Colors in atypical nevi: a computer description reproducing clinical assessment

Statistical Detection of Colors in Dermoscopic Images With a Texton-Based Estimation of Probabilities

Color has great diagnostic significance in dermatoscopy. Several diagnosis methods are based on the colors detected within a lesion. Malignant lesions frequently show more than three colors, whereas in benign lesions, three or fewer colors are usually observed. Black, red, white, and blue-gray are found more frequently in melanomas than in benign nevi. In this paper, a method to automatically identify the colors of a lesion is presented. A color label identification problem is proposed and solved by maximizing the posterior probability of a pixel to belong to a label, given its color value and the neighborhood color values. The main contribution of this paper is the estimation of the different terms involved in the computation of this probability. Two evaluations are performed on a database of 200 dermoscopic images. The first one evaluates if all the colors detected in a lesion are indeed present in it. The second analyzes if each pixel within a lesion is assigned the correct color label. The results show that the proposed method performs correctly and outperforms other methods, with an average F-measure of 0.89, an accuracy of 0.90, and a Spearman correlation of 0.831.

Incorporating Colour Information for Computer-Aided Diagnosis of Melanoma from Dermoscopy Images: A Retrospective Survey and Critical Analysis

Cutaneous melanoma is the most life-threatening form of skin cancer. Although advanced melanoma is often considered as incurable, if detected and excised early, the prognosis is promising. Today, clinicians use computer vision in an increasing number of applications to aid early detection of melanoma through dermatological image analysis (dermoscopy images, in particular). Colour assessment is essential for the clinical diagnosis of skin cancers. Due to this diagnostic importance, many studies have either focused on or employed colour features as a constituent part of their skin lesion analysis systems. These studies range from using low-level colour features, such as simple statistical measures of colours occurring in the lesion, to availing themselves of high-level semantic features such as the presence of blue-white veil, globules, or colour variegation in the lesion. This paper provides a retrospective survey and critical analysis of contributions in this research direction.

Predominant Dermoscopic Patterns Observed among Nevi

Background:

It has been clinically observed that patients' “normal” moles resemble each other. Whether this concept is applicable to dermoscopic practice has not been sufficiently studied.

Objective:

To investigate whether physicians evaluating dermoscopic images would identify common dermoscopic profiles of nevi within individual patients.

Methods:

Images of 205 nevi belonging to 18 patients were evaluated by 2 dermatologists for dermoscopic global pattern, color, and specific structures. We defined dermoscopic patterns as dominant if seen in ≥ 40% of the patient's nevi; a minor pattern was defined as 20 to 39%.

Results:

A dominant pattern was seen in 15 patients (83%). In 13 (72%) of the patients, ≥ 80% of the nevi were classified into one, two, or three global patterns. The reticular global pattern was the most prevalent dominant pattern, seen in 9 patients (50%); the homogeneous pattern was the most prevalent minor pattern, seen in 16 patients (89%).

Conclusion:

Individuals tend to have one to three predominant dermoscopic nevus global patterns.

Grey-blue regression in melanoma in situ-evaluation on 111 cases

As fibrosis and melanosis are often seen in malignant melanoma, the presence of dermoscopic signs of regression may represent a clue for the diagnosis of malignancy. Our aim was to assess the frequency and extent of 11 dermoscopic features of regression evaluating dermoscopic images of 111 melanomas in situ (MIS). Regression structures (grey-blue areas, white areas, peppering, and/or blue-whitish veil) were present in 80.1% of the lesions. Approximately 80% of the lesions showed regression of dermoscopic structures and light brown areas. Most lesions showed the presence of grey-blue areas (74.7%), whereas peppering was observable in 30.6% of MIS. Areas of fibrosis were mainly observable as structureless areas with a pinkish hue (50.4%). Based on our data, the reticular pattern of blue regression and light brown areas can be considered a significant discriminator and a reliable predictor of MIS.

Concentric decile segmentation of white and hypopigmented areas in dermoscopy images of skin lesions allows discrimination of malignant melanoma

Dermoscopy, also known as dermatoscopy or epiluminescence microscopy (ELM), permits visualization of features of pigmented melanocytic neoplasms that are not discernable by examination with the naked eye. White areas, prominent in early malignant melanoma and melanoma in situ, contribute to early detection of these lesions. An adaptive detection method has been investigated to identify white and hypopigmented areas based on lesion histogram statistics. Using the Euclidean distance transform, the lesion is segmented in concentric deciles. Overlays of the white areas on the lesion deciles are determined. Calculated features of automatically detected white areas include lesion decile ratios, normalized number of white areas, absolute and relative size of largest white area, relative size of all white areas, and white area eccentricity, dispersion, and irregularity. Using a back-propagation neural network, the white area statistics yield over 95% diagnostic accuracy of melanomas from benign nevi. White and hypopigmented areas in melanomas tend to be central or paracentral. The four most powerful features on multivariate analysis are lesion decile ratios. Automatic detection of white and hypopigmented areas in melanoma can be accomplished using lesion statistics. A neural network can achieve good discrimination of melanomas from benign nevi using these areas. Lesion decile ratios are useful white area features.

Reticular grey-blue areas of regression as a dermoscopic marker of melanoma in situ

BACKGROUND

By dermoscopy, regression structures are substantially defined by the presence of white and blue areas in the lesion image. As fibrosis and melanosis are often seen in malignant melanoma (MM), the presence of dermoscopic signs of regression may represent a clue for the diagnosis of malignancy.

