Reduction in size of digital images: does it lead to less detectability or loss of diagnostic information?

Abstract

OBJECTIVES To analyse the effect of reduction in size of digital images on diagnostic outcome. METHODS A series of 100 Visualix III (Gendex Dental Systems, Milano, Italy; Dentsply, Des Plaines, IL, USA) images was made of size 10 and 15 endodontic files in upper and lower (pre)molars. Three forms of image were created: (a) the original images, (b) reduced to one-half (containing one-quarter of the information of the original, and (c) zoomed-in, half-size images (magnification 2:1; original image size but with only one-quarter of the original information). Seven radiologists were asked to rate the position of the tip of the file, using a five-point confidence scale. ROC data were analysed by means of MANOVA statistics. RESULTS A significant difference was found between the three image modalities for both size 10 (P < 0.005), and size 15 (P < 0.021) files. CONCLUSIONS Reduction in size of digital images may cause less detectability as well as loss of diagnostic information.