1
|
Xian MF, Lan WT, Zhang Z, Li MD, Lin XX, Huang Y, Huang H, Chen LD, Huang QH, Wang W. Enhancing hepatocellular carcinoma diagnosis in non-high-risk patients: a customized ChatGPT model integrating contrast-enhanced ultrasound. LA RADIOLOGIA MEDICA 2025:10.1007/s11547-025-01994-0. [PMID: 40232657 DOI: 10.1007/s11547-025-01994-0] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/10/2024] [Accepted: 03/05/2025] [Indexed: 04/16/2025]
Abstract
PURPOSE This study aims to improve hepatocellular carcinoma (HCC) diagnostic accuracy in non-high-risk populations by utilizing GPTs that incorporate integrated risk coefficients, and to explore its feasibility. MATERIAL AND METHODS Between August 2016 and June 2019, patients with focal liver lesions (FLLs) in non-high-risk populations, confirmed by histopathology or clinical/imaging evidence, were retrospectively included. A logistic regression model was developed using baseline characteristics and contrast-enhanced ultrasound (CEUS) features to identify independent HCC risk factors. Three ChatGPT-based models were evaluated: ChatGPT 4o (a general-purpose model developed by OpenAI), BaseGPT (a customized model with HCC diagnostic knowledge), and RiskGPT (a further customized model integrating HCC knowledge and identified risk factors). Their intra-agreement and diagnostic performance were compared. RESULTS Logistic regression identified male, obesity, HBcAb or HBeAb positivity, elevated alpha-fetoprotein, and mild washout on CEUS as associated with HCC. RiskGPT achieved the highest area under a receiver operating characteristic curve (AUC) (0.89) and demonstrated superior accuracy (90.3%) in HCC identification; significantly outperforming both ChatGPT 4o (AUC 0.79, P = 0.002; accuracy 83.1%, P = 0.02) and BaseGPT (AUC 0.81, P = 0.008; accuracy 80.6%, P = 0.002). RiskGPT demonstrated superior sensitivity compared to ChatGPT 4o (85.5% vs. 66.3%) and outperformed BaseGPT in specificity (92.7% vs. 80.6%) and positive predictive value (85.5% vs. 67.7%) (all P < 0.001). Additionally, RiskGPT showed substantial intra-consistency in diagnosing FLLs, with a κ value of 0.78. CONCLUSION RiskGPT improves HCC diagnostic accuracy in non-high-risk patients by integrating clinical, imaging features, and risk coefficients, demonstrating significant diagnostic potential.
Collapse
Affiliation(s)
- Meng-Fei Xian
- Department of Medical Ultrasonics, Ultrasomics Artificial Intelligence X-Laboratory, Institute of Diagnostic and Interventional Ultrasound, The First Affiliated Hospital of Sun Yat-Sen University, 58 Zhongshan Road 2, Guangzhou, 510080, China
| | - Wen-Tong Lan
- Department of Endoscopy Center, The First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, 510080, China
| | - Zhe Zhang
- Department of Biomedical Engineering, College of Engineering and Applied Sciences, State University of New York at Stony Brook, Stony Brook, NY, USA
| | - Ming-De Li
- Department of Medical Ultrasonics, Ultrasomics Artificial Intelligence X-Laboratory, Institute of Diagnostic and Interventional Ultrasound, The First Affiliated Hospital of Sun Yat-Sen University, 58 Zhongshan Road 2, Guangzhou, 510080, China
| | - Xin-Xin Lin
- Department of Medical Ultrasonics, Ultrasomics Artificial Intelligence X-Laboratory, Institute of Diagnostic and Interventional Ultrasound, The First Affiliated Hospital of Sun Yat-Sen University, 58 Zhongshan Road 2, Guangzhou, 510080, China
| | - Yang Huang
- Department of Vascular Surgery, National-Guangdong Joint Engineering Laboratory for Diagnosis and Treatment of Vascular Diseases, The First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, 510080, China
| | - Hui Huang
- Department of Medical Ultrasonics, Ultrasomics Artificial Intelligence X-Laboratory, Institute of Diagnostic and Interventional Ultrasound, The First Affiliated Hospital of Sun Yat-Sen University, 58 Zhongshan Road 2, Guangzhou, 510080, China
| | - Li-Da Chen
- Department of Medical Ultrasonics, Ultrasomics Artificial Intelligence X-Laboratory, Institute of Diagnostic and Interventional Ultrasound, The First Affiliated Hospital of Sun Yat-Sen University, 58 Zhongshan Road 2, Guangzhou, 510080, China
| | - Qing-Hua Huang
- School of Artificial Intelligence, Optics and Electronics (iOPEN), Northwestern Polytechnical University, Xian, 710072, China.
| | - Wei Wang
- Department of Medical Ultrasonics, Ultrasomics Artificial Intelligence X-Laboratory, Institute of Diagnostic and Interventional Ultrasound, The First Affiliated Hospital of Sun Yat-Sen University, 58 Zhongshan Road 2, Guangzhou, 510080, China.
