Communications of the ACM
Multi-Algorithm Approach Helps Deliver Personalized Medicine for Cancer Patients
A major goal of personalized cancer medicine is to accurately predict likely responses to drug treatments based upon genomic profiles of individual patient tumors.
Credit: Wikimedia Commons
Georgia Institute of Technology (Georgia Tech) scientists have used ensemble-based machine learning (ML) algorithms to forecast patients' response to cancer-fighting drugs with high accuracy.
The researchers developed predictive ML-based models for 15 cancer types, based on National Cancer Institute-compiled data from 499 independent cell lines.
The models were validated against a clinical dataset containing seven chemotherapeutic drugs, administered either singularly or jointly, to 23 ovarian cancer patients, which yielded 91% predictive accuracy.
Georgia Tech's John F. McDonald said this preliminary finding "gives me hope that the days of being able to accurately predict optimal cancer drug therapies for individual patients is in sight."
From Georgia Tech Research News
View Full Article
Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA
No entries found