A Personalized Cancer Diagnosis using Text based Clinical Data with Machine Learning Models

Predictive analytics using machine learning in the healthcare industry is a difficult task, but it also aids doctors in accurately treating the disease to save time, money, and most importantly, lives. In order to precisely treat diseases like cancer, medical professionals devote an extensive amount of time and effort to manually analysing the gene sequences of patients. Numerous thousand genetic mutations are governed by the cancer tumour sequence. The personalised cancer diagnosis using gene mutations and clinical text through machine learning (ML)-based techniques aims to predict the type and severity of cancer diseases based on the patient’s gene mutations and clinical text data. The gene mutation data is analysed to identify the specific genetic mutations associated with the patient’s cancer using ML algorithms. Then, the extracted features from the clinical text data and the gene mutation data are combined to develop a predictive model using different ML algorithms such as Logistic Regression, Random Forest, Support Vector machine, or Naive Bayes.