||5R01CA225585-02 Interpret this number
||University Of California Los Angeles
||Reader Accuracy in Pathology Interpretation and Diagnosis: Perception and Cognition (Rapid-Pc)
Accurate pathologic diagnoses are the cornerstone of both patient care and cancer
research. The diagnostic process requires complex visual perceptual tasks interacting with cognitive
processes, yet little research has been done to understand and potentially improve how skills needed
for accurate diagnoses are acquired. There is an especially profound lack of data on the interpretative
process in the field of pathology. Our prior work identified a concerning level of diagnostic
disagreement and errors in the interpretation of breast biopsies related to cancer screening and
millions of breast biopsies are obtained each year. The proposed research will help improve the
accuracy of pathologists diagnosing breast disease and cancer.
We will examine breast pathology interpretation in residents and experienced pathologists while
they interpret medical images to understand how expertise develops across the entire diagnosis
process from primary diagnoses to second opinions. First, we will examine the development of
expertise among pathology resident trainees at ten U.S. medical schools in both cross-sectional
analysis (Aim 1a) and in a longitudinal analysis gathering data on the same residents annually over
three years of training (Aim 1b). Concurrently, we will study the perceptual and cognitive origins of
diagnostic accuracy and errors among experienced pathologists while they interpret images for
primary diagnosis (Aim 2). Pathologists from Aim 2 will then be asked to provide “second opinions”
on cases to characterize how knowledge of an initial diagnosis impacts the interpretive process when
providing diagnostic second opinions (Aim 3). Using data from Aims 1-3, we will develop and test a
pilot educational program (Aim 4) designed to facilitate the development of expertise in pathology.
The proposed work is innovative in the use of cutting-edge digital whole slide images and
recording eye tracking and high-resolution image navigation behavior and in our plan to follow the
same residents longitudinally as they progress in training over three years. Providing diagnoses using
digital imaging, recently approved by the FDA, is in the future of pathology; our work will both improve
the diagnostic accuracy of current practicing pathologists and guide training the next generation.
Strengths of our application include 1) our multidisciplinary team with experience leading similar
multi-site R01-funded studies; 2) an efficient data collection plan that leverages access to existing
well-characterized breast biopsy cases, a fully developed image viewing and tracking tool, and
synergy between the Aims; 3) unparalleled access to >200 pathologists from ten clinical sites across
the country; and 4) a history of successful physician recruitment into similar studies.
Pathways to breast cancer screening artificial intelligence algorithm validation.
, Houssami N.
, Elmore J.G.
, Buist D.S.M.
Breast (Edinburgh, Scotland), 2019-09-09; , .
What do we know about volumetric medical image interpretation?: a review of the basic science and medical image perception literatures.
, Drew T.
Cognitive research: principles and implications, 2019-07-08; 4(1), p. 21.
Financially effective test algorithm to identify an aggressive, EGFR-amplified variant of IDH-wildtype, lower-grade diffuse glioma.
, Jordan J.T.
, Rapalino O.
, Ramamurthy N.
, Jessop N.
, DeWitt J.C.
, Nardi V.
, Alvarez M.M.
, Frosch M.
, Batchelor T.T.
, et al.
Neuro-oncology, 2019-05-06; 21(5), p. 596-605.
A review of eye tracking for understanding and improving diagnostic interpretation.
, Drew T.
, Weaver D.L.
, Elmore J.G.
Cognitive research: principles and implications, 2019-02-22; 4(1), p. 7.