A multidisciplinary group including experts in medicine, physics, and biochemistry. The group aims to provide a cancer screening method using an ordinary blood test.
Given spectrographic blood-test data, classification of patients as positive/negative, healthy, benign, or malignant.
Small data – ~288 patients, ~4k features.
Each patient’s blood test is measured twice, in high-resolution and low-resolution spectrographic devices.
R programming language.
Business Impact and Insights
Machine learning model is able to classify a person as healthy or not healthy.