Meredith Cummings BSN, RN, OCN, highlights her research on machine learning and its potential application in optimizing symptom management.
Machine learning may have a growing role in predicting adverse events in patients undergoing cancer therapy, according to Meredith Cummings BSN, RN, OCN.
Cummings, who is a PhD student at the University of Pittsburgh School of Nursing, and who works in an outpatient infusion center with Allegheny Health Network Cancer Institute, recently presented on machine learning in a poster presentation, at the 48th Annual Oncology Nursing Society Congress.
The purpose of the review was to characterize the state of machine learning use in predicting treatment-related symptoms. The review was also designed to identify the gaps related to machine learning predictive models.
Investigators conducted their review in accordance with the Preferred Reporting for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews guidelines. They used PubMed to search for articles and selected those that evaluated machine learning to predict anticancer therapy symptoms for patients with head and neck cancer. Their initial search found 331 articles. After the abstracts were screened for eligibility, a total of 30 articles were evaluated.
Ultimately, a large proportion of articles included all patients with head and neck cancers (n = 11). Nine studies included patients with esophageal and nasopharyngeal cancers and 1 included patients with laryngeal cancer. The most explored treatment modality was radiation (62%). Of note, only 3 studies included external validation measures to assess the predictive models. In the synthesis of these studies, the most commonly predictive symptoms included radiation-induced xerostomia, radiation induced temporal lobe injury, and chemotherapy-induced myelosuppression. These models were able to predict treatment-related symptoms with an area under the curve (AUC) accuracy range of 0.65 to 0.85.
According to Cummings, machine learning may prove a useful tool to optimize symptom management in this patient population. She encourages nurses to familiarize themselves with machine learning and its potentially predictive capabilities.
Cummings M, Nilsen M, Bender C, Al-Zaiti S. Predicting anti-cancer treatment-related symptoms in patients with head and neck cancer using a machine learning approach: a scoping review. Poster presented at: 48th Annual Oncology Nursing Society Congress; April 26-30, 2023; San Antonio, TX. Accessed May 25, 2023. https://ons.confex.com/ons/2023/meetingapp.cgi/Paper/13510
PVd Confers a Small OS Benefit to Patients With Relapsed/Refractory Multiple Myeloma
Neoadjuvant Durvalumab Combination Misses Mark in EGFR+ NSCLC
Durvalumab Garners Similar OS, Albeit Varied PFS, Outcomes in EGFR+ and EGFR-Wildtype NSCLC Subgroups
Quizartinib Extends Overall Survival Across Multiple Subgroups in FLT3-ITD+ AML