Towards robust zero-shot Chest X-Ray (CXR) classification: Exploring data distribution bias in CXR datasets

Bhat S, B. Panambur A, Mansoor A, Georgescu B, Ghesu FC, Grbic S, Maier A (2025)

Publication Language: English

Publication Type: Conference contribution, Conference Contribution

Publication year: 2025

Event location: Regensburg DE

Abstract

In recent years, unsupervised classification models have become increasingly

significant, primarily due to the difficulties associated with data labeling

and its costs. This trend is also notable in the field of medical imaging, particularly

with Chest X-rays (CXRs). Among the various unsupervised pretraining

methodologies, image-text models like CLIP are highlighted for their considerable

enhancements in zero-shot classification. In this study, we perform a detailed

analysis of CLIP’s performance using multiple large CXR datasets, investigating

how the batch size, dataset size, and distribution biases differentially influence

outcomes across various findings. In two distinct experiments,we showan average

of 3% enhancement in the macro average zero-shot AUC scores when the batch

size is increased, and a corresponding 8% improvement for pneumothorax by the

addition of a second dataset. For pleural effusion, where performance is nearly

saturated and previous changes had little effect, we examine adding weak supervisory

meta-labels and image-to-image contrastive loss, achieving an average

1% improvement in zero-shot AUC. Consequently, our work shows incorporating

dataset insights, meta-information and contrastive learning strategies enhances

the robustness and accuracy of CLIP-CXR for specific findings.

Authors with CRIS profile

Sheethal Bhat Lehrstuhl für Informatik 14 (Bild- und Sprachverarbeitung) (LME) Andreas Maier Lehrstuhl für Informatik 5 (Mustererkennung)

Involved external institutions

Siemens AG, Healthcare Sector DE Germany (DE) Siemens Healthineers DE Germany (DE)

How to cite

APA:

Bhat, S., B. Panambur, A., Mansoor, A., Georgescu, B., Ghesu, F.C., Grbic, S., & Maier, A. (2025). Towards robust zero-shot Chest X-Ray (CXR) classification: Exploring data distribution bias in CXR datasets. In Proceedings of the Bildverarbeitung für die Medizin 2025. Regensburg, DE.

MLA:

Bhat, Sheethal, et al. "Towards robust zero-shot Chest X-Ray (CXR) classification: Exploring data distribution bias in CXR datasets." Proceedings of the Bildverarbeitung für die Medizin 2025, Regensburg 2025.

BibTeX: Download