|
|
| |
| An efficient XGBoost-DNN-based classification model for Heart Disease detection system
Download PDF
|
Mrs.M.Sharon Nisha,Dr. G. Rajakumarand Ms. R.Shirly Myrtle
|
Abstract:
Blockchain technology and machine learning are used in the Smart Health care system to provide the best solutions to a variety of issues. With these two cutting-edge developments over the last ten years. In this study, we advocate using blockchain and machine learning models to build secure, open, and intelligent platforms for the Internet of Medical Things. As a result of science and technology's extensive influence on the healthcare industry, a lot of information has been acquired. This vast amount of data has made it considerably more challenging to detect or forecast the presence of a disease in a patient at an early stage. Recent advancements in supervised machine learning algorithms, which aid medical professionals in swiftly and effectively evaluating the gathered data, have made it feasible to quickly and precisely diagnose high-risk disorders at an early stage. This might not only prevent the disease from spreading, but it could also save them a tonne of money on future medical expenses. This paper compares and contrasts a large number of supervised machine learning models for diagnosing illnesses using a wide range of performance criteria. The most popular supervised learning methods were XGBoost and Deep Neural Network (DNN). The XGBoost performed very well in predicting metabolic and cardiovascular diseases. XGBoost was better at predicting diabetes and heart disease, whereas DNN was better prediction is accordingly.
|
|
|
| | |