| |
Collapse
|
2
|
Elgenidy A, Abubasheer TM, Odat RM, Abdelrahim MG, Jibril NS, Ramadan AM, Ballut L, Haseeb ME, Ragab A, Ismail AM, Afifi AM, Mohamed BJ, Jalal PK. Assessing the Predictive Accuracy of the aMAP Risk Score for Hepatocellular Carcinoma (HCC): Diagnostic Test Accuracy and Meta-analysis. J Clin Exp Hepatol 2025; 15:102381. [PMID: 39262566 PMCID: PMC11386263 DOI: 10.1016/j.jceh.2024.102381] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 02/08/2024] [Accepted: 07/21/2024] [Indexed: 09/13/2024] Open
Abstract
Purpose We aimed to perform a meta-analysis with the intention of evaluating the reliability and test accuracy of the aMAP risk score in the identification of HCC. Methods A systematic search was performed in PubMed, Scopus, Cochrane, Embase, and Web of Science databases from inception to September 2023, to identify studies measuring the aMAP score in patients for the purpose of predicting the occurrence or recurrence of HCC. The meta-analysis was performed using the meta package in R version 4.1.0. The diagnostic accuracy meta-analysis was conducted using Meta-DiSc software. Results Thirty-five studies 102,959 participants were included in the review. The aMAP score was significantly higher in the HCC group than in the non-HCC group, with a mean difference of 6.15. When the aMAP score is at 50, the pooled sensitivity, specificity, negative likelihood ratio, and positive likelihood ratio with 95% CI was 0.961 (95% CI 0.936, 0.976), 0.344 (95% CI 0.227, 0.483), 0.114 (95% CI 0.087, 0.15), and 1.464 (95% CI 1.22, 1.756), respectively. At a cutoff value of 60, the pooled sensitivity, specificity, negative likelihood ratio, and positive likelihood ratio with 95% CI was 0.594 (95% CI 0.492, 0.689), 0.816 (95% CI 0.714, 0.888), 0.497 (95% CI 0.418, 0.591), and 3.235 (95% CI 2.284, 4.582), respectively. Conclusion The aMAP score is a reliable, accurate, and easy-to-use tool for predicting HCC patients of all stages, including early-stage HCC. Therefore, the aMAP score can be a valuable tool for surveillance of HCC patients and can help to improve early detection and reduce mortality.
Collapse
Affiliation(s)
| | - Tareq M Abubasheer
- Faculty of Medicine, Al-Quds University (Al-Azhar Branch), Gaza, Palestine
| | - Ramez M Odat
- Faculty of Medicine, Jordan University of Science and Technology, Irbid, Jordan
| | | | - Nada S Jibril
- Faculty of Medicine, Menofia University, Menofia, Egypt
| | - Aya M Ramadan
- Faculty of Medicine, Menofia University, Menofia, Egypt
| | | | | | | | | | - Ahmed M Afifi
- Department of Surgery, University of Toledo Medical Center, USA
| | - Benarad J Mohamed
- Oncology Department UClouvain, University Catholic Louvain, Brussels, Belgium
| | - Prasun K Jalal
- Division of Gastroenterology, Baylor College of Medicine, Houston, TX, 77030, USA
| |
Collapse
|
3
|
Mak LY, Liu K, Chirapongsathorn S, Yew KC, Tamaki N, Rajaram RB, Panlilio MT, Lui R, Lee HW, Lai JCT, Kulkarni AV, Premkumar M, Lesmana CRA, Hsu YC, Huang DQ. Liver diseases and hepatocellular carcinoma in the Asia-Pacific region: burden, trends, challenges and future directions. Nat Rev Gastroenterol Hepatol 2024; 21:834-851. [PMID: 39147893 DOI: 10.1038/s41575-024-00967-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Accepted: 07/10/2024] [Indexed: 08/17/2024]
Abstract
Globally, nearly half of deaths from cirrhosis and chronic liver diseases (CLD) and three-quarters of deaths from hepatocellular carcinoma (HCC) occur in the Asia-Pacific region. Chronic hepatitis B is responsible for the vast majority of liver-related deaths in the region. Metabolic dysfunction-associated steatotic liver disease (MASLD) is the most common form of CLD, affecting an estimated 30% of the adult population. Compared with people of European descent, people from the Asia-Pacific region carry more genetic variants associated with MASLD and its progression. Alcohol is a fast-growing cause of CLD and HCC in Asia as a result of the rising per-capita consumption of alcohol. Drug-induced liver injury is under-recognized and probably has a high prevalence in this region. The epidemiological and outcome data of acute-on-chronic liver failure are heterogeneous, and non-unified definitions across regions contribute to this heterogeneity. CLDs are severely underdiagnosed, and effective treatments and vaccinations are underutilized. In this Review, we highlight trends in the burden of CLD and HCC in the Asia-Pacific region and discuss the rapidly changing aetiologies of liver disease. We examine the multiple gaps in the care cascade and propose mitigating strategies and future directions.
