An Integrated Model Combined Conventional Radiomics and Deep Learning Features to Predict Early Recurrence of Hepatocellular Carcinoma Eligible for Curative Ablation: A Multicenter Cohort Study.
IF 1 4区 医学Q4 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING
Yong-Hai Li, Gui-Xiang Qian, Yu Zhu, Xue-di Lei, Lei Tang, Xiang-Yi Bu, Ming-Tong Wei, Wei-Dong Jia
{"title":"An Integrated Model Combined Conventional Radiomics and Deep Learning Features to Predict Early Recurrence of Hepatocellular Carcinoma Eligible for Curative Ablation: A Multicenter Cohort Study.","authors":"Yong-Hai Li, Gui-Xiang Qian, Yu Zhu, Xue-di Lei, Lei Tang, Xiang-Yi Bu, Ming-Tong Wei, Wei-Dong Jia","doi":"10.1097/RCT.0000000000001764","DOIUrl":null,"url":null,"abstract":"<p><strong>Objective: </strong>Hepatocellular carcinoma (HCC) is the most common primary liver malignancy. Ablation therapy is one of the first-line treatments for early HCC. Accurately predicting early recurrence (ER) is crucial for making precise treatment plans and improving prognosis. This study aimed to develop and validate a model (DLRR) that incorporates deep learning radiomics and traditional radiomics features to predict ER following curative ablation for HCC.</p><p><strong>Methods: </strong>We retrospectively analysed the data of 288 eligible patients from 3 hospitals-1 primary cohort (center 1, n=222) and 2 external test cohorts (center 2, n=32 and center 3, n=34)-from April 2008 to March 2022. 3D ResNet-18 and PyRadiomics were applied to extract features from contrast-enhanced computed tomography (CECT) images. The 3-step (ICC-LASSO-RFE) method was used for feature selection, and 6 machine learning methods were used to construct models. Performance was compared through the area under the receiver operating characteristic curve (AUC), net reclassification improvement (NRI) and integrated discrimination improvement (IDI) indices. Calibration and clinical applicability were assessed through calibration curves and decision curve analysis (DCA), respectively. Kaplan-Meier (K-M) curves were generated to stratify patients based on progression-free survival (PFS) and overall survival (OS).</p><p><strong>Results: </strong>The DLRR model had the best performance, with AUCs of 0.981, 0.910, and 0.851 in the training, internal validation, and external validation sets, respectively. In addition, the calibration curve and DCA curve revealed that the DLRR model had good calibration ability and clinical applicability. The K-M curve indicated that the DLRR model provided risk stratification for progression-free survival (PFS) and overall survival (OS) in HCC patients.</p><p><strong>Conclusions: </strong>The DLRR model noninvasively and efficiently predicts ER after curative ablation in HCC patients, which helps to categorize the risk in patients to formulate precise diagnosis and treatment plans and management strategies for patients and to improve the prognosis.</p>","PeriodicalId":15402,"journal":{"name":"Journal of Computer Assisted Tomography","volume":" ","pages":""},"PeriodicalIF":1.0000,"publicationDate":"2025-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Computer Assisted Tomography","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1097/RCT.0000000000001764","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}
引用次数: 0
Abstract
Objective: Hepatocellular carcinoma (HCC) is the most common primary liver malignancy. Ablation therapy is one of the first-line treatments for early HCC. Accurately predicting early recurrence (ER) is crucial for making precise treatment plans and improving prognosis. This study aimed to develop and validate a model (DLRR) that incorporates deep learning radiomics and traditional radiomics features to predict ER following curative ablation for HCC.
Methods: We retrospectively analysed the data of 288 eligible patients from 3 hospitals-1 primary cohort (center 1, n=222) and 2 external test cohorts (center 2, n=32 and center 3, n=34)-from April 2008 to March 2022. 3D ResNet-18 and PyRadiomics were applied to extract features from contrast-enhanced computed tomography (CECT) images. The 3-step (ICC-LASSO-RFE) method was used for feature selection, and 6 machine learning methods were used to construct models. Performance was compared through the area under the receiver operating characteristic curve (AUC), net reclassification improvement (NRI) and integrated discrimination improvement (IDI) indices. Calibration and clinical applicability were assessed through calibration curves and decision curve analysis (DCA), respectively. Kaplan-Meier (K-M) curves were generated to stratify patients based on progression-free survival (PFS) and overall survival (OS).
Results: The DLRR model had the best performance, with AUCs of 0.981, 0.910, and 0.851 in the training, internal validation, and external validation sets, respectively. In addition, the calibration curve and DCA curve revealed that the DLRR model had good calibration ability and clinical applicability. The K-M curve indicated that the DLRR model provided risk stratification for progression-free survival (PFS) and overall survival (OS) in HCC patients.
Conclusions: The DLRR model noninvasively and efficiently predicts ER after curative ablation in HCC patients, which helps to categorize the risk in patients to formulate precise diagnosis and treatment plans and management strategies for patients and to improve the prognosis.
期刊介绍:
The mission of Journal of Computer Assisted Tomography is to showcase the latest clinical and research developments in CT, MR, and closely related diagnostic techniques. We encourage submission of both original research and review articles that have immediate or promissory clinical applications. Topics of special interest include: 1) functional MR and CT of the brain and body; 2) advanced/innovative MRI techniques (diffusion, perfusion, rapid scanning); and 3) advanced/innovative CT techniques (perfusion, multi-energy, dose-reduction, and processing).