Machine learning classification of diagnostic accuracy in pathologists interpreting breast biopsies.

OBJECTIVE: This study explores the feasibility of using machine learning to predict accurate versus inaccurate diagnoses made by pathologists based on their spatiotemporal viewing behavior when evaluating digital breast biopsy images. MATERIALS AND METHODS: The study gathered data from 140 pathologists of varying experience levels who each reviewed a set of 14 digital whole slide images of breast biopsy tissue. Pathologists' viewing behavior, including zooming and panning actions, was recorded during image evaluation. A total of 30 features were extracted from the viewing behavior data, and 4 machine learning algorithms were used to build classifiers for predicting diagnostic accuracy. RESULTS: The Random Forest classifier demonstrated the best overall performance, achieving a test accuracy of 0.81 and area under the receiver-operator characteristic curve of 0.86. Features related to attention distribution and focus on critical regions of interest were found to be important predictors of diagnostic accuracy. Further including case-level and pathologist-level information incrementally improved classifier performance. DISCUSSION: Results suggest that pathologists' viewing behavior during digital image evaluation can be leveraged to predict diagnostic accuracy, affording automated feedback and decision support systems based on viewing behavior to aid in training and, ultimately, clinical practice. They also carry implications for basic research examining the interplay between perception, thought, and action in diagnostic decision-making. CONCLUSION: The classifiers developed herein have potential applications in training and clinical settings to provide timely feedback and support to pathologists during diagnostic decision-making. Further research could explore the generalizability of these findings to other medical domains and varied levels of expertise.

Retrieval practice enhances near but not far transfer of spatial memory.

Students learn more effectively through repeated retrieval of study materials relative to repeated exposure to the materials, a phenomenon known as the testing effect or retrieval practice. This pattern has been demonstrated repeatedly with verbal materials, and more recently with visuospatial materials. The extent to which retrieval practice produces spatial memories that successfully transfer to more diverse task demands remains unknown. Transferring spatial memory to novel task demands can involve challenging orientation and perspective transformations, possibly limiting the benefits of retrieval practice for application to realistic spatial tasks. In 4 experiments, participants learned a map of a large-scale urban environment, engaging in either study practice (repeated exposure) or retrieval practice (exposure and testing). Across experiments we varied the retrieval demands of the final memory test, increasing the breadth of transfer from study to test (from near to far transfer). Final memory tests included reconstructing a map from memory (Experiment 1), judgments of relative direction from an allocentric perspective (Experiment 2), judgments of relative direction from an egocentric perspective (Experiment 3), and navigating between target landmarks within the learned environment (Experiment 4). Results demonstrated that retrieval practice enhances near to medium transfer of memory for the map itself, including accessing spatial memory from varied orientations. However, it does not assist in medium to far transfer of spatial knowledge to pointing or navigation tasks performed from an alternate perspective. Results are considered in the context of domain-specific theories of spatial memory and navigation, and domain-general theories of learning strategies and transfer. (PsycINFO Database Record (c) 2019 APA, all rights reserved).