Predicting Cancer diagnosis based on preliminary screenings
I want to see if it is possible to predict the likely-hood of a cancer diagnosis based on collected data from screenings and other diagnostics performed.
I will be using machine learning algorithms in an attempt to predict cancer diagnosis based on the features available in PLCO. I am currently working with one of the SEER datasets and my algorithms have been able to predict, with a fair amount of precision, the length of time a person might live after multiple cancer diagnosis. I would like the bring some of that predictive power to these datasets and see if it is possible to predict the likelihood of a cancer diagnosis. If successful, I would hope to continue my research on further datasets.
-Predict cancer diagnosis with a better than baseline probability