RF-CNN-F: random forest with convolutional neural network features for coronary artery disease diagnosis based on cardiac magnetic resonance

Abstract

Coronary artery disease (CAD) is a prevalent disease with high morbidity and mortality rates. Invasive coronary angiography is the reference standard for diagnosing CAD but is costly and associated with risks. Noninvasive imaging like cardiac magnetic resonance (CMR) facilitates CAD assessment and can serve as a gatekeeper to downstream invasive testing. Machine learning methods are increasingly applied for automated interpretation of imaging and other clinical results for medical diagnosis. In this study, we proposed a novel CAD detection method based on CMR images by utilizing the feature extraction ability of deep neural networks and combining the features with the aid of a random forest for the very first time. It is necessary to convert image data to numeric features so that they can be used in the nodes of the decision trees. To this end, the predictions of multiple stand-alone convolutional neural networks (CNNs) were considered as input features for the decision trees. The capability of CNNs in representing image data renders our method a generic classification approach applicable to any image dataset. We named our method RF-CNN-F, which stands for Random Forest with CNN Features. We conducted experiments on a large CMR dataset that we have collected and made publicly accessible. Our method achieved excellent accuracy (99.18%) using Adam optimizer compared to a stand-alone CNN trained using fivefold cross validation (93.92%) tested on the same dataset.

Grape Disease Evaluation Using a Mixed-Model Ensemble of Convolution Neural Network and Random Forest

“…The RF is a classi er that incorporates multiple decision trees on different subsets of the provided dataset and takes the average to enhance the predicted accuracy of that dataset [4][14]. It is based on ensemble learning, which is the act of merging numerous classi ers to solve a complex problem and enhance the model's performance.…”

“…Khozeimeh et al proposed a novel coronary artery disease detection method based on cardiac magnetic resonance images by utilizing deep neural network feature extraction and combining the features with the assistance of an RF. The proposed method outperformed a stand-alone CNN trained using vefold cross validation with an accuracy of 99.18% when tested on the same dataset [4]. Moreover, Mosta z et al proposed an intelligent method for detecting Covid-19 in a chest X-ray image using an RF classi er and a hybridization of deep CNN and discrete wavelet transform optimized features.…”

Body composition predicts hypertension using machine learning methods: A Cohort Study

“…Using this algorithm makes it possible to learn how to combine the results of two or more basic machine learning algorithms in the best possible way. The advantage of this method is that it can use the capabilities of a wide range of well-performing algorithms and make a prediction that is better than the performance of basic algorithms (41).…”

Prediction of Cardiovascular Diseases by Integrating Electrocardiogram (ECG) and Phonocardiogram (PCG) Multi-Modal Features using Hidden Semi Morkov Model

“…In this paper, we examine present study on how the most modern data fusing technologies are bringing clinical and scientific advancements in the field of cardiology. Doctors and scientists will be able to deliver more rapid, efficient, as well as accurate clinical services in the diagnosis and Researchers developed Multi-Layer Acoustic Neural (MLAN) Systems to identify RHD symptoms utilizing heartbeat sounds and electrocardiogram (ECG) data in [25]. To improve accuracy in this proposed MLAN technology, novel methodologies such as multi-attribute acoustic appropriate sampling methods, cardiac sound sampling techniques, ECG information sampling procedure, and Acoustic Support Vector Machine (ASVM) are used.…”