Machine learning improves the prediction of significant fibrosis in Asian patients with metabolic dysfunction-associated steatotic liver disease - The Gut and Obesity in Asia (GO-ASIA) Study

• Published in: Alimentary pharmacology & therapeutics Vol. 59; no. 6; pp. 774 - 788
• Main Authors: Verma, Nipun, Duseja, Ajay, Mehta, Manu, De, Arka, Lin, Huapeng, Wong, Vincent Wai-Sun, Wong, Grace Lai-Hung, Rajaram, Ruveena Bhavani, Chan, Wah-Kheong, Mahadeva, Sanjiv, Zheng, Ming-Hua, Liu, Wen-Yue, Treeprasertsuk, Sombat, Prasoppokakorn, Thaninee, Kakizaki, Satoru, Seki, Yosuke, Kasama, Kazunori, Charatcharoenwitthaya, Phunchai, Sathirawich, Phalath, Kulkarni, Anand, Purnomo, Hery Djagat, Kamani, Lubna, Lee, Yeong Yeh, Wong, Mung Seong, Tan, Eunice X X, Young, Dan Yock
• Format: Journal Article
• Language: English
• Published: England Wiley Subscription Services, Inc 01.03.2024
• Subjects: Asia - epidemiology; Aspartate Aminotransferases; Biopsy; Blood Glucose; Diabetes mellitus; Fasting; Fatty liver; Female; Fibrosis; Humans; Laboratory testing; Learning algorithms; Liver - pathology; Liver Cirrhosis - complications; Liver diseases; Machine learning; Male; Metabolic syndrome; Metabolic Syndrome - complications; Non-alcoholic Fatty Liver Disease - diagnosis; Non-alcoholic Fatty Liver Disease - epidemiology; Obesity - complications; Asia; mortality; liver fibrosis; fatty liver; NASH; artificial intelligence
• ISSN: 0269-2813; 1365-2036; 1365-2036
• DOI: 10.1111/apt.17891

The precise estimation of cases with significant fibrosis (SF) is an unmet goal in non-alcoholic fatty liver disease (NAFLD/MASLD). We evaluated the performance of machine learning (ML) and non-patented scores for ruling out SF among NAFLD/MASLD patients. Twenty-one ML models were trained (N = 1153), tested (N = 283), and validated (N = 220) on clinical and biochemical parameters of histologically-proven NAFLD/MASLD patients (N = 1656) collected across 14 centres in 8 Asian countries. Their performance for detecting histological-SF (≥F2fibrosis) were evaluated with APRI, FIB4, NFS, BARD, and SAFE (NPV/F1-score as model-selection criteria). Patients aged 47 years (median), 54.6% males, 73.7% with metabolic syndrome, and 32.9% with histological-SF were included in the study. Patients with SFvs.no-SF had higher age, aminotransferases, fasting plasma glucose, metabolic syndrome, uncontrolled diabetes, and NAFLD activity score (p < 0.001, each). ML models showed 7%-12% better discrimination than FIB-4 to detect SF. Optimised random forest (RF) yielded best NPV/F1 in overall set (0.947/0.754), test set (0.798/0.588) and validation set (0.852/0.559), as compared to FIB4 in overall set (0.744/0.499), test set (0.722/0.456), and validation set (0.806/0.507). Compared to FIB-4, RF could pick 10 times more patients with SF, reduce unnecessary referrals by 28%, and prevent missed referrals by 78%. Age, AST, ALT fasting plasma glucose, and platelet count were top features determining the SF. Sequential use of SAFE < 140 and FIB4 < 1.2 (when SAFE > 140) was next best in ruling out SF (NPV of 0.757, 0.724 and 0.827 in overall, test and validation set). ML with clinical, anthropometric data and simple blood investigations perform better than FIB-4 for ruling out SF in biopsy-proven Asian NAFLD/MASLD patients.