Bringing light to the reading room

According to the American College of Radiology (ACR) Technical Standard Guidelines, there is a recommended illuminance level for reading rooms set between 25 and 75 lux. However, these guidelines are based on studies completed in 2007-2009, which predate significant advancements in display technology. No recent study has re-evaluated these lighting conditions, and more specifically, no prior studies have examined the effects of ambient lighting on the detection of intracranial hemorrhage on non-contrast brain CT.

To determine if higher ambient lighting levels (> 400 lux) significantly effects the diagnostic accuracy of board-certified neuroradiologists in detecting intracranial hemorrhage on brain CT, compared with standard low-light reading room conditions (20-70 lux).

Forty-two standard-of-care non-contrast head CT scans (axial acquisitions with coronal/sagittal reformats) from a tertiary-care hospital were selected by two independent neuroradiologists. These neuroradiologists established the presence or absence of intracranial hemorrhage. Four additional board-certified neuroradiologists independently interpreted all cases twice: first in a brightly lit environment (> 400 lux, measured at desktop level) and, after a > 6-month washout period to minimize recognition memory bias, in a standard reading room (20-70 lux). Interpretations were performed on calibrated Barco MDNC-3421 displays using the institutional Fuji PACS. Sensitivity, specificity, and accuracy were calculated for each condition and compared using paired two-tailed t-tests with Bonferroni correction for multiple comparisons.

No statistically significant differences were found between bright and dark conditions for sensitivity (0.974 vs. 0.974, p = 1.00), specificity (0.978 vs. 1.0, p = 0.49 adjusted), or overall accuracy (0.976 vs. 0.988, p = 1.00 adjusted).

Increased ambient lighting levels (> 400 lux) do not significantly affect the ability of board-certified neuroradiologists to accurately detect intracranial hemorrhage on modern high-resolution displays. These findings suggest that current guidelines should be re-evaluated with recent improvements in technology. Larger multi-center studies are warranted to confirm these results across additional pathologies and modalities.