skip to Cookie NoticeSkip to contents

Your health and safety remain our top priority: Learn about our Safe Care Commitment | Use our Prescreen app before arrival for faster entry | Read the COVID-19 Vaccine FAQs

Header Skipped.

Radiology Residency Research Opportunities

Per ACGME guidelines, all residents are required to participate in at least one research project that culminates in a publication or presentation at a national meeting. We hold an annual research fair, where residents can learn about opportunities for research throughout the department (see photo below). In addition, residents interested in pursuing more advanced research opportunities may apply for participation in one of the following research tracks.

Clinician-Scientist Research Pathway (CSRP)

Residents specifically interested in a career that includes basic science or translational research can apply into the Clinician-Scientist Research Pathway (CSRP) at the end of the R2 year. Residents in this pathway are given a total of approximately 8 months or more of dedicated research time during the R3 and R4 years during which they learn the fundamentals of performing NIH-funded basic science or translational research in an established laboratory. CSRP represents will present their research at the annual department research symposium. Publication of research is expected and grant writing is encouraged.

Current and Recent CSRP Residents

Clare Poynton, MD, PhD (Class of 2021)
I am a physician and biomedical engineer who applies methods from medical image processing, machine learning and statistics to the analysis of medical imaging data for applications ranging from early detection of disease to prognostication and image-guided intervention. My current research focuses on detecting and predicting the progression of abnormal imaging features on chest CT, termed interstitial lung abnormalities, that may progress to pulmonary fibrosis and are associated with increased mortality (BWH, Applied Chest Imaging Lab, PI: Raul San Jose Estepar, PhD). This work is part of a multi-institution study, COPDGene, that investigates the radiologic and genetic epidemiology of COPD and other smoking-related diseases. I am also interested in the potential of machine learning to improve early detection of disease in healthy populations presenting for lung or breast cancer screening and whether these technologies can be implemented in a way that reduces known disparities in healthcare outcomes.

Shruti Mishra, MD (Class of 2022)
My primary research interests are in developing MRI elastography (MRE) for neuroimaging applications, specifically for evaluating mechanical changes with functional excitation. I have been working on improving a novel MRE set-up and transducer for the use in brain MRE experiments as well as optimizing a multi-slice Ristretto sequence designed to allow interleaved acquisition of two different stimulation states for use in functional studies. I am currently working on continuing to improve our sequence for mapping the human visual cortex using a flashing checkerboard visual stimulation with hopes of subsequently mapping somatosensory cortex activation in human subjects (P.I. Samuel Patz, Department of Radiology).

Khushboo Jhala, MD, MBA (Class of 2022)
My research interests are in strategy and development for tertiary care hospital systems. Specifically, I have worked on scenario planning for radiology departments in hospital systems facing market shocks with shifts reimbursement models (with Dr. Steven Seltzer). Additionally, I have worked on innovation adoption strategies across multiple sectors. This includes studying why good ideas fail just as much as why they succeed. Most recently, we are evaluating the integration of virtual communication networks in hospital systems in the post-Covid era (PI: Ramin Khorasani). Lastly, in order to help teach concepts in strategy and development to the rising generation of radiologists, we are creating the first radiology residency business curriculum, partnering with Harvard Business School.

Center for Evidence-Based Imaging (CEBI)

Residents specifically interested in evidence-based imaging and quality improvement can apply at the end of the R2 year into a combined Residency and Fellowship Training Program at the Center for Evidence-Based Imaging (CEBI). This program is run in conjunction with the Harvard T.H. Chan School of Public Health’s Summer Program in Clinical Effectiveness. This program combines 12 months of translational and performance improvement research with 12 months of a radiology clinical fellowship in any radiology subspecialty. Residents are expected to remain at BWH for the clinical fellowship component of this program to maximize learning and continuity.

