Work

MedCutMix: A Data-Centric Approach to Improve Radiology Vision-Language Pre-Training with Disease Awareness

Year: 2026

Type: conference-paper

Abstract: Vision-Language Pre-training (VLP) is drawing increasing interest for its ability to minimize manual annotation requirements while enhancing semantic understanding in downstream tasks. However, its reliance on image-text datasets poses challenges due to privacy concerns and the high cost of obtaining paired annotations. Data augmentation emerges as a viable strategy to address this issue, yet existing methods often fall short of capturing the subtle and complex variations in medical data due to limited diversity. To this end, we propose MedCutMix, a novel multi-modal disease-centric data augmentation method. MedCutMix performs diagnostic sentence CutMix within medical reports and establishes the cross-attention between the diagnostic sentence and medical image to guide attentive manifold mix within the imaging modality. Our approach surpasses previous methods across four downstream radiology diagnosis datasets, highlighting its effectiveness in enhancing performance and generalizability in radiology VLP.

Source: IEEE International Conference on Acoustics Speech and Signal Processing

DOI: 10.1109/icassp55912.2026.11462043

Authors: Sinuo Wang, Yutong Xie, Yuyuan Liu, Qi Wu

Institutions: The University of Adelaide, Mohamed bin Zayed University of Artificial Intelligence, University of Oxford

Language: English

Topic: Multimodal Machine Learning Applications

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

SDG: Good health and well-being

Open Access status: green

OpenAlex ID: W4414744657

FWCI:

Cites:

Cited by:

Related to: