JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Combined radiomics-clinical model to predict platinum-sensitivity in advanced high-grade serous ovarian carcinoma using multimodal MRI

  • 0Department of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, Republic of Korea.
Frontiers in Oncology +

|

February 8, 2024

|

February 8, 2024

View abstract on PubMed

Summary

This summary is machine-generated.

Predicting platinum sensitivity in ovarian cancer is improved by combining radiomics features from magnetic resonance imaging (MRI) with clinical data. This integrated approach enhances prediction accuracy for advanced ovarian cancer patients.

Area Of Science

  • Oncology
  • Radiology
  • Medical Imaging

Background

  • Platinum-based chemotherapy is a cornerstone treatment for ovarian cancer.
  • Predicting platinum sensitivity is crucial for optimizing treatment strategies.
  • Radiomics offers a non-invasive method to extract quantitative features from medical images.

Purpose Of The Study

  • To predict platinum sensitivity in ovarian cancer patients using a radiomics framework.
  • To evaluate the efficacy of a combined radiomics and clinical data model.

Main Methods

  • 96 ovarian cancer patients underwent multimodal MRI (diffusion-weighted, T1-weighted, T2-weighted).
  • 293 radiomic features were extracted and selected using the least absolute shrinkage and selection operators.
  • Two prediction models were developed: one with radiomics features and another combining radiomics with clinical factors (age, stage, tumor marker, treatment course).
  • Models were validated using five-fold cross-validation.

Main Results

  • The radiomics-only model achieved an AUC of 0.65 for predicting platinum sensitivity.
  • The combined radiomics-clinical model demonstrated improved performance with an AUC of 0.77.
  • Radiomic features related to tumor heterogeneity were key predictors.

Conclusions

  • A combined radiomics and clinical data model effectively predicts platinum sensitivity in advanced ovarian cancer.
  • Integrating radiomics enhances the predictive power beyond traditional clinical factors.

Keywords: