +2Medical Image SynthesisSynthetic Data Augmentation

  3. Unified Ultrasound Intelligence Toward an End-to-End Agentic System

    Apr 18, 2026Chen Ma, Yunshu Li, Junhu Fu +3Medical Imaging Foundation ModelsMulti-Task Learning