Though there is a high prevalence of heart disease in the world, several heart conditions often go unrecognized by healthcare professionals. Cardiologists often face difficult decisions in recognizing physical changes in the heart and defining whether they were related to disease or simply just to aging. Clinical decision support for heart disease has been created to assist in early diagnosis of heart diseases which can also help with physician guided management of the disease based on each patient’s clinical picture.
Automatic detection of unwanted biases in the model e.g. on the basis of sociodemographic features. The solution has an interpretability tool that allows the physician to focus on the symptoms and tests that matter for each individual patient (achieved AUC of 0.9 for predicting heart disease).
Clinical decision support for heart disease allows cost optimization, and prevention of heart failure. The solution can also enable detection of left ventricular dysfunction in people without noticeable symptoms. Leveraging this solution, will enable physicians and providers to early risk prediction and diagnosis of serious and complex heart problems. It can also play an important role in education and used by medical students, residents, fellows and experienced surgeons learning new or uncommon procedures with respect to cardiovascular diseases