Since cardiac arrhythmias are a serious threat to public health, quick and accurate diagnostic techniques need to be developed. The prediction of cardiac arrhythmia using machine learning algorithms—Decision Tree Classifier and Gaussian Naive Bayes, in particular—is investigated in this work. The dataset used in this study includes a variety of physiological indicators that were gathered from a varied population, including blood pressure, heart rate, and electrocardiogram (ECG) readings. To improve input feature quality and minimize dimensionality, preprocessing and feature selection are first performed. The pre-processed data is then used to train Gaussian Naive Bayes and Decision Tree Classifier models, which are used to predict whether cardiac arrhythmia will occur or not. Our findings show that both algorithms perform well in identifying cases of cardiac arrhythmia, with encouraging results. While Gaussian Naive Bayes uses probabilistic modeling to produce predictions, Decision Tree Classifier provides understanding, enabling the identification of the most significant features. To guarantee the models' usefulness in clinical practice, they are further evaluated for sensitivity, specificity, and precision. Healthcare professionals may find the combination of Gaussian Naive Bayes and Decision Tree Classifier to be a reliable method for predicting cardiac arrhythmias, which could help with early diagnosis and intervention. These results support continued efforts to improve cardiac arrhythmia detection efficiency and accuracy, which will ultimately improve patient outcomes.
Cardiac arrhythmia, Machine learning, Decision Tree Classifier, Gaussian Naive Bayes, Diagnosis.
IRE Journals:
Ashish Verma , Bhavesh Sutrave , S K Singh , Rimsy Dua
"Deep Learning-based Intelligent Diagnosis of Cardiac Heartbeat Disorders" Iconic Research And Engineering Journals Volume 7 Issue 8 2024 Page 113-118
IEEE:
Ashish Verma , Bhavesh Sutrave , S K Singh , Rimsy Dua
"Deep Learning-based Intelligent Diagnosis of Cardiac Heartbeat Disorders" Iconic Research And Engineering Journals, 7(8)