Research Statement

 

In my time with the UCLA, I have been able to submit and receive several new grants, begin collaboration within the department and university network, publish extensively in high impact journals, lead and enact meaningful changes to the UCLA National Clinician Scholars Program, and firmly establish my research team and myself in a supportive and collaborative research environment. It is a pleasure and an honor to be a faculty member in the UCLA Division of General Internal Medicine and Health Services Research.

I have a longstanding interest in the topics of cancer screening, diagnostic accuracy, and technology. Over the past 20 years, my research programs have expanded scientific understanding of the most effective methods to study diagnostic accuracy, including potential patient, physician, technology, and system factors associated with accuracy. I have led several multi-site R01 studies evaluating diagnostic accuracy of radiologists and pathologists and continue to secure NIH awards as well as private, non-profit foundation funding. My work continues to explore and describe diagnostic variability and the impact of new technology. I am continuing to publish major articles on cancer screening and new screening modalities, including new work into AI and machine learning, as I would like to harness technology to provide a better scientific understanding of the physician's decision-making process.

In my first year at UCLA I received, as Principal Investigator, a new 5-year R01 and a new 3-year U01. This current year (2018-19) I received new 3-year funding from the Melanoma Research Foundation and supported my mentee, Dr. Lee on his first R01 (UCLA will have a subcontract of >$1 million in this 5-year award).

Since my start at UCLA in November of 2017, I have published 56 peer-reviewed papers, 6 editorials and have numerous Up-To-Date topics. As examples, I most recently published a paper in JAMA Network Open, which assessed machine learning for automated differentiation of breast cancer and high-risk proliferative lesions (Mercan E, et al, JAMA Netw Open. 2019 Aug). This work represents the main results of Dr. Mercan's PhD thesis in computer science on which I served on her dissertation committee.

Funding for my work is currently supported through several current grants, on which I am PI, including:

  • ME 190069. Department of Defense Office of Congressionally Directed Medical Research Programs (CDMRP) (2020 - 23), Improving the Diagnosis of Melanoma and Precursor Lesions Among Veterans: Developing AI Techniques and Teledermatopathology;
  • Melanoma Research Alliance (2019-22), Applying AI to Assess Histologic Features to Improve Melanoma Diagnosis;
  • R01 CA225585-02:S1. NIH-NCI National Cancer Institute (2019-22), Reader Accuracy in Pathology Interpretation and Diagnosis: Perception and Cognition (RAPID-PC) (Minority Diversity supplement);
  • R01-CA225585. NIH-NCI National Cancer Institute (2018-23), Reader Accuracy in Pathology Interpretation and Diagnosis: Perception and Cognition (RAPID-PC);
  • U01 CA231782. NIH (2018-21), A Unified Machine Learning Package for Cancer Diagnosis;
  • R01-CA200690. NIH (2016-21), Improving Melanoma Pathology Accuracy through Computer Vision Techniques the IMPACT Study;
  • R01-CA201376. NIH (2016-21), Reducing Errors in the Diagnosis of Melanoma and Melanocytic Lesions (REMI); and
  • NIH, External Validation, Refinement, and Clinical Translation of an Artificial Intelligence Algorithm for Improved Breast Cancer Screening Accuracy. (Received fundable score and awaiting award start date).

 

Joann G. Elmore, MD, MPH