Effect of display type and room illuminance in chest radiographs

Objectives To compare diagnostic accuracy in the detection of subtle chest lesions on digital chest radiographs using medical-grade displays, consumer-grade displays, and tablet devices under bright and dim ambient light. Methods Five experienced radiologists independently assessed 50 chest radiographs (32 with subtle pulmonary findings and 18 without apparent findings) under bright (510 lx) and dim (16 lx) ambient lighting. Computed tomography was used as the reference standard for interstitial and nodular lesions and follow-up chest radiograph for pneumothorax. Diagnostic accuracy and sensitivity were calculated for assessments carried out in all displays and compared using the McNemar test. The level of significance was set to p < 0.05. Results Significant differences in sensitivity between the assessments under bright and dim lighting were found among consumer-grade displays in interstitial opacities with, and in pneumothorax without, Digital Imaging and Communication in Medicine-Grayscale Standard Display Function (DICOM-GSDF) calibration. Compared to 6 megapixel (MP) display under bright lighting, sensitivity in pneumothorax was lower in the tablet device and the consumer-grade display. Sensitivity in interstitial opacities was lower in the DICOM-GSDF calibrated consumer-grade display. Conclusions A consumer-grade display with or without DICOM-GSDF calibration or a tablet device is not suitable for reading digital chest radiographs in bright lighting. No significant differences were observed between five displays in dim light. Key Points • Ambient lighting affects performance of consumer-grade displays (with or without DICOM-GSDF calibration). • Bright light decreases detection of pneumothorax on non-medical displays. • Bright light decreases detection of interstitial opacities on DICOM-GSDF-calibrated, consumer-grade displays. • Dim light is sufficient to detect subtle chest lesions from all displays.