Newswise – Tampa, Florida (September 15th, 2025)) – New research Featured in JCO Clinical Cancer Informatics Demonstrate which patients can predict a machine learning model that incorporates patient report results and wearable sensor data Non-small cell lung cancer You are at the highest risk of needing emergency treatment during treatment. This study was guided by researchers and clinicians Moffit Cancer Center.
Patients receiving systemic therapy for non-small cell lung cancer often experience treatment-related toxicities that can result in unexpected emergency care visits. In this study, Moffitt researchers tested whether integrating multiple sources of patient-generated health data, including self-reported quality of life investigations and wearable device metrics such as sleep and heart rate, could improve forecasts beyond standard clinical and demographic information.
The team used an explanatory machine learning approach called the Bayesian network to build a predictive model among 58 patients monitored on a Fitbit device and surveyed in the survey. Machine learning models, including patient-reported results and wearable sensor data, far outperform the models, based solely on clinical data on their ability to distinguish between high-risk and low-risk patients.
“Combining the information allows patients to provide their symptoms and provide ongoing monitoring from wearable devices to better identify those at the highest risk of treatment complications,” he said. Dr. Brian D. Gonzalezauthors and researchers of Moffitt's Ministry of Health's outcomes and actions. “Our goal is to intervene early on the clinician's tools, improve patient experience and potentially prevent hospitalizations.”
Findings suggest that integrating multidimensional data into machine learning models could enhance personalized cancer care and enable providers to actively deal with toxins before they escalate. Although this study was limited to a single center and a modest sample size, researchers say this approach is promising for a wider range of applications.
“What makes this approach powerful is not only the accuracy of predictions, but also the ability to understand why models reach those predictions,” he said. Yi Luo, Ph.D.co-authors and researchers of Moffitt's Faculty of Mechanical Sciences. “By using explanatory machine learning methods, we can see how factors such as symptom reporting, sleep quality, and lab outcomes interact to influence risk. This transparency is important in that we can build trust with our clinicians and use models to guide real-world decisions in cancer treatment.”
Future studies will expand the model to include additional clinical and molecular data, and validate the results in a larger multicenter cohort.
This study was supported by the National Institutes of Health (P30 CA076292).
aBout Moffitt Cancer Center Moffit I'm dedicated to one life-saving mission. It contributes to the prevention and treatment of cancer. There are only one of 57 Tampa-based facilities Comprehensive Cancer Centers Designated by the National Cancer InstituteMoffitt's distinction that recognizes scientific excellence, interdisciplinary research, robust training and education. Moffitt's specialist nursing staff are recognized by the American Nurses Credentialing Center using Magnet® Status, its best distinction. For more information, please call 1-888-Moffitt (1-888-663-3488) moffitt.organd follow the momentum Facebook, x, Instagram and YouTube.
###
