Implementing ML for breast cancer risk prediction
Breast cancer affects about 1 in 8 women in the US. In 2020, more than 2.3 million women were diagnosed with breast cancer worldwide and 685,000 died as a result. These grim statistics provide enough information for us to conclude that breast cancer is a deadly disease. There have been several procedures used to diagnose breast cancer, from genomic testing to imaging-based studies. One common method is to look at the characteristics of the cell nuclei derived from imaging studies and classify them as malignant (M) or benign (B). In this example implementation, we will use this method to predict whether a breast mass is M or B using cell nuclei features. This prediction can be generated at various stages of the progression of the disease as the features of the cell nuclei change. This will help us determine whether a patient is at risk of developing a malignant breast mass over a period of time. Early determination of this risk and timely...