Cancer Prediction using Machine Learning: A Survey

Abstract

Carcinoma is a terrible illness that has been rising in morbidity at an alarming rate in recent years all over the world. Computer-assisted cancer prediction has made significant progress thanks to the rapid growth of computer science and machine learning technology. A novel technique that includes many machine learning models and uses deep learning in an ensemble manner. Five distinct categorization models using meaningful gene data are been derived from differential gene expression analyses. The results of the five classifications are then combined using a deep learning algorithm. To choose features in T2DM, Fisher's score, RFE, and a decision tree are been used. The prevalence of diabetes was predicted using random forest, logistic regression, SVM, and MLP. The MIMICIII data collection is being used to construct and train several algorithms aimed at predicting DM patient death by using deep learning model. Breast cancer (BC) is also one of the most common reasons of worry across the world. It was the world's second most often diagnosed cancer and the fifth leading cause of mortality. High precision outcomes are frequently obtained at the expense of sensitivity. Distance-based clustering with Euclidean distance, the k-means method, and discretization are among the machine learning techniques employed.