Learning from data to predict future symptoms of oncology patients

• Published in: PloS one Vol. 13; no. 12; p. e0208808
• Main Authors: Papachristou, Nikolaos, Puschmann, Daniel, Barnaghi, Payam, Cooper, Bruce, Hu, Xiao, Maguire, Roma, Apostolidis, Kathi, Conley, Yvette P, Hammer, Marilyn, Katsaragakis, Stylianos, Kober, Kord M, Levine, Jon D, McCann, Lisa, Patiraki, Elisabeth, Furlong, Eileen P, Fox, Patricia A, Paul, Steven M, Ream, Emma, Wright, Fay, Miaskowski, Christine
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 31.12.2018; Public Library of Science (PLoS)
• Subjects: Algorithms; Anxiety; Anxiety - psychology; Artificial intelligence; Biology and Life Sciences; Cancer; Chemotherapy; Computer and Information Sciences; Computer applications; Correlation analysis; Critical components; Depression - psychology; DNA methylation; Epidemiology; Female; Humans; Male; Medicine and Health Sciences; Mental depression; Midwifery; Models, Psychological; Neoplasms - psychology; Neural networks; Neural Networks, Computer; Nonlinear analysis; Nursing schools; Oncology; Patients; Physical Sciences; Prediction models; Regression analysis; Research and Analysis Methods; Signal processing; Sleep; Software; Support Vector Machine; Support vector machines; Symptom management; Teaching methods; Variables; Greece; Glasgow Scotland; San Francisco California; United Kingdom > UK; United States > US; Ireland; Scotland; California
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0208808

Effective symptom management is a critical component of cancer treatment. Computational tools that predict the course and severity of these symptoms have the potential to assist oncology clinicians to personalize the patient's treatment regimen more efficiently and provide more aggressive and timely interventions. Three common and inter-related symptoms in cancer patients are depression, anxiety, and sleep disturbance. In this paper, we elaborate on the efficiency of Support Vector Regression (SVR) and Non-linear Canonical Correlation Analysis by Neural Networks (n-CCA) to predict the severity of the aforementioned symptoms between two different time points during a cycle of chemotherapy (CTX). Our results demonstrate that these two methods produced equivalent results for all three symptoms. These types of predictive models can be used to identify high risk patients, educate patients about their symptom experience, and improve the timing of pre-emptive and personalized symptom management interventions.