Gong Feng1, Kenneth I Zheng2, Yang-Yang Li3, Rafael S Rios2, Pei-Wu Zhu4, Xiao-Yan Pan5, Gang Li2, Hong-Lei Ma2, Liang-Jie Tang2, Christopher D Byrne6, Targher Giovanni7, Na He8, Man Mi1, Yong-Ping Chen2,9,10, Ming-Hua Zheng2,9,10. 1. Xi'an Medical University, Xi'an, China. 2. NAFLD Research Center, Department of Hepatology, the First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China. 3. Department of Pathology, the First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China. 4. Department of Laboratory Medicine, the First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China. 5. Department of Endocrinology, the First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China. 6. Southampton National Institute for Health Research Biomedical Research Centre, University Hospital Southampton, Southampton General Hospital, Southampton, UK. 7. Section of Endocrinology, Diabetes and Metabolism, Department of Medicine, University and Azienda Ospedaliera Universitaria Integrata of Verona, Verona, Italy. 8. Department of Gastroenterology, the First Affiliated Hospital of Xi'an Medical University, Xi'an, China. 9. Institute of Hepatology, Wenzhou Medical University, Wenzhou, China. 10. Key Laboratory of Diagnosis and Treatment for The Development of Chronic Liver Disease in Zhejiang Province, Wenzhou, China.
Abstract
BACKGROUND: The presence of significant liver fibrosis is a key determinant of long-term prognosis in non-alcoholic fatty liver disease (NAFLD). We aimed to develop a novel machine learning algorithm (MLA) to predict fibrosis severity in NAFLD and compared it with the most widely used non-invasive fibrosis biomarkers. METHODS: We used a cohort of 553 adults with biopsy-proven NAFLD, who were randomly divided into a training cohort (n=278) for the development of both logistic regression model (LRM) and MLA, and a validation cohort (n=275). Significant fibrosis was defined as fibrosis stage F≥2. MLA and LRM were derived from variables that were selected using a least absolute shrinkage and selection operator (LASSO) logistic regression algorithm. RESULTS: In the training cohort, the variables selected by LASSO algorithm were body mass index, pro-collagen type III, collagen type IV, aspartate aminotransferase and albumin-to-globulin ratio. The diagnostic accuracy of MLA showed the highest values of area under the receiver operator characteristic curve (AUROC: 0.902, 95%CI 0.869-0.904) for identifying fibrosis F≥2. The LRM AUROC was 0.764, 95%CI 0.710-0.816) and significantly better than the AST-to-Platelet ratio (AUROC 0.684, 95%CI 0.605-0.762), FIB-4 score (AUROC 0.594, 95%CI 0.503-0.685) and NAFLD Fibrosis Score (AUROC 0.557, 95%CI 0.470-0.644). In the validation cohort, MLA also showed the highest AUROC (0.893, 95%CI 0.864-0.901). The diagnostic accuracy of MLA outperformed that of LRM in all subgroups considered. CONCLUSIONS: Our newly developed MLA algorithm has excellent diagnostic performance for predicting fibrosis F≥2 in patients with biopsy-confirmed NAFLD. This article is protected by copyright. All rights reserved.
BACKGROUND: The presence of significant liver fibrosis is a key determinant of long-term prognosis in non-alcoholic fatty liver disease (NAFLD). We aimed to develop a novel machine learning algorithm (MLA) to predict fibrosis severity in NAFLD and compared it with the most widely used non-invasive fibrosis biomarkers. METHODS: We used a cohort of 553 adults with biopsy-proven NAFLD, who were randomly divided into a training cohort (n=278) for the development of both logistic regression model (LRM) and MLA, and a validation cohort (n=275). Significant fibrosis was defined as fibrosis stage F≥2. MLA and LRM were derived from variables that were selected using a least absolute shrinkage and selection operator (LASSO) logistic regression algorithm. RESULTS: In the training cohort, the variables selected by LASSO algorithm were body mass index, pro-collagen type III, collagen type IV, aspartate aminotransferase and albumin-to-globulin ratio. The diagnostic accuracy of MLA showed the highest values of area under the receiver operator characteristic curve (AUROC: 0.902, 95%CI 0.869-0.904) for identifying fibrosis F≥2. The LRM AUROC was 0.764, 95%CI 0.710-0.816) and significantly better than the AST-to-Platelet ratio (AUROC 0.684, 95%CI 0.605-0.762), FIB-4 score (AUROC 0.594, 95%CI 0.503-0.685) and NAFLD Fibrosis Score (AUROC 0.557, 95%CI 0.470-0.644). In the validation cohort, MLA also showed the highest AUROC (0.893, 95%CI 0.864-0.901). The diagnostic accuracy of MLA outperformed that of LRM in all subgroups considered. CONCLUSIONS: Our newly developed MLA algorithm has excellent diagnostic performance for predicting fibrosis F≥2 in patients with biopsy-confirmed NAFLD. This article is protected by copyright. All rights reserved.
Authors: Paolo Sorino; Angelo Campanella; Caterina Bonfiglio; Antonella Mirizzi; Isabella Franco; Antonella Bianco; Maria Gabriella Caruso; Giovanni Misciagna; Laura R Aballay; Claudia Buongiorno; Rosalba Liuzzi; Anna Maria Cisternino; Maria Notarnicola; Marisa Chiloiro; Francesca Fallucchi; Giovanni Pascoschi; Alberto Rubén Osella Journal: Sci Rep Date: 2021-10-12 Impact factor: 4.379