Kashish Agarwal*, Ayush Singh, Hrithik Maheshwari
According to World Health Organization (WHO) Heart diseases are the major cause of death all across the globe, estimating about 17.9 million lives (32% of overall deaths) each year. This group of disorder includes coronary heart disease, cerebrovascular disease, rheumatic heart disease and other conditions. The most of the daily activity behavioural risk factors of heart disease and stroke includes unbalanced diet, physical inactivity, inertness, consumption of tobacco and alcohol. These risk factors may show up among people as raised blood pressure, raised blood sugar level, raised blood lipids and obesity. These intermediate risks factors can be measured in primary care facilities and helps in indicating increased risk of heart complications such as heart attack, stroke, heart failure.
As a traditional method, detection of disease is done by a doctor based on the laboratory test reports. This process involves consultation with multiple doctors by the patient in order to decrease the human error coefficient which not only costs a lot of money but also takes huge time. As a solution to solve this problem, various machine learning based techniques are used to provide non-invasive solutions. In this paper, we propose to use such machine techniques which can be used to check whether a patient has some kind of heart disease or not. We evaluate our approach on several benchmark datasets and show that it outperforms existing state-of-the-art and makes significant contribution.