Development and Evaluation of a Multiparametric Ultrasound Approach for Quantifying Hepatic Steatosis

Area of Science:

• Medical Imaging
• Hepatology
• Biophysics

Background:

• Noninvasive methods are vital for detecting and monitoring liver steatosis.
• Current methods require improvement for accurate and accessible assessment.

Purpose of the Study:

• To evaluate ultrasound markers for liver steatosis quantification.
• To develop an ultrasound-based fat fraction (UFF) correlated with MRI-PDFF.

Main Methods:

• Assessed three ultrasound markers (attenuation, backscattering, Nakagami distribution) using intercostal and subcostal approaches.
• Calculated UFF by combining markers and compared it with MRI-PDFF in 133 patients.
• Utilized polynomial fits, Pearson correlation, and ROC curves for analysis.

Main Results:

• Individual ultrasound markers showed good correlation with PDFF (r² 0.51-0.72).
• The combined UFF demonstrated a strong correlation with PDFF (r² = 0.79).
• ROC analysis showed high performance in distinguishing steatosis grades (AUROCs 0.94-0.98).

Conclusions:

• UFF is a reliable, noninvasive method for liver steatosis quantification.
• Combining intercostal and subcostal ultrasound measurements enhances UFF accuracy.