Communications of the ACM
Cancer Diagnosis, Treatment Could Get a Boost from ML
The new analysis technique is compatible with "liquid biopsy" testing equipment already approved in the U.S. and in use in cancer clinics, which could speed the new method’s path to helping patients.
Credit: RiseApps
A multi-institutional team of researchers led by the University of Wisconsin-Madison designed machine learning (ML) algorithm to help physicians diagnose specific types of cancer and select the most effective treatment.
The researchers utilized cell-free DNA in blood samples from a previous study of nearly 200 patients, and samples from more than 300 people treated for breast, lung, prostate, or bladder cancers.
They split each sample group into two portions, using one to train the ML algorithm to find patterns among cell-free DNA fragments and testing the program on the other.
The algorithm could translate a liquid biopsy's results into a cancer diagnosis and a patient's specific cancer type with more than 80% accuracy.
The approach also could differentiate adenocarcinoma from neuroendocrine prostate cancer.
From University of Wisconsin-Madison News
View Full Article
Abstracts Copyright © 2023 SmithBucklin, Washington, D.C., USA
No entries found