Related Experiment Video
Updated: May 9, 2025

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
Preoperative prediction of early recurrence in hepatocellular carcinoma using simultaneous multislice diffusion
Yingyi Wu1, Zheng Ye1, Ting Yang1
1Department of Radiology, West China Hospital, Sichuan University, Chengdu, China.
Objective:
This study aimed to evaluate the value of simultaneous multislice diffusion kurtosis imaging (SMS-DKI) for predicting early recurrence (within 2 years) in hepatocellular carcinoma (HCC) and to develop a predictive model.
Materials And Methods:
This prospective study included 67 HCC patients who underwent SMS-DKI on a 3-T MRI between June 2021 and January 2023. Diffusion parameters, including the apparent diffusion coefficient (ADC), SMS-mean kurtosis (SMS-MK), and SMS-mean diffusivity (SMS-MD), along with radiological features, were analyzed. Logistic regression models were used to predict early recurrence, internally validated using 10-fold cross-validation, and assessed using AUC, calibration curves, and decision curve analysis (DCA).
Results:
Among 67 patients (58 males; mean age, 53.5 ± 9.9 years), 30 (44.8%) experienced early recurrence. The early recurrence had significantly lower ADC (1.12 vs 1.22 × 10-3 mm2/s) and SMS-MD (1.45 vs 1.70 × 10-3 mm2/s), and higher SMS-MK (0.91 vs 0.75). SMS-MK showed the highest AUC (0.90, 95% CI: 0.80-0.96). Multivariate analysis identified SMS-MK (OR = 3.43 [1.31-8.89]), tumor size (OR = 4.22 [1.58-7.76]), non-smooth tumor margin (OR = 2.68 [1.58-7.96]), and complete capsule (OR = 0.22 [0.02-0.79]) as independent predictors of early recurrence. Based on these four parameters, the final model achieved an AUC of 0.94 (95% CI: 0.88-1.00). Calibration curves and DCA confirmed clinical utility.
Conclusion:
SMS-DKI enhances early recurrence prediction in HCC. The predictive model, incorporating SMS-MK, tumor size, and key radiological features, demonstrated good prognostic value.
Key Points:
Question Can SMS-DKI predict HCC early recurrence within 2 years post-surgery? Findings Higher SMS-MK, larger tumor size, non-smooth margins, and incomplete capsule predict HCC early recurrence (model AUC = 0.94). Clinical relevance Integrating preoperative SMS-DKI biomarkers (SMS-MK) with tumor size and capsule status stratifies early HCC recurrence risk, guiding surgical planning and postoperative management.
More Related Videos
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018
07:32Author Spotlight: Investigating Immune Cell Dynamics in the Tumor Microenvironment — Challenges and Innovations in Cancer Prognosis
Published on: April 12, 2024