Collapse
Affiliation(s)
- Lung-Yi Mak
- The University of Hong Kong, Hong Kong, China
| | - Ken Liu
- The University of Sydney, Sydney, Australia
| | | | | | | | | | | | - Rashid Lui
- The Chinese University of Hong Kong, Hong Kong, China
| | - Hye Won Lee
- Yonsei University College of Medicine, Seoul, Korea
| | | | - Anand V Kulkarni
- Department of Hepatology, Asian Institute of Gastroenterology, Hyderabad, India
| | - Madhumita Premkumar
- Department of Hepatology, Postgraduate Institute of Medical Education and Research, Chandigarh, India
| | | | - Yao Chun Hsu
- Department of Medical Research, E-Da Hospital, Kaohsiung, Taiwan; School of Medicine and Graduate Institute of Medicine, I-Shou University, Kaohsiung, Taiwan
- School of Medicine and Graduate Institute of Medicine, I-Shou University, Kaohsiung, Taiwan
| | - Daniel Q Huang
- Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore.
- Division of Gastroenterology and Hepatology, National University Hospital, Singapore, Singapore.
| |
Collapse
|
4
|
Soliman R, Lok J, Ajaz S, Agarwal K, Guerra M. Predictive performance of hepatocellular carcinoma risk scores in chronic hepatitis C patients with advanced fibrosis after achieving sustained virological response: Insights from European Association for the study of the Liver Policy recommendations. Eur J Intern Med 2024; 129:155-157. [PMID: 39237432 DOI: 10.1016/j.ejim.2024.08.018] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 05/28/2024] [Revised: 08/13/2024] [Accepted: 08/16/2024] [Indexed: 09/07/2024]
Affiliation(s)
- Riham Soliman
- Institute of Liver Studies. King´s College Hospital, NHS Trust, London, England.
| | - James Lok
- Institute of Liver Studies. King´s College Hospital, NHS Trust, London, England
| | - Saima Ajaz
- Institute of Liver Studies. King´s College Hospital, NHS Trust, London, England
| | - Kosh Agarwal
- Institute of Liver Studies. King´s College Hospital, NHS Trust, London, England
| | - María Guerra
- Institute of Liver Studies. King´s College Hospital, NHS Trust, London, England
| |
Collapse
|
5
|
Tavabie OD, Salehi S, Aluvihare VR. The challenges and potential in developing microRNA associated with regeneration as biomarkers to improve prognostication for liver failure syndromes and hepatocellular carcinoma. Expert Rev Mol Diagn 2024; 24:5-22. [PMID: 38059597 DOI: 10.1080/14737159.2023.2292642] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/16/2023] [Accepted: 12/05/2023] [Indexed: 12/08/2023]
Abstract
INTRODUCTION Determining the need for liver transplantation remains critical in the management of hepatocellular carcinoma (HCC) and liver failure syndromes (including acute liver failure and decompensated cirrhosis states). Conventional prognostic models utilize biomarkers of liver and non-liver failure and have limitations in their application. Novel biomarkers which predict regeneration may fulfil this niche. microRNA are implicated in health and disease and are present in abundance in the circulation. Despite this, they have not translated into mainstream clinical biomarkers. AREAS COVERED We will discuss current challenges in the prognostication of patients with liver failure syndromes as well as for patients with HCC. We will discuss biomarkers implicated with liver regeneration. We then provide an overview of the challenges in developing microRNA into clinically tractable biomarkers. Finally, we will provide a scoping review of microRNA which may have potential as prognostic biomarkers in liver failure syndromes and HCC. EXPERT OPINION Novel biomarkers are needed to improve prognostic models in liver failure syndromes and HCC. Biomarkers associated with liver regeneration are currently lacking and may fulfil this niche. microRNA have the potential to be developed into clinically tractable biomarkers but a consensus on standardizing methodology and reporting is required prior to large-scale studies.
Collapse
Affiliation(s)
| | - Siamak Salehi
- Institute of Liver Studies, King's College Hospital, London, UK
| | | |
Collapse
|
6
|
Singal AG, Parikh ND, Hoshida Y. Precision HCC Surveillance: It Is All in the Number (Needed). Dig Dis Sci 2023; 68:720-722. [PMID: 36302995 DOI: 10.1007/s10620-022-07735-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Accepted: 09/27/2022] [Indexed: 12/09/2022]
Affiliation(s)
- Amit G Singal
- Division of Digestive and Liver Diseases, UT Southwestern Medical Center, 5959 Harry Hines Blvd, POB 1, Suite 420, Dallas, TX, 75390-8887, USA.
- Division of Gastroenterology and Hepatology, University of Michigan, Ann Arbor, MI, 48320, USA.
| | - Neehar D Parikh
- Division of Digestive and Liver Diseases, UT Southwestern Medical Center, 5959 Harry Hines Blvd, POB 1, Suite 420, Dallas, TX, 75390-8887, USA
- Division of Gastroenterology and Hepatology, University of Michigan, Ann Arbor, MI, 48320, USA
| | - Yujin Hoshida
- Division of Digestive and Liver Diseases, UT Southwestern Medical Center, 5959 Harry Hines Blvd, POB 1, Suite 420, Dallas, TX, 75390-8887, USA
- Division of Gastroenterology and Hepatology, University of Michigan, Ann Arbor, MI, 48320, USA
| |
Collapse
|