CERSI Collaborators: Christian Pavlovich, MD

FDA Collaborators: Danica Marinac-Dabic, MD, PhD; Benjamin Fisher, PhD; Charles Viviano, MD, PhD

CERSI Subcontractors: Weill Cornell Medicine Art Sedrakyan, MD, PhD; Jim C. Hu, MD, MPH; Jialin Mao, MD, MS; Miko Yu, MA; Sendong Zhao, PHD, Vahan Simonyan, PhD

Project Start Date: September 2019

Regulatory Science Challenge

Traditionally, prostate cancer has been treated with whole gland surgical and radiation therapies. However, there is growing interest in partial gland ablation (PGA), which aims to decrease side effects, including erectile dysfunction and urinary incontinence. Considering the growing interest in PGA, there is a role for actively monitoring the safety and effectiveness outcomes with the various available medical devices used in this procedure, such as high intensity focused ultrasound and cryoablation (treatment to kill cancer cells with extreme cold).

Project Description and Goals

The goal of this project is to develop and apply natural language processing (NLP) tools to perform active surveillance of medical devices used for PGA. NLP is a type of methodology that enables computers to search and analyze large amounts of data stored in the natural (human) language. NLP tools will be used to analyze data elements from text in electronic health records (EHR), such as magnetic resonance imaging (MRI) radiology reports as well as pathology reports from tissue samples (biopsy). A radiology report provides clinical findings of the MRI imaging study and a pathology report provides additional information about the diagnosis and disease state. The NLP tools developed and applied in this project may facilitate an efficient, cost-effective active surveillance of devices performing PGA. These NLP tools may also be expanded for application to other medical devices and disease areas.