Introduction: Beyond Uniform Fat Accumulation

Clinicians traditionally viewed hepatic steatosis as a relatively uniform process. However, recent evidence suggests that intrahepatic fat spatial heterogeneity plays a critical role in distinguishing between various steatotic liver disease (SLD) subcategories. Understanding these distribution patterns is essential for improving diagnostic accuracy and tailoring patient management. A landmark multicenter study has now utilized artificial intelligence and advanced MRI to map these variations across metabolic, alcoholic, and viral-related liver diseases.

Mapping Intrahepatic Fat Spatial Heterogeneity with AI

The research team analyzed data from 682 patients diagnosed with hepatic steatosis, including those with metabolic dysfunction-associated steatotic liver disease (MASLD), alcohol-related liver disease (ALD), and SLD with chronic hepatitis B (SLD&CHB). Researchers employed a fully automated nnU-Net model for whole-liver segmentation. They coupled this with two-point Dixon MRI-based fat fraction (FF) mapping to evaluate spatial patterns. The AI model achieved impressive Dice similarity coefficient (DSC) scores, ensuring high precision in separating the left and right lobes for detailed analysis.

Distinct Regional Fat Patterns and Gradients

The study revealed that MASLD patients exhibited the highest fat burden and the most significant spatial variation. Furthermore, the researchers identified a higher fat concentration in the right lobe compared to the left lobe specifically in MASLD and ALD cases. Conversely, this lobar asymmetry was not statistically significant in patients with SLD&CHB. Moreover, the study investigated the periportal-to-peripheral FF gradients. In MASLD, this gradient increased significantly as steatosis severity progressed from mild to severe. However, the opposite trend occurred in ALD and SLD&CHB, where the gradient actually decreased with worsening disease levels.

Clinical Implications for Liver Diagnostics

These findings suggest that fat distribution is not random but follows disease-specific biological rules. Therefore, spatial profiling could serve as a non-invasive biomarker to differentiate between SLD subcategories. Specifically, the high heterogeneity and right-lobe dominance in MASLD may help identify patients at higher metabolic risk. Consequently, these insights allow for more localized monitoring and could influence how clinicians perform biopsies or interpret regional imaging findings in the future.

Frequently Asked Questions

What is the clinical significance of intrahepatic fat spatial heterogeneity?

It helps clinicians differentiate between MASLD, ALD, and SLD&CHB by identifying unique regional fat accumulation patterns and gradients that are specific to each disease etiology.

How does MASLD fat distribution differ from other liver diseases?

MASLD typically shows the highest overall fat fraction and the greatest spatial variability, with fat preferentially accumulating in the right lobe and increasing in a periportal-to-peripheral gradient as the disease worsens.

Can AI improve the accuracy of liver fat quantification?

Yes, fully automated AI models like nnU-Net provide precise and consistent segmentation of liver lobes, allowing for more detailed analysis of spatial fat distribution than manual methods.

Disclaimer: This content is for informational and educational purposes only and does not constitute medical advice. It is not intended to be a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a medical condition. Refer to the latest local and national guidelines for clinical practice.

References

Zhao YC et al. Spatial characterization of intrahepatic fat deposition across steatotic liver disease subcategories: a multicenter study. BMC Med. 2026 Feb 18. doi: 10.1186/s12916-026-04706-1. PMID: 41703539.