Selection of neural network architecture and data augmentation procedures for predicting the course of cardiovascular diseases

Abstract

The article solves the problem of creating models for predicting the course and complications of cardiovascular diseases. Artificial neural networks based on the Keras library are used. The original dataset includes 1700 case histories. In addition, the dataset augmentation procedure was used. As a result, the overall accuracy exceeded 84%. Furthermore, optimizing the network architecture and dataset has increased the overall accuracy by 17% and precision by 7%.

Predicting Myocardial Infarction Complications and Outcomes With Deep Learning

“…The model optimized with adaptive moment estimation (Adam) algorithm, batch normalization, and dropout layers are also used against overfitting problem and improve performance. The proposed model in this study has better performance than the state-of-the-art literature approaches [19,20,22,24].…”

“…In the second scenario, data was augmented as proposed in [22]. The main goal of the data augmentation was the augment the positive outcomes.…”

“…Study [22] was proposed by the same authors as study [24]. [22] was presented as a further study of [24] to predict more positive outcomes.…”

“…Study [22] was proposed by the same authors as study [24]. [22] was presented as a further study of [24] to predict more positive outcomes. Dorrer and friends aimed to predict the course of CVD with data augmentation.…”

“…Due to the scarcity of positive output data, the authors proposed a data augmentation method and performed their testing on various deep neural network models. The authors had 72.14% total accuracy score from their final proposed model [22].…”