Modeling eye movement patterns to characterize perceptual skill in image-based diagnostic reasoning processes

Eye movements reflect expertise development in hybrid search

Domain-specific expertise changes the way people perceive, process, and remember information from that domain. This is often observed in visual domains involving skilled searches, such as athletics referees, or professional visual searchers (e.g., security and medical screeners). Although existing research has compared expert to novice performance in visual search, little work has directly documented how accumulating experiences change behavior. A longitudinal approach to studying visual search performance may permit a finer-grained understanding of experience-dependent changes in visual scanning, and the extent to which various cognitive processes are affected by experience. In this study, participants acquired experience by taking part in many experimental sessions over the course of an academic semester. Searchers looked for 20 categories of targets simultaneously (which appeared with unequal frequency), in displays with 0-3 targets present, while having their eye movements recorded. With experience, accuracy increased and response times decreased. Fixation probabilities and durations decreased with increasing experience, but saccade amplitudes and visual span increased. These findings suggest that the behavioral benefits endowed by expertise emerge from oculomotor behaviors that reflect enhanced reliance on memory to guide attention and the ability to process more of the visual field within individual fixations.

Focal lung pathology detection in radiology: Is there an effect of experience on visual search behavior?

In radiology, 60% to 80% of diagnostic errors are perceptual. The use of more efficient visual search behaviors is expected to reduce these errors. We collected eye-tracking data from participants with different levels of experience when interpreting chest X-rays during the completion of a pathology-detection task. Eye-tracking measures were assessed in the context of three existing visual search theories from the literature to understand the association between visual search behavior and underlying processes: the long-term working memory theory, the information-reduction hypothesis, and the holistic model of image perception. The most experienced participants (radiology residents) showed the highest level of performance, albeit their visual search behaviors did not differ from the intermediate group. This suggests that radiology residents better processed the represented information on the X-ray, using a visual search strategy similar to the intermediate group. Since similar visual search resulted in more information extraction in the radiology residents compared with the intermediates, we suggest that this result might support the long-term working memory theory. Furthermore, compared with novices, intermediates and radiology residents fixated longer on areas that were more important to avoid missing any pathology, which possibly confirms the information-reduction hypothesis. Finally, the larger distances between fixations observed in more experienced participants could support the holistic model of image perception. In addition, measures of generic skills were related to a lower time cost for switching between global and local information processing. Our findings suggest that the three theories may be complementary in chest X-ray interpretation. Therefore, a unified theory explaining perceptual-cognitive superiority in radiology is considered.

Saliency-driven system models for cell analysis with deep learning

BACKGROUND AND OBJECTIVES

Saliency refers to the visual perception quality that makes objects in a scene to stand out from others and attract attention. While computational saliency models can simulate the expert's visual attention, there is little evidence about how these models perform when used to predict the cytopathologist's eye fixations. Saliency models may be the key to instrumenting fast object detection on large Pap smear slides under real noisy conditions, artifacts, and cell occlusions. This paper describes how our computational schemes retrieve regions of interest (ROI) of clinical relevance using visual attention models. We also compare the performance of different computed saliency models as part of cell screening tasks, aiming to design a computer-aided diagnosis systems that supports cytopathologists.

METHOD

We record eye fixation maps from cytopathologists at work, and compare with 13 different saliency prediction algorithms, including deep learning. We develop cell-specific convolutional neural networks (CNN) to investigate the impact of bottom-up and top-down factors on saliency prediction from real routine exams. By combining the eye tracking data from pathologists with computed saliency models, we assess algorithms reliability in identifying clinically relevant cells.

RESULTS

The proposed cell-specific CNN model outperforms all other saliency prediction methods, particularly regarding the number of false positives. Our algorithm also detects the most clinically relevant cells, which are among the three top salient regions, with accuracy above 98% for all diseases, except carcinoma (87%). Bottom-up methods performed satisfactorily, with saliency maps that enabled ROI detection above 75% for carcinoma and 86% for other pathologies.

CONCLUSIONS

ROIs extraction using our saliency prediction methods enabled ranking the most relevant clinical areas within the image, a viable data reduction strategy to guide automatic analyses of Pap smear slides. Top-down factors for saliency prediction on cell images increases the accuracy of the estimated maps while bottom-up algorithms proved to be useful for predicting the cytopathologist's eye fixations depending on parameters, such as the number of false positive and negative. Our contributions are: comparison among 13 state-of-the-art saliency models to cytopathologists' visual attention and deliver a method that the associate the most conspicuous regions to clinically relevant cells.

Does effective gaze behavior lead to enhanced performance in a complex error-detection cockpit task?

The purpose of the current study was to examine the relationship between expertise, performance, and gaze behavior in a complex error-detection cockpit task. Twenty-four pilots and 26 non-pilots viewed video-clips from a pilot's viewpoint and were asked to detect malfunctions in the cockpit instrument panel. Compared to non-pilots, pilots detected more malfunctioning instruments, had shorter dwell times on the instruments, made more transitions, visited task-relevant areas more often, and dwelled longer on the areas between the instruments. These results provide evidence for three theories that explain underlying processes for expert performance: The long-term working memory theory, the information-reduction hypothesis, and the holistic model of image perception. In addition, the results for generic attentional skills indicated a higher capability to switch between global and local information processing in pilots compared to non-pilots. Taken together, the results suggest that gaze behavior as well as other generic skills may provide important information concerning underlying processes that can explain successful performance during flight in expert pilots.