Dr. Lungren 2016 AUR GERRAF Recipient
2016 Fellows: Matthew P. Lungren, MD, MPH, Stanford University School of Medicine; Paggie Kim, MD, Loma Linda University.
Photo credit: Association of University Radiologists
July 26, 2016
The Association of University Radiologists (AUR) announced Matthew P. Lungren, MD, MPH, an Assistant Professor of Pediatric Radiology, a recipient of the 2016 GE Radiology Research Academic Fellowship (GERRAF) at its 64th Annual Meeting in San Diego, California March 2016.
An abridged version of his proposal, "Use of Predictive Modeling and Machine Learning to Develop Clinical Decision Support for Pulmonary Embolism CT Imaging in Children" follows: Due to the growing burden of venous thromboembolism (VTE) and its life-long complications, over 100-fold increase in the past two decades, there is an urgent need for guidelines for the surveillance of VTE in children. An evidence-based diagnostic approach for imaging of VTE has not been established in the pediatric literature and evidence from adult studies have been shown to be inapplicable to the pediatric population. Currently pediatric physicians must rely on their best judgment instead of on evidence-based quantitative analysis. This is made more difficult, in part, because childhood VTE risk factors are shown in small series to be disparate, some overlapping, and cannot all be accounted for in routine clinical decision making. The overall concept of this proposal is that the availability of vast clinical EMR repositories can lend themselves to a treasure trove of point-of-care, relevant, actionable data generating practice-based evidence to inform clinical management. Radiology reports are an important potential source of information for these analyses and the use of machine learning techniques may allow for unlocking the radiology report information.We believe that the imaging report narrative can be correlated with clinical data to devise accurate predictive models of medical imaging yield. We propose to achieve this by creating a predictive model that leverages real-time EMR clinical and laboratory data and prior imaging report data to arrive at a patient-specific imaging prediction. This will allow pediatric clinicians the ability to leverage aggregate patient data for decision making at the point of care.
The GERRAF program was initiated in 1992 to help meet the scholarly and research support needs of radiology. The specific purpose of the GERRAF Award is to bring the benefits of radiological advances to medical practice and the public. The mission is twofold. First, it is to develop a cadre of academic radiologists trained in patient-oriented and health services research and demonstrate the value of such training to the field of radiology. Second, it is to provide an opportunity for a critical mass of young radiologists from a wide geographic distribution of academic health centers to receive exceptional training in clinical research methodology. The Fellowship also serves to develop the academic careers of its recipients through a program of mentoring and networking, each being a key element of professional success.
Funding is made possible by an unrestricted grant provided by GE Healthcare.
Congratulations, Dr. Lungren!