The influence of computer-aided polyp detection systems on reaction time for polyp detection and eye gaze

Abstract

Background and study aims: Multiple computer-aided systems for polyp detection (CADe) are currently introduced into clinical practice, with an unclear effect on examiner behavior. In this work, we aimed to measure the influence of a CADe system on reaction time, mucosa misinterpretations, and changes in visual gaze pattern. Patients and methods: Participants with variable levels of experience in colonoscopy examined video sequences while eye movement was tracked. Using a crossover design, videos were presented in two assessments with and without CADe (GI Genius, Medtronic) support. Reaction time for polyp detection and eye-tracking metrics were evaluated. Results: 21 Participants performed 1218 experiments. CADe was with a median of 1.16sec significantly faster in detecting polyps compared to the users with 2.97sec (99%CI;0.40-3.43 and 2.53-3.77sec, respectively). However, the reaction time of the user with the use of CADe with a median of 2.9sec (99%CI;2.55-3.38sec) was similar than without its use. CADe increased the misinterpretations of normal mucosa and reduced the eye travel distance. Conclusions: This work confirms that CADe systems detect polyps faster than humans. In addition, they led to increased misinterpretations of normal mucosa and decreased eye travel distance. Possible consequences of these findings might be prolonged examination time and deskilling.

Development and evaluation of a deep learning model to improve the usability of polyp detection systems during interventions

“…Another possibility would be to investigate (e.g. by eye‐tracking) whether the examiner's attention could be better focused on the intervention by reducing distracting CADe signals 17 . Other limitations include that, to facilitate the generation of a large‐annotated training dataset in a short period of time, no predefined protocol was used for video selection.…”

“…whether the examiner's attention could be better focused on the intervention by reducing distracting CADe signals. 17 CADe systems already achieved a remarkable benefit in randomized controlled trials. Future developments of those systems include improving usability by adding customizable features.…”

“…by eye‐tracking) whether the examiner's attention could be better focused on the intervention by reducing distracting CADe signals. 17 Other limitations include that, to facilitate the generation of a large‐annotated training dataset in a short period of time, no predefined protocol was used for video selection. A quantitative identification of the causes of false positive detections of our CNN need to be evaluated in future studies.…”

Pilot study of a new freely available computer-aided polyp detection system in clinical practice

“…Therefore, the authors concluded that more data for non-Asian populations is necessary. Furthermore, examiners focus on the center of the endoscopic image and CADe systems improve detection in the image’s periphery [ 9 ]. Lastly, to our knowledge there is no data regarding usability and acceptance of CADe systems in clinical practice.…”

Targeting the low detector with artificial intelligence

“…In the second paper, Troya et al tracked the eye movement for polyp detection, analyzing its interaction with a computeraided polyp detection system (CADe) [2]. Novice endoscopists, which may approximate low detectors, showed a trend for a slower reaction time in polyp detection by nearly 1 second compared with experienced endoscopists.…”