OBJECTIVES

To assess the frequency and extent of dermoscopic signs of regression in melanoma in situ (MIS) and to describe its dermoscopic features.

METHODS

Dermoscopic images of 85 MIS, 85 invasive MMs and 85 dermoscopically equivocal lesions with a histological diagnosis of naevus were evaluated by three dermatologists, who assessed the presence of 11 parameters of regression.

RESULTS

The number of regression parameters per lesion increased according to melanoma thickness. White areas, the grey-blue veil and widespread blue areas were more frequent in invasive MMs than in the other two lesion groups, whereas light brown areas and regression of dermoscopic structures were more frequent in MIS. Peppering was observable in the same percentage of MIS and invasive MMs. Blue areas were more frequently structureless in equivocal lesions and invasive MMs, whereas the reticular pattern prevailed in MIS.

CONCLUSIONS

Frequency, morphology, extent and distribution of regression vary according to melanoma thickness and diameter. Lesions with reticular blue regression and light brown areas should undergo surgical excision for the suspicion of MIS. Moreover, the identification of the reticular pattern of blue regression can be considered a significant discriminator and a reliable predictor of MIS.

A retrospective study addressed to understanding what predicts severe histological dysplasia/early melanoma in excised atypical melanocytic lesions

BACKGROUND

Atypical naevi are common benign skin lesions but are also recognized both as precursors of and risk factors for melanoma. It is therefore imperative to excise those lesions that are either likely to progress or are already progressing to melanoma. Clinically, however, it may be difficult to distinguish these from benign atypical naevi with bland histology.

OBJECTIVES

To analyse the clinical characteristics of excised melanocytic lesions and to identify the predictors of severe histological atypia/melanoma in situ and invasive melanoma.

METHODS

The case notes of 434 patients who had melanocytic lesions removed at a pigmented lesion clinic were studied retrospectively. A single pathologist reviewed the excised lesions and clinical characteristics predictive of malignancy were identified.

RESULTS

The best predictors of melanoma were older age, history of change and site on an extremity, but only older age was predictive of severe histological atypia/melanoma in situ as opposed to mild to moderate atypical histology.

CONCLUSIONS

These results confirm the difficulty of differentiating accurately between benign atypical naevi and borderline lesions or early melanoma in a clinical setting. It is therefore necessary to have a sufficiently low threshold for excision to avoid missing early melanomas, particularly in older patients presenting with lesions on the extremities.

Diving into the blue: In vivo microscopic characterization of the dermoscopic blue hue

BACKGROUND

In dermoscopy the presence of a blue hue is a clue for malignancy, although a blue tint is sometimes observable in benign lesions.

OBJECTIVE

To identify the in vivo confocal microscopy correlates of the blue hue for improving diagnostic accuracy for melanoma.

METHODS

Fifty-seven melanomas, 41 junctional, 88 compound, and 27 Spitz nevi were studied by dermoscopy, confocal microscopy, and histopathology.

RESULTS

Confocal microscopy enabled the distinction between blue areas and blue veil, the former characterized by plump cells corresponding to melanophages and inflammatory infiltrate at histology, the latter by the contemporary presence of epidermal and dermal features consistent with diagnosis of melanoma, such as disarranged pattern, pagetoid cells, cytologic and architectural atypias, nonhomogeneous and cerebriform clusters, and dermal nucleated cells.

LIMITATIONS

Confocal microscopy failed to accurately distinguish Spitz nevi, because of the presence of cytoarchitectural disarray in the epidermis and the upper dermis.

CONCLUSION

Confocal microscopy enabled the in vivo identification of characteristic cytological substrates correlated with the blue features in dermoscopy.

Algorithmic reproduction of asymmetry and border cut-off parameters according to the ABCD rule for dermoscopy

BACKGROUND

Semiquantitative algorithms were applied to dermoscopic images to improve the clinical diagnosis for melanoma.

OBJECTIVE

The aim of the study was to develop a computerized method for automated quantification of the 'A' (asymmetry) and 'B' (border cut-off) parameters, according to the ABCD rule for dermoscopy, thus reproducing human evaluation.

METHODS

Three hundred and thirty-one melanocytic lesion images, referring to 113 melanomas and 218 melanocytic nevi, acquired by means of a digital videodermatoscope, were considered. Images were evaluated by two experienced observers and by using computer algorithms developed by us. Clinical evaluation of asymmetry was performed by attributing scores to shape asymmetry and asymmetry of pigment distribution and structures, whereas computer evaluation of shape and pigment distribution asymmetries were based on the assessment of differences in area and lightness in the two halves of the image, respectively. Borders were evaluated both by clinicians and by the computer, by attributing a score to each border segment ending abruptly. Differences between nevus and melanoma values were evaluated using the chi-square test, while Cohen's Kappa index for agreement was employed for the evaluation of the concordance between human and computer.

RESULTS

Pigment distribution asymmetry appears the most striking parameter for melanoma diagnosis both for human and for automated diagnosis. A good concordance between clinicians and computer evaluation was achieved for all asymmetry parameters, and was excellent for border cut-off evaluation.

CONCLUSIONS

These algorithms enable a good reproduction of the 'A' and 'B' parameters of the ABCD rule for dermoscopy, and appear useful for diagnostic and learning purposes.