Survival Prediction of Heart Failure Patients Using Lasso Algorithm and Gaussian Naive Bayes Classifier

EasyChair Preprint 6912

Abstract

Cardiovascular diseases kill approximately 17 million people globally per annum , and that they mainly exhibit as myocardial infarctions and heart failures. Heart failure (HF) occurs when  heart  cannot  pump  enough  blood  to  satisfy  the  requirements  of  the  body.  Available electronic medical records of patients quantify symptoms, body features, and clinical laboratory test  values,  which  may  be  wont  to  perform  biostatistics  analysis  aimed  toward  highlighting patterns   and   correlations   otherwise   undetectable   by   medical   doctors.   Health   plans   must prioritize disease management efforts to scale back hospitalization and mortality rates in heart disease  patients.  We  developed  a  risk  model  to  predict  the  5-year  risk  of  mortality  or hospitalization for heart disease among patients at an outsized health maintenance organization.While  performing  partitioning  recursively,  it  sequences  partitioning greedily instead of finding the optimal partitioning sequence. In proposed system, the LASSO algorithm  is  used  to  select  features  and  classification  using  Gaussian  Naïve  Bayes,  and investigate   the   results. Lasso and ridge regression with Gaussian Naïve Bayes (GNB) classifiers has given better results in most of the casess

Keyphrases: Lasso algorithm, MachineLearning, Patient

@booklet{EasyChair:6912,
  author    = {Velmurugan Sathya Narayanan and Kavin N Raj and Kishore Kumar and Manoj Kumar},
  title     = {Survival Prediction of Heart Failure Patients Using Lasso Algorithm and Gaussian Naive Bayes Classifier},
  howpublished = {EasyChair Preprint 6912},
  year      = {EasyChair, 2021}}