Moderate evidence
More
AI model spots fatty liver disease with 0.926-0.972 accuracy score in small study of 179 people
Study title: Machine learning-based discrimination of metabolic dysfunction-associated steatotic liver disease: implications of tryptophan metabolites as potential biomarkers
- Evidence type
- Observational study
- Sample
- n = 179
- Sources
- 1 sourcePeer-reviewed
- Published
- 1 min read
- Truvace record
- TRV-2026-1223cite or check it
Study title: Machine learning-based discrimination of metabolic dysfunction-associated steatotic liver disease: implications of tryptophan metabolites as potential biomarkers
In a 179-person study of 80 healthy controls and 99 MASLD patients published September 2026, investigators used non-targeted and targeted metabolomics to identify dysregulated tryptophan metabolites and trained seven machine learning classifiers for MASLD discrimination, selecting a Random Forest model that combines kynurenine, neopterin and five routine labs.
Read the study abstract
Objectives Early screening of individuals at high risk of metabolic dysfunction-associated steatotic liver disease (MASLD) facilitates timely intervention and may reduce the risk of progression to cirrhosis. This study aimed to develop a reliable and clinically practical model for MASLD discrimination. Methods A total of 179 individuals were recruited, including 80 healthy controls and 99 MASLD patients. Non-targeted and targeted metabolomics were performed to identify and validate candidate biomarkers associate…

The gain
In a 179-person study (80 healthy, 99 with fatty liver disease), a model combining kynurenine, neopterin and five routine lab tests separated patients from healthy people with AUC 0.926-0.972.
In brief
The approach matters because early identification of high-risk MASLD could enable timely intervention to reduce progression to cirrhosis, but the reported excellent performance is from a small internal cohort and the authors frame the tool as promising and requiring further validation and mechanistic work.
Main points
- Study recruited 179 individuals including 80 healthy controls and 99 MASLD patients.
- Non-targeted metabolomics identified tryptophan metabolism as the most significantly enriched pathway in MASLD.
- Targeted metabolomics verified serum tryptophan metabolites were significantly dysregulated between groups.
- Final model used kynurenine, neopterin, triglyceride, high-density lipoprotein cholesterol, uric acid, alanine aminotransferase, and glucose.
The full story
Researchers performed non-targeted metabolomics to find enriched pathways, then targeted metabolomics to validate tryptophan metabolites, and built seven machine learning algorithms assessed by accuracy, sensitivity, specificity and AUC with Shapley Additive Explanations for interpretation.
The selected Random Forest model integrated kynurenine and neopterin with triglyceride, HDL cholesterol, uric acid, ALT and glucose, achieving AUC 0.926-0.972 across evaluations in the 179-person cohort.
Sources
1 on file- Clinica Chimica Actadoi:10.1016/j.cca.2026.122854
Talk it over
The Random Forest model combined kynurenine and neopterin with five routine lab tests and reached AUC 0.926-0.972. How would you weigh that strong score against the fact that it comes from one small cohort of 179 people?
AI model spots fatty liver disease with 0.926-0.972 accuracy score in small study of 179 peoplehttps://truvace.com/item/machine-learning-based-discrimination-of-metabolic-dysfunction-associated-steato
The Random Forest model combined kynurenine and neopterin with five routine lab tests and reached AUC 0.926-0.972. How would you weigh that strong score against the fact that it comes from one small cohort of 179 people?
Answer in the threadThe study compared 80 healthy controls with 99 people who have MASLD. How could leaving out people with other liver or metabolic conditions change how well the model works in a real clinic?
Answer in the threadThe authors say the model needs further validation before clinical use. What finding in a new group of patients would settle whether the tryptophan metabolites add anything beyond the routine labs?
Answer in the thread
If you have more time
Look up what the five routine lab tests alone could do. Why would that comparison matter for deciding if kynurenine and neopterin are worth measuring?
Answer in the threadThe authors want early identification to reduce progression to cirrhosis. How would a clinic or health system you know decide who gets tested, and what evidence would it need?
Answer in the thread
ace
What readers say