Key result
Deep learning model DIONCO predicts 5-year MACE risk in oncology patients.
Why the study?
The study aimed to develop an artificial intelligence application to automatically predict cardiovascular risk and major cardiovascular events in oncologic patients receiving specific treatments.
A deep learning MultiLayer Perceptron network (DIONCO) is being developed and trained to automatically predict the 5-year risk of cardiac dysfunction in oncology patients.
Requires external validation before any clinical use; leaves open deep learning for cardio-oncology risk prediction.
In this research paper we utilized MultiLayer Perceptron Networks to automaticaly assess cardiac risk calculation in oncologic patients receiving different types of specific treatments. An artificial intelligence/deep learning application was developed to calculate the risk of cardiac dysfunction (low, moderate, high) based on several clinical and imaging parameters. The main goal was to help cardiologists in their practice by automatically predicting the cardiovascular risk for a given oncological patient, in other words to determine which patient will have a major cardiovascular event in the near future (the next 5 years), such as sudden death or heart failure, therefore, doctors will have to prescribe a more aggressive medical treatment. We proposed, for supervised learning, a deep learning MLP network with 4 layers sequential deep neural network model, which has multiple inputs and three outputs, because our model must predict the overall cardiac risk for the patient. The enrollment of the 50 patients from the total batch of patients who are loaded into the Artificial Intelligence / Deep learning application to determine the risk of cardiac dysfunction automatically using the artificial intelligence application has been completed. For this, we started training the DIONCO neural network to calculating the accuracy of the final results and calculating the risk score for a new patient. Also in this report, we decided the implementation of a new improved network for calculating cardiac risk in oncological patients.
No takes yet. Share an insight, caveat, or question.
Popa et al. (2025) studied Cardiac dysfunction in oncology patients (n=50). MultiLayer Perceptron Network (DIONCO) was evaluated on Risk of cardiac dysfunction (low, moderate, high). A deep learning MultiLayer Perceptron network (DIONCO) was developed and trained on 50 oncology patients to automatically predict the 5-year risk of major cardiovascular events.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: