High-resolution computer display of portable, digital, chest radiographs of adults: suitability for primary interpretation.

Abstract

The suitability of using a high-resolution computer-display system for primary interpretation of adult chest radiographs obtained with portable apparatus and storage phosphor technology was evaluated with receiver operating characteristic analysis and subjective methods. Sixty portable digital chest radiographs showing abnormalities that approached the resolution limits of our computed radiography system were selected. The original digital data making up the 2140 x 1760 x 10 bit images were transferred to a cathode ray tube (CRT) display system composed of two Megascan monitors. Postprocessing rendered two images, one for each monitor, to emulate the two-on-one hard-copy format produced by the computed radiography system. Each image set was interpreted independently by three radiologists experienced with hard-copy format. Limited level and window adjustments were allowed during CRT interpretation. For both CRT and hard-copy display, images were graded on the basis of the interpreter's confidence in the presence of pneumothorax and parenchymal abnormalities. Three types of foreign devices were subjectively analyzed: endotracheal tubes, nasogastric tubes, and temporary epicardial pacer wires. No significant differences were found in detecting pneumothorax or focal infiltrates. One radiologist found the hard copy better for detecting diffuse infiltrates (p = .02); two radiologists favored CRT for visualizing nasogastric tubes (p < .005, p < .02); and one radiologist favored CRT for visualizing temporary epicardial pacer wires (p = .05). We conclude that an optimized high-resolution CRT system is quite promising for primary interpretation of digital portable chest radiographs, but further investigation and greater statistical power are necessary to confirm our results.