Contrast-Enhanced CT-Based Deep Learning Radiomics Nomogram for the Survival Prediction in Gallbladder Cancer

Area of Science:

• Oncology
• Medical Imaging
• Artificial Intelligence

Background:

• Accurate prognostic models are crucial for developing effective treatment strategies in gallbladder cancer (GBC).
• Predicting patient survival post-surgical resection is vital for individualized patient management.

Purpose of the Study:

• To develop and validate an integrated prognostic model for predicting survival in gallbladder cancer (GBC) patients after surgical resection.
• To combine clinical features, radiomics, and deep learning from computed tomography (CT) images for enhanced survival prediction.

Main Methods:

• A retrospective study of 167 GBC patients who underwent surgical resection.
• Extraction of handcrafted radiomics features and application of a DenseNet121 model for deep learning signature development.
• Integration of clinical, radiomics, and deep learning signatures using logistic regression to form a unified multimodal model.

Main Results:

• The integrated multimodal model achieved an area under the curve (AUC) of 0.870 and a concordance index (C-index) of 0.736 in the test set.
• Kaplan-Meier analysis confirmed significantly lower survival probability in the high-risk group compared to the low-risk group (log-rank p < 0.05).

Conclusions:

• The developed nomogram effectively predicts survival for GBC patients post-surgery.
• The model aids in identifying high-risk patients, facilitating personalized management and improving outcomes for gallbladder cancer.