Designing Personalized Treatment Plans for Breast Cancer Patients: A Predictive Analytics Approach
Posted: 26 Jul 2017 Last revised: 11 Jan 2019
Date Written: July 24, 2017
Contemporary treatments for breast cancer are complex and multimodal, involving highly specialized medical professionals working across a variety of settings. This poses many challenges to the design and delivery of treatment to individual patients. This paper proposes a novel method to optimize treatment plans using predictive analytics. Specifically, unlike the traditional method which prescribes homogeneous plans for all patients, our method customizes the treatment plans by accurate predictions of the amount and distribution of tumor cells based on patients’ data from different sources. Using extensive simulation experiments based on different real-world complications, we demonstrate that our personalized treatment plans which account for patients’ clinical profiles can significantly improve the local tumor control while reducing the side effects. Our findings show the great promise of meaningful use of electronic health records and effective sharing of patient information through different stages of care delivery process. The proposed simulation framework can be used as an effective and economical decision support tool for clinical researchers in suggesting and screening effective treatment plans to be tested in future clinical trials.
Keywords: clinical decision support, health information sharing, personalized medicine, predictive analytics, treatment planning
Suggested Citation: Suggested Citation