Predicting Treatment Response to Transcatheter Arterial Chemoembolization in Hepatocellular Carcinoma Patients using...

Area of Science:

• Radiology
• Artificial Intelligence
• Oncology

Background:

• Hepatocellular carcinoma (HCC) is a primary liver cancer.
• Transarterial chemoembolization (TACE) is a standard treatment for HCC.
• Predicting patient response to TACE is crucial for effective treatment planning.

Purpose of the Study:

• To assess a deep learning model's effectiveness in predicting early response to TACE in HCC patients.
• To evaluate the precision of a novel LeNet-based deep learning model with an attention mechanism.

Main Methods:

• Retrospective analysis of 111 HCC patients undergoing TACE.
• Utilized pre- and post-TACE MRI scans.
• Developed a deep learning model (LeNet with attention) for response prediction.
• Validated using ROC curves and confusion matrices based on mRECIST criteria.

Main Results:

• The deep learning model achieved an AUC of 0.760 (training) and 0.729 (test).
• Prediction accuracy was 70.7% (training) and 72.3% (test).
• 50.5% of patients showed an objective response to TACE.

Conclusions:

• Deep learning models using MRI data can effectively forecast HCC patient response to TACE.
• The developed LeNet model with attention mechanism offers valuable insights for treatment strategy formulation.