AI enables early detection of fatty liver disease via chest X-rays

Current diagnostic tools such as ultrasound, CT scans, and MRI require access to specialized equipment and are often associated with higher costs and limited availability - especially in resource-limited settings. By contrast, chest X-rays are performed more routinely, are less expensive, and expose patients to relatively low radiation levels. Though primarily used to assess the lungs and heart, these images also capture parts of the liver, presenting an underexplored opportunity for liver disease screening. 

To explore this potential, a team of medical and AI researchers at Osaka Metropolitan University, led by Associate Professors Sawako Uchida-Kobayashi and Daiju Ueda, created an artificial intelligence model capable of detecting signs of fatty liver disease from standard chest X-rays. 

In their retrospective study, 6,599 chest X-ray images from 4,414 patients were analyzed. The model was trained using controlled attenuation parameter (CAP) scores, a widely accepted method to assess fat accumulation in the liver. Evaluation results showed a high diagnostic performance, with an AUC (area under the receiver operating characteristic curve) between 0.82 and 0.83, indicating strong predictive accuracy. 

The development opens up possibilities for incorporating AI-based liver screening into routine radiological workflows, especially where advanced imaging equipment is not available. 

“The development of diagnostic methods using easily obtainable and inexpensive chest X-rays has the potential to improve fatty liver detection. We hope it can be put into practical use in the future,” said Professor Uchida-Kobayashi. 

MEDICA-tradefair.com; Source: Osaka Metropolitan University