Classification of Lentigo Maligna at Patient-Level by Means of Reflectance Confocal Microscopy Data

Abstract

Reflectance confocal microscopy is an appropriate tool for the diagnosis of lentigo maligna. Compared with dermoscopy, this device can provide abundant information as a mosaic and/or a stack of images. In this particular context, the number of images per patient varied between 2 and 833 images and the objective, ultimately, is to be able to discern between benign and malignant classes. First, this paper evaluated classification at the image level, with the help of handcrafted methods derived from the literature and transfer learning methods. The transfer learning feature extraction methods outperformed the handcrafted feature extraction methods from literature, with a F 1 score value of 0.82. Secondly, this work proposed patient-level supervised methods based on image decisions and a comparison of these with multi-instance learning methods. This study achieved comparable results to those of the dermatologists, with an auc score of 0.87 for supervised patient diagnosis and an auc score of 0.88 for multi-instance learning patient diagnosis. According to these results, computer-aided diagnosis methods presented in this paper could be easily used in a clinical context to save time or confirm a diagnosis and can be oriented to detect images of interest. Also, this methodology can be used to serve future works based on multimodality.

A Clinical Perspective on the Automated Analysis of Reflectance Confocal Microscopy in Dermatology

“…Although the study was performed on a limited number of patients ( n = 5 in each group), the results showed promise for developing quantitative metrics for melanoma diagnosis based on RCM images. For lentigo maligna, RCM images of 135 malignant tumors (115 lentigo maligna or lentigo maligna melanoma and 20 basal cell carcinoma) and 88 benign tumors were analyzed using a series of extraction methods, textural interpretations, and machine learning algorithms [62]. All tested algorithms achieved sensitivities and specificities greater than 75% [62].…”

“…For lentigo maligna, RCM images of 135 malignant tumors (115 lentigo maligna or lentigo maligna melanoma and 20 basal cell carcinoma) and 88 benign tumors were analyzed using a series of extraction methods, textural interpretations, and machine learning algorithms [62]. All tested algorithms achieved sensitivities and specificities greater than 75% [62].…”

“…Texton-based algorithm [43] Texton-based algorithm [47][48][49][50][51][52] Computation of DEJ roughness [61] Transformation analysis [67] Logistic regression [24] Conditional random fields [54,55] Classification and Regression Trees [58,62] Support vector machine [41,68] Support vector machine [25] Spatial Poisson process [53] Convolutional neural network [60,62,65] Random Forest Classifier [69] Hybrid deep learning [26] Support vector machine [47][48][49][50][51][52] Support vector machine [59,62] 3D reconstruction [70] Convolutional neural network [12,44] Recurrent neural network [56,57] Recurrent neural network plus Convolutional neural network [56,57] Tree-of-shapes [71] Wavelet transform [46] Bayesian Model [63] Wavelet decomposition with support vector machine [64] articles that analyze automatic image processing of RCM images. Technical details of each algorithm were not analyzed in detail, as the goal of this review was to give clinicians an idea of how RCM image interpretation could be applied to the clinical setting using developed software mechanisms.…”

Super‐high magnification dermoscopy can aid the differential diagnosis between melanoma and atypical naevi

“…It only included pigmented lesions that were difficult to diagnose using conventional dermoscopy, as demonstrated by the low specificity of the seven‐point checklist, and it is possible that the inclusion of all pigmented melanocytic lesions could increase the list of differences between benign and malignant melanocytic tumours under D400. In the near future, it is possible that clinical diagnosis of MM will be performed with the aid of artificial intelligence, and it might be important to have as much information as possible such as clinical data, D20 and D400 dermoscopic images, and RCM images to improve the diagnostic accuracy of the neural networks 4,11 …”

Improving dermal level images from reflectance confocal microscopy using wavelet‐based transformations and adaptive histogram equalization

“…[10][11][12] A few studies have analyzed wavelets for dermatological applications, including image processing, machine learning, and diagnostic classification. [13][14][15][16][17][18][19] Several types of wavelets are employed for noise reduction and contrast enhancement; we elected to use fractional wavelet transforms (FRWT), which are powerful algorithms enabling detail preservation and structure enhancement through iterative signal expansion and shrinking. They are superior to other methods for noise reduction due to their ability to separate signals from noise and edge detection mimicking human vision.…”