Application of Machine Learning in case analysis for potential complications of Hypertension

Abstract

Worldwide, “Hypertension” involves numerous risk factors and can be cured by medication as well as changing lifestyle. Hypertension becomes the global burden. The health complications can be prevented by early diagnosis and prediction. This research has the objective to learn the procedure in order to develop and compare prognostic models to recognize the person who has at the high risk of evolving hypertension. Such process does not need any clinical procedures. This comprehensive review illustrates the accurate utilization of AI algorithm in order to envisage the clinical outcomes in big data such that the information consists the high volume of data with diverse value, velocity, veracity and variety, veracity. Several examples have been considered here that used the latest technologies. For instance, Support Vector Machine (SVM) and Deep Learning approaches are included in this review. In the first example, deep learning and support vector machine (SVM) anticipated the manifestation of cardiovascular measures with the accuracy of 56%–57%. Next case involves the use of neural network algorithm on the database of 378,256 patients. It estimated the manifestation of cardiovascular measures taken at the duration of ten years and shown the results as specificity (71%) and sensitivity (68%). The third study involves the usage of machine learning algorithm in order to find the chances of hypertension in 1,504,437 with specificity (99%), sensitivity (51%) and also represents the area under the curve of 87%. In last, study involves the usage of advanced devices like portable gadgets and wearable biosensors for the assessment of the high-risk individuals who are suffering from hypertension using photoplethysmography. It can discriminate the individuals who have the high chances of developing the hypertension by measuring the positive predictive value (> 90%) and sensitivity (> 80%). Such above mentioned outcomes shows the variations according to the conventional risk aspects for atherosclerotic disease as well as demographics.