1
|
Verma N, Duseja A, Mehta M, De A, Lin H, Wong VWS, Wong GLH, Rajaram RB, Chan WK, Mahadeva S, Zheng MH, Liu WY, Treeprasertsuk S, Prasoppokakorn T, Kakizaki S, Seki Y, Kasama K, Charatcharoenwitthaya P, Sathirawich P, Kulkarni A, Purnomo HD, Kamani L, Lee YY, Wong MS, Tan EXX, Young DY. 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. Aliment Pharmacol Ther 2024; 59:774-788. [PMID: 38303507 DOI: 10.1111/apt.17891] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/29/2023] [Revised: 11/28/2023] [Accepted: 01/20/2024] [Indexed: 02/03/2024]
Abstract
BACKGROUND The precise estimation of cases with significant fibrosis (SF) is an unmet goal in non-alcoholic fatty liver disease (NAFLD/MASLD). AIMS We evaluated the performance of machine learning (ML) and non-patented scores for ruling out SF among NAFLD/MASLD patients. METHODS 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). RESULTS 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). CONCLUSIONS 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.
Collapse
Affiliation(s)
- Nipun Verma
- Department of Hepatology, Postgraduate Institute of Medical Education and Research, Chandigarh, India
| | - Ajay Duseja
- Department of Hepatology, Postgraduate Institute of Medical Education and Research, Chandigarh, India
| | - Manu Mehta
- Department of Hepatology, Postgraduate Institute of Medical Education and Research, Chandigarh, India
| | - Arka De
- Department of Hepatology, Postgraduate Institute of Medical Education and Research, Chandigarh, India
| | - Huapeng Lin
- Department of Medicine and Therapeutics, The Chinese University of Hong Kong, Hong Kong, China
| | - Vincent Wai-Sun Wong
- Department of Medicine and Therapeutics, The Chinese University of Hong Kong, Hong Kong, China
| | - Grace Lai-Hung Wong
- Department of Medicine and Therapeutics, The Chinese University of Hong Kong, Hong Kong, China
| | - Ruveena Bhavani Rajaram
- Gastroenterology and Hepatology Unit, Department of Medicine, Faculty of Medicine, University of Malaya Medical Centre, Kuala Lumpur, Malaysia
| | - Wah-Kheong Chan
- Gastroenterology and Hepatology Unit, Department of Medicine, Faculty of Medicine, University of Malaya Medical Centre, Kuala Lumpur, Malaysia
| | - Sanjiv Mahadeva
- Gastroenterology and Hepatology Unit, Department of Medicine, Faculty of Medicine, University of Malaya Medical Centre, Kuala Lumpur, Malaysia
| | - Ming-Hua Zheng
- NAFLD Research Centre Department of Hepatology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China
| | - Wen-Yue Liu
- Department of Endocrinology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China
| | - Sombat Treeprasertsuk
- Division of Gastroenterology, King Chulalongkorn Memorial Hospital, Chulalongkorn University, Bangkok, Thailand
| | - Thaninee Prasoppokakorn
- Division of Gastroenterology, King Chulalongkorn Memorial Hospital, Chulalongkorn University, Bangkok, Thailand
| | - Satoru Kakizaki
- Department of Clinical Research, National Hospital Organization Takasaki General Medical Centre, Takasaki, Japan
| | - Yosuke Seki
- Weight Loss and Metabolic Surgery Centre, Yotsuya Medical Cube, Tokyo, Japan
| | - Kazunori Kasama
- Weight Loss and Metabolic Surgery Centre, Yotsuya Medical Cube, Tokyo, Japan
| | | | - Phalath Sathirawich
- Division of Gastroenterology, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand
| | - Anand Kulkarni
- Asian Institute of Gastroenterology Hospital, Hyderabad, India
| | - Hery Djagat Purnomo
- Faculty of Medicine, Diponegoro University, Kariadi Hospital, Semarang, Indonesia
| | | | - Yeong Yeh Lee
- School of Medical Sciences Universiti Sains Malaysia, Kota Bharu, Malaysia
| | - Mung Seong Wong
- School of Medical Sciences Universiti Sains Malaysia, Kota Bharu, Malaysia
| | - Eunice X X Tan
- Department of Medicine, National University Singapore, Singapore, Singapore
| | - Dan Yock Young
- Department of Medicine, National University Singapore, Singapore, Singapore
| |
Collapse
|