Predicting Severity in Cancer Patients Based On Their Lifestyle and Psychological Wellbeing Using Machine Learning

  • Koduru Manisha, Jasti Sreedevi


Cancer is groups of diseases in which the cells are aggressive, invasive, and metastatic. Cancer causes a residual disability, cell cycle variations and demands distinctive instructions to the patients about treatment modalities like chemotherapy, radiotherapy with proper monitoring and care. The increasing trends in population growth and advancing age have increased the global burden of cancer. The development of cancer is contingent on lifestyle and psychological factors. The living physical environment, sleep-cycle variations, and coping with varying levels of distress may impact the prognosis of the disease in these patients. Once an Individual is diagnosed with cancer, he starts thinking if he is financially stable to avail the medical treatment. The chemo and radiotherapy adds up in causing weakened immune system, rashes, alopecia, gastrointestinal disturbances along with mental health problems and psychological distress. These factors further aggravate the disease severity [1]. The aim of the study is to analyze the impact of lifestyle and psychological factors on the severity of the disease in cancer patients.

A questionnaire-based study was conducted in the omega group of hospitals, oncology department with consent.60 cancer patients in their stage I-II of the disease, aged between 40 to 70 years were considered under the study. Based on the questionnaire the severity of disease in cancer patients was predicted by using Binary Logistic Regression model which is a part of Machine learning. This model has been written in Python in Jupyter Notebook using Anaconda software. The accuracy has been found 96% for this model. The confusion matrix was generated to find the Accuracy, Precision, F1-Score and recall for validating the accuracy.