Early Detection of Breast Cancer Based on Patient Symptom Data Using Naive Bayes Algorithm on Genomic Data

The Naive Bayes algorithm is employed to facilitate early detection of breast cancer, a prevalent condition affecting women. Timely detection plays a crucial role in ensuring effective treatment and improved survival rates. Considering the limitations of conventional methods like mammography, we explored the potential of genomic analysis, made possible by recent technological advancements. Genomic data from breast cancer patients were collected and analyzed to identify genetic patterns associated with the disease. Patient symptom data was also incorporated. Utilizing the Naive Bayes algorithm, we processed this dataset and classified patients into either the “breast cancer” or “non-breast cancer” categories based on the probability of symptoms in patients with similar genomic profiles. Our findings indicate that this approach holds promise in accurately detecting breast cancer at an early stage, with the potential to enhance prognosis and treatment outcomes. However, further validation through clinical studies involving larger and diverse samples is necessary. The results demonstrated that the Naive Bayes algorithm achieved a high accuracy rate of 90.4% in classifying early symptoms of breast cancer into two distinct categories.