Visual interpretation of Pediatric Panoramic Radiographs: An Eye-Tracking Mixed-Method Study

Abstract

Objectives: This study aimed to (1) investigate the eye-tracking (ET) performance of postgraduate  students (PGs) and dental interns in detecting abnormalities in pediatric panoramic radiographs, (2) explore the association between different ET measurements and students’ odds of detecting abnormalities, and (3) assess the participants’ experiences with utilizing eye-tracking technology (ETT) as a learning aid. Methods: Participants consisted of 30 pediatric PGs and 32 dental interns at King Abdulaziz University Dental Hospital. Nine pediatric panoramic radiographs were used to develop the test. A RED-m® SMI ET system was used to track the participant’s gaze patterns, and abnormalities were identified using BeGaze software. The parameters extracted from the software were entry time, dwell time, fixation time, fixation count, and revisit count. Univariate, bivariate, and multivariable analyses were conducted. Semi-structured interviews were analyzed and categorized into themes using NVivo 11. Results: PGs displayed a higher proportion of correctly identified abnormal radiographs compared to dental interns (p = 0.003). PGs had significantly more revisits (p = 0.003) for each area of interest (AOI) than dental interns. With every unit increase in revisits, the odds of correct detection increased by 1.17 (p = 0.009). The interview data analysis uncovered five themes: ETT experience, challenges encountered, ETT as an educational tool, anticipated improvements, and final recommendations. Conclusion: Using an ET machine, pediatric PGs demonstrated a higher likelihood of correctly identifying abnormal radiographs and significantly more frequent revisits to AOIs than dental interns. Clinical significance: This study objectively assessed the process of detecting pediatric dental radiographic lesions among two different educational levels and provided insight into the use of ETT as an innovative pedagogical aid in dental education.