Using AI and Gd-EOB-DTPA-enhanced MR imaging to assess liver function, comparing the MELIF score with the ALBI score

Area of Science:

• Hepatology
• Medical Imaging
• Artificial Intelligence in Medicine

Background:

• Effective monitoring of liver disease progression is crucial for treatment optimization, particularly for reversible conditions.
• Current tools often lack sensitivity for early or mild liver disease stages.
• There is a growing need for advanced liver function monitoring tools due to anticipated increases in liver disease prevalence.

Purpose of the Study:

• To evaluate the performance of the novel T1 relaxometry-based MELIF score in assessing liver function.
• To compare the MELIF score with the established ALBI score for monitoring liver disease.
• To assess the utility of the MELIF score in identifying liver dysfunction and cirrhosis.

Main Methods:

• Retrospective analysis of liver function scores (ALBI and MELIF) in 150 patients.
• Comparison of MELIF and ALBI scores against MELD and Child-Pugh scores using Pearson correlation.
• Assessment of diagnostic accuracy using Area Under the Curve (AUC) for discriminating liver function and cirrhosis.

Main Results:

• The MELIF score demonstrated superior performance compared to the ALBI score in discriminating between good and impaired liver function (AUC: MELIF 0.80 vs. ALBI 0.77).
• MELIF showed a higher accuracy in distinguishing patients with and without cirrhosis (AUC: MELIF 0.83 vs. ALBI 0.79).
• The AI-driven MELIF score is fully automated and shows potential for monitoring mild liver disease progression.

Conclusions:

• The AI-based MELIF score is a promising tool for monitoring liver function, especially in early or reversible stages of liver disease.
• MELIF offers advantages over the ALBI score in assessing liver function and detecting cirrhosis.
• This novel score may help bridge the gap in monitoring tools for potentially reversible liver conditions before chronic disease develops.