Researchers of a study examined whether unsupervised machine learning, based on baseline patient-reported symptom severity, could help stratify older adults with advanced cancer into different risk groups for adverse outcomes. The machine learning algorithm classified patients into low-, medium-, and high-severity clusters based on their symptom severities. Patients in the moderate-severity cluster had a higher likelihood of hospitalization compared with those in the low-severity cluster. Both the moderate- and high-severity clusters were associated with a higher risk of death, but not with toxic effects.

The study demonstrates that unsupervised machine learning can effectively categorize patients into distinct symptom severity clusters, and those with higher pretreatment severity are more likely to experience hospitalization and death.

Reference: Xu H, Mohamed M, Flannery M, et al. An Unsupervised Machine Learning Approach to Evaluating the Association of Symptom Clusters With Adverse Outcomes Among Older Adults With Advanced Cancer: A Secondary Analysis of a Randomized Clinical Trial. JAMA Netw Open. 2023; 1;6(3):e234198. doi: 10.1001/jamanetworkopen.2023.4198.