Recent CEBI Residents

Tony Trinh, MD (CEBI)
As a part of the BWH radiology residency "3+2” training pathway, I was able to participate in the Center for Evidence Based Imaging (CEBI) Informatics fellowship during my fourth year of radiology residency training. Preparation for this fellowship required enrollment in the Harvard School of Public Health’s Program in Clinical Effectiveness (PCE), which provides a foundation of epidemiological and biostatistics training in preparation for future research endeavors (the HSPH program tuition is subsidized by CEBI). After completion of the PCE, I have spent 80% of my time on imaging informatics research with translation of results to clinical operations initiatives. These projects are focused on quality, safety, and efficiency of care delivery related to medical imaging.

The Center for Clinical Data Science (CCDS)

The MGH & BWH Center for Clinical Data Science (CCDS) is a collaboration between Mass General and Brigham and Women's Hospitals to develop, evaluate, clinically translate, and potentially commercialize artificial intelligence for healthcare. By using AI techniques such as machine learning and convolutional neural networks, the center is building systems to improve the detection, diagnosis, treatment, and management of diseases as well as improve clinical operations. Residents may choose to spend elective time in the R4 year at the CCDS.

Current CCDS Residents

Jisoo Kim, MD
Clare Poynton, MD, PhD

Recent CCDS Residents

Matthew Haber, MD (Class of 2020)
I am interested in applications of machine learning to improve upon screening and diagnosis of neurologic disease that might be difficult for the human eye to detect, particularly in a fast-paced clinical setting. My specific research emphasis is on automated detection of idiopathic normal pressure hydrocephalus (iNPH) on head CT using a deep convolutional neural network (PI: Katherine Andriole, PhD). Our goal is to develop a model which can screen for iNPH, a surgically treatable yet underdiagnosed form of dementia, using a robust clinical ground truth.

Olga Laur, MD (Class of 2020)
Sacroiliitis represents an inflammation of the sacroiliac joint, which has a range of radiographic findings including erosions, reactive sclerosis and ankylosis. Diagnosis of sacroiliitis and its consistent staging is of utter importance in rheumatologic care for ankylosing spondylitis as it represents one of the main diagnostic criteria necessary for administration of appropriate pharmaceutical treatments. However, identification of sacroiliitis on a radiograph can be challenging as it can stem from a subtle finding, with the joint often obscured by overlying bowel gas. In addition, there is a low inter-reader agreement in staging sacroiliitis, which affects clinical care as under staging or missing subtle findings of the disease may prevent patient access to appropriate medications and result in high morbidity and mortality with the progression of the disease.

The goal of this project was to provide automated identification and staging of sacroiliitis and its differentiation from normal pelvic radiographs and its mimics such as osteoarthritis. Automatic detection of sacroiliitis may serve as an “expert opinion” and be a helpful tool for a radiologist to arrive at the correct diagnosis without a delay in patient’s care.

Cory Robinson, MD (Class of 2020)
My data science pathway research project involves using deep learning to identify adrenal glands and adrenal masses on full CT abdomen/pelvis series, with the goal of aiding the radiologist in identifying and characterizing adrenal masses, to help standardize interpretation of these incidental findings, reduce variation in recommendations, and improve patient care. Future endeavors include automatically comparing masses to prior imaging to assess stability and applying the ACR guidelines for management of incidental adrenal masses. (Center for Clinical Data Science, PI: Katherine Andriole, and Brigham and Women’s Hospital, PI: William Mayo-Smith).

Travis Caton, MD (Class of 2019)
Walter Wiggins, MD (Class of 2019)

Image Guided Therapy

In partnership with the National Center for Image Guided Therapy (NCIGT), our department has an NIH R25 fellowship, which could be incorporated into a 3+2 program for the appropriate resident. The Advanced Multimodality Image Guided Operating (AMIGO) suite is the clinical test-bed for research in the NCIGT and is the main area with R25 fellows complete cancer therapy-focused research projects.


For over a century, a leader in patient care, medical education and research, with expertise in virtually every specialty of medicine and surgery.

About BWH