The goal was to develop prediction models of malignancy risk of abnormalities (aka nodules) found in patients’ lungs. The predictive models are based on a diverse set of machine learning algorithms whose inputs are various measured characteristics of the nodules and patient demographics, such as:
- Patient demographics such as years smoking, age, gender, race, etc.
In addition, computer vision algorithms were developed to extract these characteristics directly and ‘auto-magically’ from the images so as to automate the detection process. The core technologies were then deployed to a cloud based infrastructure providing a software-as-a-service (behind web APIs) which physicians could access as an aide in the detection of cancer and the quantification of malignancy risk for a given nodule.
Finally, plugins for multiple DICOM viewers were developed to facilitate physician interaction with said services.