Comparison of dermatoscopic ABCD rule and risk stratification in the diagnosis of malignant melanoma

For didactic and documentation purposes the dermatoscopic ABCD rule and the dermatoscopic risk stratification have been proposed. The aim of this investigation was to compare the ability of the 2 methods to separate patients with cutaneous malignant melanoma from individuals with other pigmented skin lesions. Three dermatologists, experienced users of dermatoscopy, assessed macroscopic clinical and dermatoscopic slides from 258 patients referred to the skin cancer outpatient clinic by the ABCD rule and risk stratification methods. Diagnostic performance of the 2 methods was compared by receiver operating characteristics curve analysis. When all pigmented skin lesions were compared, there was a trend for the observers to perform better using risk stratification. When only lesions with a well-defined pigment network were included, the diagnostic performance of the risk stratification method was superior to the dermatoscopic ABCD rule (areas under the receiver operating characteristics curve median 0.93 vs. 0.80, p<0.004) for all observers. The agreement between the 2 methods was moderate to substantial (kappa coefficient 0.53-0.62). More melanomas were identified when the rules were combined. The dermatoscopic ABCD rule has been accepted as a standard for identifying melanomas with the dermatoscope, but should be considered secondary to pigment network analysis.     

Preoperative assessment of melanoma thickness by ABCD score of dermatoscopy

BACKGROUND: Epiluminescence microscopy is a useful tool for the noninvasive diagnosis of malignant melanoma based on criteria that have been correlated with specific histologic characteristics. Previous studies have also shown a good correlation between the frequency of some dermatoscopic criteria based on pattern analysis (pigment network, blue-gray areas, vascular pattern) and tumor thickness. This technique could be useful in the preoperative assessment of tumor thickness as an indication for sentinel node biopsy. OBJECTIVE: Our purpose was to evaluate the possible role of the total dermatoscopy score (TDS) assigned to a series of lesions in accordance with the ABCD rule of dermatoscopy as a preoperative predictor of melanoma thickness. METHODS: For 84 cutaneous melanomas (17 were in situ melanomas; 67 invasive cases with mean thickness 0. 93 mm; range, 0.2-3.9 mm; standard deviation, 0.63) consecutively excised at the Department of Dermatology of Florence, TDS was established by two observers blinded as to the tumor thickness. The performance of different cut-off points of TDS in the diagnosis of melanoma with Breslow's thickness more than 0.75 or 1.00 mm was investigated by receiver operating characteristic (ROC) analysis. Accuracy of classification was evaluated in terms of sensitivity, specificity, and area under ROC curves (AUC). RESULTS: The TDS assigned to cutaneous melanomas by the ABCD rule increased with the thickness of the lesions from in situ melanomas to melanomas of intermediate thickness (0.75-1.50 mm). As a preoperative method for the detection of melanomas with a Breslow thickness greater than 0. 75 mm, a TDS cut-off point of 6.80 showed 80% sensitivity, 84% specificity, and 82% diagnostic accuracy (AUC value, 0.90). TDS performed better in the diagnosis of a thickness threshold of 0.75 than 1.00 mm (higher AUC value). CONCLUSION: The TDS calculated by the ABCD rule of dermatoscopy-a simplified approach to dermatoscopic diagnosis of melanoma-provides useful information for the preoperative assessment of melanoma thickness greater than 0.75 mm.     

Morphologic changes of pigmented skin lesions: a useful extension of the ABCD rule for dermatoscopy

BACKGROUND: Epiluminescence microscopy (ELM) significantly increases the early diagnosis of pigmented skin lesions (PSL) using established criteria and pattern analysis. The ABCD rule for dermatoscopy (ie, ELM) provides a simplified approach to the interpretation of ELM images on the basis of asymmetry (A), border (B), color (C), and dermatoscopic structure (D). OBJECTIVE: We set out to determine whether the diagnostic accuracy of the ABCD scoring algorithm can be significantly improved by incorporating information about morphologic changes of the lesion observed and provided by the patient. METHODS: We prospectively collected 356 small pigmented skin lesions (< 1 cm) including 73 (20.5%) melanomas. Before excision all patients were asked whether the lesion had changed in size, color, or shape within the last year or whether they experienced any sign of ulceration or spontaneous bleeding. ELM images of the lesions were evaluated according to the ABCD rule for dermatoscopy to yield a semiquantitative score. Accuracy of diagnosis was evaluated in terms of sensitivity, specificity, and area under receiver operating characteristic curves (AUC). RESULTS: The frequency of reported changes was significantly higher for melanomas than benign PSL (65.8% vs 29.7%, P < .001). In a multivariate model morphologic change was a significant independent predictor of malignancy (odds ratio = 3.17, 95% confidence interval [CI]: 1.96 to 5.14, P < .001). The mean final score achieved when using the enhanced ABCD-E criteria including morphologic change (E) was significantly higher for melanomas (5.7, 95% CI: 5.3 to 6.0) than benign PSL (2.9, 95% CI: 2.8 to 3.1, P < .001). Diagnostic accuracy was significantly higher when the lesions were evaluated by the enhanced ABCD-E criteria as compared with the standard ABCD score (AUC(ABCD) = 0.87 vs AUC(ABCD-E) = 0.90; P = .006). CONCLUSION: Information about morphologic changes of PSL as reported by the patient is a useful extension of the ABCD rule for dermatoscopy.     

