A research group led by Prof. LI Hai from the Hefei Institutes of Physical Science of the Chinese Academy of Sciences has developed MultiXpert, a new artificial intelligence (AI) medical diagnostic system for zero-shot intelligent diagnosis of chest X-rays.

The results were published in Information Processing and Management.

Chest X-rays are commonly used to screen for diseases such as pneumonia, nodules, and pneumothorax; however, manual interpretation is time-consuming and dependent on the expertise of the reader. Although traditional AI systems can match experts in some tasks, they rely on large annotated datasets, which makes them less effective for new diseases or differences between hospitals. Consequently, their generalization is limited, and they struggle to meet the needs of complex clinical settings.

In this study, the researchers proposed a multimodal, dual-stream collaborative enhancement approach and developed MultiXpert, a high-precision zero-shot diagnostic framework that requires no additional annotated data. The model processes image and text information simultaneously, integrating large language models with radiology expert knowledge to refine lesion descriptions. This deep fusion of visual and linguistic information enables the AI to "understand" chest X-rays in a manner similar to clinical reasoning, even for previously unseen diseases.

Experiments show that MultiXpert improves the average area under the receiver operating characteristic curve (AUC) by 7.5% across four single-label public datasets. In zero-shot settings, MultiXpert outperforms mainstream vision-language models by an average of 3.9%. On multicenter private datasets from ten hospitals, MultiXpert outperformed traditional supervised models, demonstrating strong cross-center generalization and clinical adaptability.

This work offers a new approach for zero-shot chest X-ray diagnosis and marks a shift in medical AI from relying on annotations to achieving autonomous understanding, according to the researchers.