Cancer Diagnosis through Hidden Markov Model and Gaussian Mixture based Novel DNA Sequencing Approach

Abstract

The research paper focuses on cancer prediction of patients based on DNA sequencing. The whole study is designed around collecting different DNA sequencing samples for patients over the years. The proposed technique in the current research paper is based on the hybridization of the Hidden Markov Model and Gaussian Mixture clustering. HMM, and GM is proposed to identify the expected probability of cancer of the patients. This hybrid model specifies the Bayesian-hidden-Markov model and Gaussian- Mixture clustering approach that is used to identify the genetic variation present in the human Genome. These changes in the Genome may cause cancer. The proposed algorithm is the hybridization of the Bayesian-hidden-Markov model and Gaussian-Mixture clustering approach which provides the optimization of results. The result shows the prediction with better accuracy. The proposed approach shows the cancer prediction with a higher level of accuracy with an improvement of 4.45%. The error rate for the prediction has improved by 2.25%.