Comparison of dermatoscopic diagnostic algorithms based on calculation: The ABCD rule of dermatoscopy,the seven‐point checklist,the three‐point checklist and the CASH algorithm in dermatoscopic evaluation of melanocytic lesions

Dermatoscopic analysis of melanocytic lesions using the CASH algorithm has rarely been described in the literature. The purpose of this study was to compare the sensitivity, specificity, and diagnostic accuracy rates of the ABCD rule of dermatoscopy, the seven‐point checklist, the three‐point checklist, and the CASH algorithm in the diagnosis and dermatoscopic evaluation of melanocytic lesions on the hairy skin. One hundred and fifteen melanocytic lesions of 115 patients were examined retrospectively using dermatoscopic images and compared with the histopathologic diagnosis. Four dermatoscopic algorithms were carried out for all lesions. The ABCD rule of dermatoscopy showed sensitivity of 91.6%, specificity of 60.4%, and diagnostic accuracy of 66.9%. The seven‐point checklist showed sensitivity, specificity, and diagnostic accuracy of 87.5, 65.9, and 70.4%, respectively; the three‐point checklist 79.1, 62.6, 66%; and the CASH algorithm 91.6, 64.8, and 70.4%, respectively. To our knowledge, this is the first study that compares the sensitivity, specificity and diagnostic accuracy of the ABCD rule of dermatoscopy, the three‐point checklist, the seven‐point checklist, and the CASH algorithm for the diagnosis of melanocytic lesions on the hairy skin. In our study, the ABCD rule of dermatoscopy and the CASH algorithm showed the highest sensitivity for the diagnosis of melanoma.     

Früherkennung eines malignen Melanoms mittels computergestüzt videodokumentierter Auflichtmikroskopie von multiplen Pigmentmalen

Summary: We report on a 34 years old male patient with multiple melanocytic nevi, in whom the dermatoscopic video image of a pigmented lesion on his lower limb was computerstored among others. Its surface measured 10,6 mm².
By means of the computerbased ?Tübingen” mole‐analyzer several parameters of its margin and structure were calculated. Following the ABCD‐rules of dermatoscopy a score of 4.0 was calculated for this lesion. After one year the pigmented lesion measured more than 15,6 mm². Most parameters of the computerized mole‐analyzer have become worse. The dermatoscopic ABCD score was calculated 7.5 now. The subsequent exzision biopsy of the lesion disclosed a superficial type malignant melanoma with Clark II level and a Breslow thickness of 0.43 mm. The present case demonstrates that using the technique of computerized video image dermatoscopy it is possible to detect early malignant melanoma in a stage when final cure is still highly probable.     

Dermatoscopic prediction of melanoma thickness using latent trait analysis and likelihood ratios.

Breslow thickness and Clark level can be used to determine surgical procedures for cutaneous malignant melanoma and patient eligibility for experimental adjuvant therapy. Efforts to predict the thickness of melanomas using dermatoscopy have focused on differences between single dermatoscopic findings. The aim of this study was to develop a method for preoperative identification of melanomas of > or = 1 mm Breslow thickness using the entire range of dermatoscopic findings. Sixty-five melanomas were assessed for the presence of 22 dermatoscopic features. Ten dermatoscopic features showed differences in thick and thin melanomas and were selected for further analysis. A latent trait analysis construct implied that a progression in dermatoscopic features was associated with advancement of melanomas. Early melanomas are characterized by a light brown colour, a pigment network and irregularity or heterogeneity. Gray-blue areas, white scar-like areas and an atypical vascular pattern gradually displace these features. Likelihood ratios were determined for these 6 dermatoscopic findings and an algorithm for calculating the probability of thick malignant melanoma was established.     

