Hypertrophic Cardiomyopathy Diagnosis Based on Cardiovascular Magnetic Resonance Using Deep Learning Techniques

Background: Hypertrophic cardiomyopathy (HCM) can lead to serious cardiac problems. HCM is often diagnosed by an expert using cardiovascular magnetic resonance (CMR) images obtained from patients. In this research, we aimed to develop a deep learning technique to automate HCM diagnosis. Methods: CMR images of 37421 healthy and 21846 HCM patients (53% females, mean age 48.2±19.5 years) were obtained for two years. Three experts inspected the images to determine the presence of HCM. New data augmentation method was used to generate new images by employing colour filtering on the existing ones. To classify the augmented images, we used a deep convolutional neural network (CNN). To the best of our knowledge, this is the first time CNN is used for HCM diagnosis. We designed our CNN from scratch to reach acceptable diagnosis accuracy. Findings: The designed algorithms achieved an accuracy of 95.23%, recall of 97.90%, and specificity of 93.06% on the original dataset compared with expert opinions. The same performance metrics for the designed algorithm on the augmented dataset were 98 . 53%, 98.70%, and 95.21%, respectively. We experimented with different optimizers (e.g. Adadelta and Adagrad) and other data augmentation methods (e.g. height shift and rotation) to further evaluate the proposed method. Using our data augmentation method, 98.53% accuracy was achieved, which is higher than the best accuracy (95.83%) obtained by the other previous data augmentation methods.   Interpretation: Our machine learning method could achieve high performance for HCM diagnosis. The advantages of employing the proposed method are eliminating contrast agent and their complications, decreased CMR examination time, lower costs for patients and cardiac imaging centres. Funding Information: None. Declaration of Interests: The authors declare that there is no conflict of interest. Ethics Approval Statement: Ethical approval for this study was obtained by the Omid Hospital Ethics Committee.