Evaluation of students’ gaze patterns, diagnosis speed and diagnosis accuracy when interpreting clinical findings / Noor Halilah Buari... [et al.]

Time and speed are vital aspects of clinical diagnosis decision-making. This study aimed to investigate the gaze patterns, diagnosis speed and accuracy with and without the assistance of clinical history while interpreting clinical findings. This cross-sectional study employed convenience sampling to recruit 28 normally sighted final year students with ongoing clinical training. Each student was shown six clinical findings, half of which accompanied a brief clinical history in prose, and the other half were not. First, the clinical history was presented to be read by the participants, followed by providing clinical findings regarding fundus picture images. The participants were asked to make a diagnosis based on a clinical finding presented to them. The Dikablis eye tracker was used to record and track the gaze patterns during the treatment. The assessment had no time restriction, and the gaze patterns (number of fixations, fixation duration, number of saccadic, and saccadic angles) were retrieved from the D-Lab software. Diagnostic speed was calculated based on the time taken for the students to provide a clinical diagnosis. Diagnosis accuracy was the score of correct or incorrect of the given diagnosis. Comparison of gaze patterns in interpreting clinical findings with clinical history and without clinical history showed no statistically significant difference for all gaze patterns including the number of fixation (p=0.20), fixation durations (p=0.98), number of saccadic (p=0.33) and saccadic angle (p=0.77). There was also no statistically significant difference in both diagnosis accuracy (p=0.14) and diagnosis speed (p= 0.20) between both conditions. However, there was a strong correlation between the number of fixations and diagnosis speed with (r = 0.708, p < 0.05) and without (r = 0.618, p < 0.05) clinical history. A moderate correlation was found between the number of saccades and diagnosis speed with (r = 0.578, p < 0.05) and without (r = 0.424, p < 0.05) clinical history. In conclusion, a brief clinical history does not appear to influence the gaze patterns, diagnosis speed and accuracy in evaluating the clinical findings. However, the gaze patterns highly correlated with the diagnosis speed in clinical decision-making. These findings indicate cognitive processing during clinical decision-making, which might benefit clinical educators in enhancing the clinical teaching approach and quality.