Improving diagnostic sensitivity of combined dermoscopy and reflectance confocal microscopy imaging through double reader concordance evaluation in telemedicine settings: A retrospective study of 1000 equivocal cases

Abstract

BackgroundReflectance confocal microscopy (RCM) is an imaging device that permits non-invasive visualization of cellular morphology and has been shown to improve diagnostic accuracy of dermoscopically equivocal cutaneous lesions. The application of double reader concordance evaluation of dermoscopy-RCM image sets in retrospective settings and its potential application to telemedicine evaluation has not been tested in a large study population.ObjectiveTo improve diagnostic sensitivity of RCM image diagnosis using a double reader concordance evaluation approach; to reduce mismanagement of equivocal cutaneous lesions in retrospective consultation and telemedicine settings.Methods1000 combined dermoscopy-RCM image sets were evaluated in blind by 10 readers with advanced training and internship in dermoscopy and RCM evaluation. We compared sensitivity and specificity of single reader evaluation versus double reader concordance evaluation as well as the effect of diagnostic confidence on lesion management in a retrospective setting.ResultsSingle reader evaluation resulted in an overall sensitivity of 95.2% and specificity of 76.3%, with misdiagnosis of 8 melanomas, 4 basal cell carcinomas and 2 squamous cell carcinomas. Combined double reader evaluation resulted in an overall sensitivity of 98.3% and specificity of 65.5%, with misdiagnosis of 1 in-situ melanoma and 2 basal cell carcinomas.ConclusionEvaluation of dermoscopy-RCM image sets of cutaneous lesions by single reader evaluation in retrospective settings is limited by sensitivity levels that may result in potential mismanagement of malignant lesions. Double reader blind concordance evaluation may improve the sensitivity of diagnosis and management safety. The use of a second check can be implemented in telemedicine settings where expert consultation and second opinions may be required.

On the spectral signature of melanoma: a non-parametric classification framework for cancer detection in hyperspectral imaging of melanocytic lesions

“…It can also minimize false negative rates, obtaining a minimum false positive rate of 9.5%. These statistics are either comparable or an improvement over other state-of-the-art algorithms for spectral images of nevi [22,25,27,31,33,38,40], while its theoretical background simplifies its usage, as it has but one degree of freedom in its operation. In order to avoid complicating the clinical translation of this methodology, we seek to adapt the method to the dermatologists' needs.…”

“…For every patient, a concise protocol was followed, for which about 30 minutes were required per lesion, of which up to 2 minutes were used for long-exposure acquisition [22]. Nevi were inspected by dermatologists, whereas melanomas also underwent histological examination to prove their malignancy, as usual in these types of clinical studies [31,40]. The images vary in spatial resolution, but are consistent in spectral resolution per pixel.…”

Accuracy of tele‐consultation on management decisions of lesions suspect for melanoma using reflectance confocal microscopy as a stand‐alone diagnostic tool

“…retrospectively evaluated 100 equivocal cases with combined dermoscopy‐RCM image sets and found single‐reader sensitivity of 89% and specificity of 66%. Witkowski et al . had 10 readers retrospectively evaluate 1000 combined dermoscopy‐RCM image sets and found single‐reader sensitivity of 95% and specificity of 76%.…”

Clinical noninvasive imaging and spectroscopic tools for dermatological applications: Review of recent progress

“…Notably, experience in feature identification plays a role in achieving an appropriate level of diagnostic accuracy. In fact, it has been reported a minimum of 6 months training, with daily clinical exposure for at least 3 months to RCM and to at least 2000 RCM cases were needed for accurate clinical decision making [72]. Interobserver variations in recognizing RCM features skin tumor lesions were investigated in a study involving nine dermatologists evaluating 100 RCM data sets of melanocytic tumors, epithelial skin cancers and benign nonmelanocytic lesions [73].…”