Predictive Human Heart Anomaly Detector

Abstract

Cardiovascular diseases (CVDs), commonly known as heart attack or strokes are the number 1 cause of death globally, taking an estimated 17.9 million lives each year, accounting to about 31% of total deaths across the world every year, according to the World Health Organization. One of the primary reasons behind CVD is hyper homocysteinea, that is higher level of an amino acid name homocysteine, causing blood clots in the arteries which results is a heart attack. This work proposes a novel system to detect homocysteine level in a non-invasive way through detection of hypothyroid as well as Vitamin B-12 deficiency. Neural networks have been used in this predictive model with an accuracy of 80%.