TPNET: A time-sensitive small sample multimodal network for cardiotoxicity risk prediction.

in IEEE journal of biomedical and health informatics by Yuan He, Fengyun Zhang, Kaimiao Hu, Changming Sun, Jie Geng, Ning Ren, Ran Su

TLDR

  • The study developed a deep learning model that uses tissue Doppler imaging and clinical data to predict cancer therapy-related cardiac dysfunction (CTRCD) in breast cancer patients, achieving an accuracy of 83%. The model has the potential to identify unseen pathogenic signs and may lead to new diagnostic and therapeutic strategies.

Abstract

Cancer therapy-related cardiac dysfunction (CTRCD) is a potential complication associated with cancer treatment, particularly in patients with breast cancer, requiring monitoring of cardiac health during the treatment process. Tissue Doppler imaging (TDI) is a remarkable technique that can provide a comprehensive reflection of the left ventricle's physiological status. We hypothesized that the combination of TDI features with deep learning techniques could be utilized to predict CTRCD. To evaluate the hypothesis, we developed a temporal-multimodal pattern network for efficient training (TPNET) model to predict the incidence of CTRCD over a 24-month period based on TDI, function, and clinical data from 270 patients. Our model achieved an area under curve (AUC) of 0.83 and sensitivity of 0.88, demonstrating greater robustness compared to other existing visual models. To further translate our model's findings into practical applications, we utilized the integrated gradients (IG) attribution to perform a detailed evaluation of all the features. This analysis has identified key pathogenic signs that may have remained unnoticed, providing a viable option for implementing our model in preoperative breast cancer patients. Additionally, our findings demonstrate the potential of TPNET in discovering new causative agents for CTRCD.

Overview

  • The study aimed to develop a model that could predict cancer therapy-related cardiac dysfunction (CTRCD) using tissue Doppler imaging (TDI) features and deep learning techniques.
  • The study used a temporal-multimodal pattern network for efficient training (TPNET) model to predict CTRCD in 270 patients over a 24-month period based on TDI, function, and clinical data.
  • The primary objective of the study is to identify a robust model that can effectively predict CTRCD in patients undergoing cancer treatment, particularly in those with breast cancer.

Comparative Analysis & Findings

  • The TPNET model achieved an area under curve (AUC) of 0.83 and sensitivity of 0.88, outperforming existing visual models in predicting CTRCD.
  • The study found that the TPNET model was able to identify key pathogenic signs that may have remained unnoticed, providing a viable option for implementing the model in preoperative breast cancer patients.
  • The findings of the study demonstrate the potential of TPNET in discovering new causative agents for CTRCD, highlighting the importance of using multimodal data and deep learning techniques in cancer research.

Implications and Future Directions

  • The study highlights the importance of monitoring cardiac health in patients undergoing cancer treatment, particularly in those with breast cancer, to identify potential complications early on.
  • Future studies should focus on validating the TPNET model in larger and more diverse cohorts of patients, as well as exploring the application of the model in other cancer types.
  • The use of deep learning techniques and multimodal data has the potential to improve our understanding of cancer-related cardiac dysfunction and may lead to the development of new diagnostic and therapeutic strategies.
Read Full Article