Abstract
Even expert radiologists will sometimes fail to detect the presence of a pulmonary nodule in a chest X-ray image, with estimated rates of missed detection of 20–30%. The challenging nature of this real-world search task lies not only in the visual contrast or the size of the nodule, but also in the heterogeneity of nodule appearance and the variability of the local anatomical background. For this study, we developed image processing software to create hundreds of visually realistic simulated nodules to investigate the visual-cognitive bases of nodule detection and the impact of extended training. First, we tested radiologist participants (n=10) with both real and computer-simulated nodules at a challenging nodule localization task. Performance accuracy was significantly better for real nodules than for the subtle simulated nodules that were created (70.5% vs. 59.0% accuracy, p < 0.005). Of greater interest, radiologists performed no greater than chance level at discriminating whether nodules were real or computer-simulated (mean accuracy, 52.9%). Next, we evaluated the impact of training naive undergraduate participants at a localization task involving simulated nodules. After 3–4 training sessions with 600 simulated cases, we observed significant improvements in performance for both simulated nodules (30.3% accuracy pre-test, 78.2% accuracy post-test, p < 0.00001) and real nodules (37.5% pre-test, 62.5% accuracy post-test, p < 0.0005.). In a final study, we found that undergraduates who had extended training with either light or dark polarity nodules showed much better subsequent performance at localizing nodules of the corresponding trained polarity. Our results demonstrate that marked improvements in nodule detection can be achieved by implementing a training regimen with numerous realistic examples; moreover, such training leads to learning of a polarity-specific perceptual template of nodule appearance. These findings suggest promising avenues for the development of a learning-based paradigm to facilitate the training of future radiologists.
Acknowledgement: R01EY029278