Ovarian Cancer Prediction using Machine Learning

Ovarian cancer is the type of cancer that has the highest recurrence rate in women and poses a serious threat to women. Due to the lack of observable signs, this quiet invader often goes undetected in the beginning, leaving women susceptible to its sneaky development. Once it has advanced and its symptoms are obvious, it becomes arduous to treat. Machine learning is emerging as a ray of hope in the healthcare scene in the current era of rapid technological innovation. Machine learning provides a promising path for the early diagnosis and prediction of ovarian cancer by harnessing the power of algorithms and data analytics, perhaps turning the tide against this tough enemy. The dataset [17] comprises of 349 instances and 49 features. The goal here is to derive a model that helps in classification of instances into two classes: Benign Ovarian Tumors (BOT) and Ovarian Cancer (OC). Therefore, methodical machine learning algorithms are employed to detect ovarian cancer. Our research findings revealed that out of all the classifiers, the Random Forest Model has the highest accuracy (91.43 %). This approach is capable of accurately classifying BOT and OC and could have immense effect on diagnosis of patients, leading to better treatments and higher survival rate.