JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Prediction of Local Tumor Progression After Thermal Ablation of Colorectal Cancer Liver Metastases Based on Magnetic Resonance Imaging Δ-Radiomics

  • 0Department of medical college, School of Medicine, Shaoxing University, Shaoxing.
Journal of Computer Assisted Tomography +

|

December 4, 2024

|

December 4, 2024

View abstract on PubMed

Summary

This summary is machine-generated.

Predicting local tumor progression after thermal ablation for colorectal cancer liver metastases is improved by combining clinical data with MRI-based radiomics. This approach enhances prognostic accuracy for patients undergoing ablation therapy.

Area Of Science

  • Medical Imaging
  • Oncology
  • Radiology

Background

  • Colorectal cancer liver metastases (CRLMs) are a common clinical challenge.
  • Thermal ablation is a key treatment modality for CRLMs.
  • Predicting local tumor progression (LTP) post-ablation is crucial for patient management.

Purpose Of The Study

  • To enhance the predictability of local tumor progression (LTP) after thermal ablation in patients with CRLMs.
  • To integrate magnetic resonance imaging (MRI) Δ-radiomics and clinical features for improved LTP prediction.
  • To develop and validate predictive models for LTP following thermal ablation.

Main Methods

  • Retrospective analysis of 37 patients with CRLM (57 tumors).
  • Calculation of Δ-radiomics by subtracting pre- and post-treatment radiomics features.
  • Development of preoperative lesion, postoperative ablation area, and Δ models using LASSO and logistic regression.
  • Creation of a composite model combining clinical features and Δ-radiomics for LTP prediction.

Main Results

  • Local tumor progression (LTP) occurred in 35% of lesions.
  • Clinical model identified tumor size and ΔCEA as significant LTP risk factors.
  • The Δ model showed high AUC values (e.g., 0.875 in testing Delay AUC).
  • The combined model achieved optimal performance (e.g., 0.917 in testing Delay AUC).

Conclusions

  • Combined models integrating clinical data and Δ-radiomics are valuable for predicting LTP post-thermal ablation in CRLM patients.
  • This approach offers improved prognostic utility for managing CRLMs treated with thermal ablation.

Keywords: