Multimodal data integration for predicting progression risk in castration-resistant prostate cancer using deep
Chuan Zhou1,2, Yun-Feng Zhang3, Sheng Guo3
1The First Clinical Medical College of Lanzhou University, Lanzhou, China.
Frontiers in Oncology
|March 29, 2024
Summary
This study developed an AI model integrating imaging and pathology data to predict prostate cancer progression to castration-resistant prostate cancer. The combined model significantly improved prediction accuracy, aiding patient management.
Area of Science:
- Oncology
- Medical Imaging
- Artificial Intelligence
Background:
- Advanced prostate cancer (PCa) frequently progresses to castration-resistant PCa (CRPC), a condition with a poor prognosis.
- Multiparametric magnetic resonance imaging (mpMRI) and histopathology offer valuable prognostic information.
- Artificial intelligence (AI) can integrate multimodal data for enhanced prognostic capabilities.
Purpose of the Study:
- To construct an AI-based model for predicting CRPC progression.
- To integrate multimodal data from mpMRI and histopathology for improved prediction.
- To develop a tool to guide patient prognosis and management strategies.
Main Methods:
- Retrospective analysis of data from 399 PCa patients.
- Delineation of regions of interest (ROIs) from T2WI, DWI, and ADC MRI sequences.
- Deep learning model training using pathological H&E slides and radiomic features.
- Construction of a joint combined model nomogram, assessed with ROC, calibration, and decision curve analysis.
Main Results:
- The combined AI model achieved an Area Under the Curve (AUC) of 0.86.
- Deep learning models (ResNet-50) showed strong performance for radiomics (AUC 0.768) and pathomics (AUC 0.752).
- The combined model demonstrated good calibration and clinical net benefit, with improved prediction of PCa to CRPC progression.
Conclusions:
- Integrating multimodal data, including mpMRI and histopathology, significantly enhances the prediction of prostate cancer progression to CRPC.
- The developed AI model offers a valuable tool for guiding patient prognosis and management decisions.
- AI-driven multimodal data integration represents a promising approach for improving outcomes in advanced prostate cancer.
Related Concept Videos
02:43Mouse Models of Cancer Study
4.7K
Mice have long served as models for studying human biology and pathology because of their phylogenetic and physiological similarity with humans. They are also easy to maintain and breed in the laboratory, and hence, many inbred strains are now available for research. Studies on mice have contributed immeasurably to our understanding of cancer biology.
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
4.7K
01:21Cancer Survival Analysis
863
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
863