Construction of a predictive model for hepatocellular carcinoma recurrence after radical resection based on MSCT features(PDF)
《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]
- Issue:
- 2026年第2期
- Page:
- 261-267
- Research Field:
- 医学人工智能
- Publishing date:
Info
- Title:
- Construction of a predictive model for hepatocellular carcinoma recurrence after radical resection based on MSCT features
- Author(s):
-
LIU Yangjun; LIU Shuzhen; LI Peng; LI Jinan
- Department of General Surgery, the First Affiliated Hospital of Jinzhou Medical University, Jinzhou 121000, China
- Keywords:
- Keywords: hepatocellular carcinoma radical resection recurrence multi-slice spiral computed tomography predictive model
- PACS:
- R318；R735.7
- DOI:
- DOI:10.3969/j.issn.1005-202X.2026.02.017
- Abstract:
- Objective To identify the multi-slice spiral computed tomography (MSCT) features associated with recurrence after radical resection of hepatocellular carcinoma (HCC), and to construct a predictive model. Methods MSCT images of HCC patients who underwent radical resection at the First Affiliated Hospital of Jinzhou Medical University between January 2022 and April 2024 were collected retrospectively. According to the occurrence of postoperative recurrence, the patients were divided into the recurrence group (n=54) and the non-recurrence group (n=218), and compared for tumor morphological characteristics, hemodynamic features, and invasion features. Multivariate Logistic stepwise regression analysis was used to screen for independent risk factors of postoperative HCC recurrence, and a predictive model was constructed, whose predictive efficiency was evaluated using a nomogram, receiver operating characteristic (ROC) curve, and calibration curve. Results Compared with the non-recurrence group, the recurrence group exhibited larger maximum tumor diameter and tumor volume, higher incidences of lobulation sign, spiculation sign, and vacuole sign, as well as greater arterial-phase CT values and arterial phase-plain scan enhancement difference (△CT values) (all P<0.05). Multivariate Logistic analysis identified the maximum tumor diameter, tumor volume, arterial-phase CT values, and [Δ]CT values as independent predictors of HCC recurrence after radical resection (P<0.05). A nomogram was generated based on the constructed predictive model. ROC curve analysis showed that the area under the curve (AUC) of the predictive model was 0.860, indicating high predictive value. The calibration curve revealed a high degree of consistency between the predicted probability of the model and the actual observation results, confirming its excellent calibration and practicality. Conclusion Maximum tumor diameter, tumor volume, arterial-phase CT values, and △CT values are independent risk factors for HCC recurrence after radical resection. The predictive model constructed on above basis exhibits high predictive efficacy, and is recommended for clinical application.
Last Update: 2026-01-27