Construction of an Artificially Intelligent Model for Accurate Detection of HCC by Integrating Clinical, Radiological, and Peripheral Immunological Features.
{"title":"Construction of an Artificially Intelligent Model for Accurate Detection of HCC by Integrating Clinical, Radiological, and Peripheral Immunological Features.","authors":"Yangyang Wang, Shengqiang Chi, Yu Tian, Xueyao Li, Hang Zhang, Yiting Xu, Chao Huang, Yiwei Gao, Gaowei Jin, Qihan Fu, Wanyue Cao, Cao Chen, Haonan Ding, Yuquan Zhang, Yupeng Hong, Junjian Li, Xu Sun, Enliang Li, Yuhua Zhang, Weiyun Yao, Runtian Liu, Yongfei Hua, Haifeng Huang, Minghui Xu, Bo Zhang, Weifeng Tao, Tianxing Yang, Yuming Gao, Xiaoguang Wang, Cheng Lin, Jingsong Li, Qi Zhang, Tingbo Liang","doi":"10.1097/JS9.0000000000002281","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Integrating comprehensive information on hepatocellular carcinoma (HCC) is essential to improve its early detection. We aimed to develop a model with multi-modal features (MMF) using artificial intelligence (AI) approaches to enhance the performance of HCC detection.</p><p><strong>Materials and methods: </strong>A total of 1,092 participants were enrolled from 16 centers. These participants were allocated into the training, internal validation, and external validation cohorts. Peripheral blood specimens were collected prospectively and subjected to mass cytometry analysis. Clinical and radiological data were obtained from electrical medical records. Various AI methods were employed to identify pertinent features and construct single-modal models with optimal performance. The XGBoost algorithm was utilized to amalgamate these models, integrating multi-modal information and facilitating the development of a fusion model. Model evaluation and interpretability were demonstrated using the SHapley Additive exPlanations method.</p><p><strong>Results: </strong>We constructed the electronic health record, BioScore, RadiomicScore, and DLScore models based on clinical, radiological, and peripheral immunological features, respectively. Subsequently, these single-modal models were amalgamated to develop an all-in-one MMF model. The MMF model exhibited enhanced performance compared to models comprising only single-modal features in detecting HCC. This superiority in performance was confirmed through the internal and external validation cohorts, yielding area under the curve (AUC) values of 0.985 and 0.915, respectively. Additionally, the MMF model improved the detection ability in subpopulations of HCCs that were negative for alpha-fetoprotein and those with small size, with AUC values of 0.974 and 0.996, respectively.</p><p><strong>Conclusions: </strong>Integrating multi-modal features improved the performance of the model for HCC detection.</p>","PeriodicalId":14401,"journal":{"name":"International journal of surgery","volume":" ","pages":""},"PeriodicalIF":12.5000,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International journal of surgery","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1097/JS9.0000000000002281","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"SURGERY","Score":null,"Total":0}
引用次数: 0
Abstract
Background: Integrating comprehensive information on hepatocellular carcinoma (HCC) is essential to improve its early detection. We aimed to develop a model with multi-modal features (MMF) using artificial intelligence (AI) approaches to enhance the performance of HCC detection.
Materials and methods: A total of 1,092 participants were enrolled from 16 centers. These participants were allocated into the training, internal validation, and external validation cohorts. Peripheral blood specimens were collected prospectively and subjected to mass cytometry analysis. Clinical and radiological data were obtained from electrical medical records. Various AI methods were employed to identify pertinent features and construct single-modal models with optimal performance. The XGBoost algorithm was utilized to amalgamate these models, integrating multi-modal information and facilitating the development of a fusion model. Model evaluation and interpretability were demonstrated using the SHapley Additive exPlanations method.
Results: We constructed the electronic health record, BioScore, RadiomicScore, and DLScore models based on clinical, radiological, and peripheral immunological features, respectively. Subsequently, these single-modal models were amalgamated to develop an all-in-one MMF model. The MMF model exhibited enhanced performance compared to models comprising only single-modal features in detecting HCC. This superiority in performance was confirmed through the internal and external validation cohorts, yielding area under the curve (AUC) values of 0.985 and 0.915, respectively. Additionally, the MMF model improved the detection ability in subpopulations of HCCs that were negative for alpha-fetoprotein and those with small size, with AUC values of 0.974 and 0.996, respectively.
Conclusions: Integrating multi-modal features improved the performance of the model for HCC detection.
期刊介绍：
The International Journal of Surgery (IJS) has a broad scope, encompassing all surgical specialties. Its primary objective is to facilitate the exchange of crucial ideas and lines of thought between and across these specialties.By doing so, the journal aims to counter the growing trend of increasing sub-specialization, which can result in "tunnel-vision" and the isolation of significant surgical advancements within specific specialties.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
