An IoT-based Framework for Detecting Heart Conditions using Machine Learning

A lot of diseases may be preventable if they can analyzed or predicted from patient historical and family data. Predicting diagnosis depends on the gathered clinical physiological data patients. The more collected medical healthcare data, knowledge support system support. Hence, real monitoring for patients is trend this decade based Internet Things technologies (IoT). IoT models facilitate human life by easily collecting remotely recognizing that are treatable it diagnosed early. This paper proposes a framework consisting two models: (i) heart attack detection model (HADM); (ii) Electrocardiosignal ECG heartbeat multiclass-classification (ECG-HMCM). Gridsearch used to hyperparameters optimization different machine learning (ML) techniques. dataset in HADM consists 1190 14 features. As foundation diagnosing cardiovascular disease arrhythmia hence, we propose an multi-class classification using MIT-BIH Arrhythmia PTB Diagnostic signals which contains five categories with 109446 samples. K Nearest Neighbor (KNN) technique applied build ECG-HMCM addition algorithm hyperparameter aiming improve accuracy achieved 97.5%. proposed aims remotely. outcomes experiments show suggested works well practical setting.