.png){kind=icon}
.png){kind=icon}
Return
Expertise under the microscope: processing histopathological slides
.png){kind=icon}
OA
AI
DOI：10.1111/medu.12385.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
ObjectivesAlthough the obvious goal of training in clinical pathology is to bring forth capable diagnosticians, developmental stages and their characteristics are unknown. This study therefore aims to find expertise-related differences in the processing of histopathological slides using a combination of eye tracking data and verbal data. MethodsParticipants in this study were 13 clinical pathologists (experts), 12 pathology residents (intermediates) and 13 medical students (novices). They diagnosed 10 microscopic images of colon tissue for 2seconds. Eye movements, the given diagnoses, and the vocabulary used in post hoc verbal explanations were registered. Eye movements were analysed according to changes over trial time and the processing of diagnostically relevant areas. The content analysis of verbal data was based on a categorisation system developed from the literature. ResultsAlthough experts and intermediates showed equal levels of diagnostic accuracy, their visual and cognitive processing differed. Whereas experts relied on their first findings and checked the image further for other abnormalities, intermediates tended to double-check their first findings. In their explanations, experts focused on the typicality of the tissue, whereas intermediates mainly mentioned many specific pathologies. Novices looked less often at the relevant areas and were incomplete, incorrect and inconclusive in their explanations. Their diagnostic accuracy was correspondingly poor. ConclusionsThis study indicates that in the case of intermediates and experts, different visual and cognitive strategies can result in equal levels of diagnostic accuracy. Lessons for training underline the relevance of the distinction between normal and abnormal tissue for novices, especially when the mental rotation of 2-D images is required. Intermediates need to be trained to see deviations in abnormalities. Feedback and an educational design that is specific to these developmental stages might improve training. Discuss ideas arising from the article at '
Keywords:
VIRTUAL SLIDES
PERFORMANCE
KNOWLEDGE
PATHOLOGY
MODEL
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
5.2
Papers:
8.2K
Citations:
1.3W