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

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

OBJECTIVES

Reflectance confocal microscopy (RCM) generates scalar image data from serial depths in the skin, allowing in vivo examination of cellular features. The maximum imaging depth of RCM is approximately 250 µm, to the papillary dermis, or upper reticular dermis. Frequently, important diagnostic features are present in the dermis, hence improved visualization of deeper levels is advantageous.

METHODS

Low contrast and noise in dermal images were improved by employing a combination of wavelet-based transformations and contrast-limited adaptive histogram equalization.

RESULTS

Preserved details, noise reduction, increased contrast, and feature enhancement were observed in the resulting processed images.

CONCLUSIONS

Complex and combined wavelet-based enhancement approaches for dermal level images yielded reconstructions of higher quality than less sophisticated histogram-based strategies. Image optimization may improve the diagnostic accuracy of RCM, especially for entities with dermal findings.

Classification of Wideband Tympanometry by Deep Transfer Learning with Data Augmentation for Automatic Diagnosis of Otosclerosis

Otosclerosis is a common disease of the middle ear leading to stapedial fixation. Its rapid and non-invasive diagnosis could be achieved through wideband tympanometry (WBT), but the interpretation of the raw data provided by this tool is complex and time-consuming. Convolutional neural networks (CNN) could potentially be applied to this situation to help the clinicians categorize WBT data. A dataset containing 135 samples from 80 patients with otosclerosis and 55 controls was obtained. We designed a lightweight CNN to categorize samples into the otosclerosis and control. Receiver operating characteristic (ROC) analysis showed an area under the curve (AUC) of 0.95 0.011, and the F1-score was 0.89 0.031 (r=10). The performance was further improved by data augmentation schemes and transfer learning strategies (AUC: 0.97 0.010, F1-score: 0.94 0.016, p<0.05, ANOVA). Finally, the most relevant diagnostic features employed by the CNN were assessed via the activation pattern heatmaps. These results are crucial for the visual interpretation of WBT graphic outputs which clinicians use in routine, and for a better understanding of the WBT signal in relation to the ossicular mechanics.

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

BACKGROUND

Dermoscopy is the most widely used noninvasive imaging technique for the clinical diagnosis of melanoma (MM). Super-high (× 400) magnification dermoscopy (D400) has recently been developed; compared with traditional dermoscopy, it can reveal additional features, down to the identification of single melanocytes in the skin.

OBJECTIVES

To evaluate which structures are visible at D400 and to compare them in atypical naevi and MMs.

METHODS

A prospective observational multicentre study was conducted. We enrolled patients who were identified as having atypical melanocytic skin lesions by clinical and/or × 20 magnification dermoscopy (D20) examination, and who were assigned to either excision or follow-up. Lesions were imaged by videodermoscopy at D20 and D400. The presence of pigmented cells and their features were assessed at D400.

RESULTS

In total, there were 79 patients with 57 naevi and 31 MMs. Of the total 88 lesions, 63 (71.6%) were given a histological diagnosis, while the others were followed up for ≥ 12 months, during which they showed no change and were all diagnosed as naevi. Pigmented cells were identified in > 90% of the lesions at D400. Compared with naevi, MMs had a higher frequency of scattered, large, irregular (in shape and size), dendritic/roundish, violet/blue pigmented cells under D400 (P < 0.001). Moreover, dots (P < 0.01), out-of-focus blue structureless areas (P < 0.01) and vessels (P < 0.001) were also more frequent in MMs than in naevi at D400.

CONCLUSIONS

This study showed that D400 can reveal many elements not otherwise visible in traditional D20 dermoscopy, such as pigmented cells and their morphology, which could be useful for the diagnosis of MM.

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

BACKGROUND AND OBJECTIVES

Non-invasive optical imaging has the potential to provide a diagnosis without the need for biopsy. One such technology is reflectance confocal microscopy (RCM), which uses low power, near-infrared laser light to enable real-time in vivo visualization of superficial human skin from the epidermis down to the papillary dermis. Although RCM has great potential as a diagnostic tool, there is a need for the development of reliable image analysis programs, as acquired grayscale images can be difficult and time-consuming to visually assess. The purpose of this review is to provide a clinical perspective on the current state of artificial intelligence (AI) for the analysis and diagnostic utility of RCM imaging.

STUDY DESIGN/MATERIALS AND METHODS

A systematic PubMed search was conducted with additional relevant literature obtained from reference lists.

RESULTS

Algorithms used for skin stratification, classification of pigmented lesions, and the quantification of photoaging were reviewed. Image segmentation, statistical methods, and machine learning techniques are among the most common methods used to analyze RCM image stacks. The poor visual contrast within RCM images and difficulty navigating image stacks were mediated by machine learning algorithms, which allowed the identification of specific skin layers.

CONCLUSIONS

AI analysis of RCM images has the potential to increase the clinical utility of this emerging technology. A number of different techniques have been utilized but further refinements are necessary to allow consistent accurate assessments for diagnosis. The automated detection of skin cancers requires more development, but future applications are truly boundless, and it is compelling to envision the role that AI will have in the practice of dermatology. Lasers Surg. Med. © 2020 Wiley Periodicals LLC.