In collaboration with Prof. Mark Phillips, I am exploring a couple of approaches to improving decision making in the radiation oncology setting.
First, we are exploring the ability of Bayesian Nets to detect incorrect, or sub-optimal radiation treatment plans. Our approach is novel, in that we are exploring the ability to create and modify the topology of these Bayes nets automatically, as new information and data become available.
Second, in the follow-up setting for radiation treatment patient, we are looking to improve decision making for those patients who may need additional imaging or treatment. As an example domain, we focus on CNS disease, such as brain cancers, or brain metastases. Our approach here is patient-centered; we plan to provide a smart-phone app for patients to log self-reported symptoms and standard neurological test that assess cognitive and motor skills. These results can then be transmitted as needed to the clinical decision maker for followup. Our approach is in contrast to the current standard of care, where the followup patient is scheduled for visits every 3 months to assess symptoms.
Project Keywords: Clinical Informatics