Abstract
One of the most challenging problems in clinical dermatology is the early detection of melanoma. Reflectance confocal microscopy (RCM) is an added tool to dermoscopy improving considerably diagnostic accuracy. However, diagnosis strongly depends on the experience of physicians. High-definition optical coherence tomography (HD-OCT) appears to offer additional structural and cellular information on melanocytic lesions complementary to that of RCM. However, the diagnostic potential of HD-OCT seems to be not high enough for ruling out the diagnosis of melanoma if based on morphology analysis. The aim of this paper is first to quantify in vivo optical properties such as light attenuation in melanocytic lesions by HD-OCT. The second objective is to determine the best critical value of these optical properties for melanoma diagnosis. The technique of semi-log plot whereby an exponential function becomes a straight line has been implemented on HD-OCT signals coming from four successive skin layers (epidermis, upper papillary dermis, deeper papillary dermis and superficial reticular dermis). This permitted the HD-OCT in vivo measurement of skin entrance signal (SES), relative attenuation factor normalized for the skin entrance signal (µraf1) and half value layer (z1/2). The diagnostic accuracy of HD-OCT for melanoma detection based on the optical properties, µraf1, SES and z1/2 was high (95.6, 82.2 and 88.9 %, respectively). High negative predictive values could be found for these optical properties (96.7, 89.3 and 96.3 %, respectively) compared to morphologic assessment alone (89.9 %), reducing the risk of mistreating a malignant lesion to a more acceptable level (3.3 % instead of 11.1 %). HD-OCT seems to enable the combination of in vivo morphological analysis of cellular and 3-D micro-architectural structures with in vivo analysis of optical properties of tissue scatterers in melanocytic lesions. In vivo HD-OCT analysis of optical properties permits melanoma diagnosis with higher accuracy than in vivo HD-OCT analysis of morphology alone.
Highlights
One of the most challenging problems in clinical dermatology is the early detection of melanoma [4, 42, 46]
Non-invasive imaging techniques have been introduced to improve the early detection of melanoma which can be often challenging with the naked eye alone [10, 17]
Almost no melanomas were misclassified and undertreated when both techniques were used in combination: sensitivity 97.82 %, specificity 92.44 %, positive predictive value (PPV) 87.37 % and negative predictive value (NPV) 98.75 % [1]
Summary
One of the most challenging problems in clinical dermatology is the early detection of melanoma [4, 42, 46]. Since clinical diagnosis may be difficult, non-invasive imaging techniques have been developed to enhance early diagnosis in challenging cases [10, 17]. Dermoscopy and reflectance confocal microscopy (RCM) are both able to considerably improve the diagnostic accuracy for melanoma, especially when used synergistically [11, 32, 33, 48]. New diagnostic tools providing automated classification of pigmented skin lesions usable by non-experts have been. Spectral methods fall into this class of emerging new techniques holding the promise to provide quantitative criteria for melanoma diagnosis and to improve early diagnosis [37]. Multispectral information can be assessed both in the spatial domain (multispectral digital dermoscopy [22]) or in the frequency domain (spectroscopic methods such as diffuse-reflectance spectroscopy [21, 41, 54], Raman spectroscopy [36, 52] and fluorescence spectroscopy [15, 34])
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