Dermatoscopy of amelanotic and hypomelanotic melanoma

Amelanotic melanoma is a subtype of cutaneous melanoma without pigment. The clinical diagnosis is challenging because it may mimic benign or malignant melanocytic and non‐melanocytic neoplasms and inflammatory skin diseases. In synchrony with the improvement of the diagnosis of pigmented lesions, dermatoscopy may assist the clinician in the diagnosis of non‐pigmented skin neoplasms in general and of amelanotic melanoma in particular. We have searched the literature to extract the most relevant dermatoscopic clues to diagnose amelanotic and hypomelanotic melanomas by dermatoscopy. In addition we present eight consecutive cases and discuss their clinical and dermatoscopic characteristics in the light of published data.     

Modified ABC-point list of dermoscopy: A simplified and highly accurate dermoscopic algorithm for the diagnosis of cutaneous melanocytic lesions

BACKGROUND: The use of dermoscopy (epiluminescence microscopy, surface microscopy, dermatoscopy) improves clinical diagnostic sensitivity by 10% to 27%, particularly achieved by different algorithms or scores. OBJECTIVE: We sought to develop a simplified and highly accurate dermoscopic-point list for cutaneous melanocytic lesions. METHOD: We studied consecutive patients with suspicious melanocytic lesions, which were excised and histopathologically examined at our institution. On the basis of the ABCD rule of Stolz, Menzies score, and the modified ABCD rule of Kittler, a simplified ABC-point list was developed. Simple points were given for the following: asymmetry of outer shape (A) or differential structures inside the lesion in at least 1 axis ((A)); the abrupt cutoff of network at the border in at least one quarter of circumference (B); 3 or more colors (C); 3 or more differential structures (D); or noticed change (evolution) in the last 3 months (E). Using 20-fold magnification of computer dermoscopy, the sensitivity, specificity, and diagnostic accuracy were examined in 269 cutaneous melanocytic lesions. Of these, 84 (31.2%) were cutaneous melanomas. Also, the sensitivity, specificity, and diagnostic accuracy were investigated with a 7-point checklist and the 7 features for melanoma. RESULTS: With the ABC-point list for the diagnosis of cutaneous melanoma, sensitivity was 90.5%, specificity was 87%, and diagnostic accuracy was 88.1%, confirmed by cross-validation. The ABCD rule resulted in 90.5%, 72.4%, and 78.1%; Menzies score in 95.2%, 77.8%, and 83.3%; 7-point checklist in 90.5%, 87%, and 88.1%; and 7 features for melanoma in 94%, 74.6%, and 80.7%, respectively, CONCLUSIONS: The ABC-point list is simpler than the already established algorithms. Despite its simplicity, a high sensitivity, specificity, and diagnostic accuracy was achieved. This simplified approach in dermoscopic diagnostics may contribute to further spread and enable to learn and use this method more easily.     

Principles of dermatoscopy of pigmented skin lesions

There has been a dramatic increase in the incidence of malignant melanoma in most parts of the world. Because the tumor thickness is the most important prognostic factor for the prognosis of the malignant melanoma, the early detection of thin melanomas is essential. Dermatoscopy allows the physician to discriminate between melanocytic and nonmelanocytic lesions with high diagnostic accuracy and to detect initial malignant melanomas. We review the principles of dermatoscopy and the differential diagnosis of pigmented skin lesions. Before using the ABCD rule of dermatoscopy to classify melanocytic lesions into benign, suspicious, or malignant, the distinction between melanocytic and nonmelanocytic lesions is necessary. An essential prerequisite for the usefulness of this technique is adequate training.     

Current technologies for the in vivo diagnosis of cutaneous melanomas

The rising incidence of cutaneous malignant melanoma has been observed in the past decades. Currently, there is no cure for metastatic melanoma; only early diagnosis followed by prompt excision of cutaneous lesions ensures a good prognosis. The clinical ABCD rule is created as a framework for differentiating melanomas from benign pigmented skin lesions, and it serves as the basis for current clinical diagnosis. The ABCD rule relies on four simple clinical morphologies of melanoma: 1) Asymmetry, 2) Border irregularity, 3) Color variegation, and 4) Diameter greater than 6 mm. Although it is valuable, it has its limitations. Currently, the diagnostic accuracy for physicians is about 65%. This statistic implies that 1) melanomas with subtle signs are missed as benign lesions, and 2) benign lesions are over diagnosed as melanomas, which lead to unnecessary biopsies.     

Dermatoscopy of "dysplastic nevi": a beacon in diagnostic darkness

The sequential progression model for melanocytic tumours from common nevus to malignant melanoma was proposed by Clark almost 30 years ago. The "dysplastic nevus" has frequently been considered a logical offspring of this concept and as a direct precursor of melanoma, analogous to the epithelial dysplasia-carcinoma sequence. Despite the use of modern molecular methods, there is no consensus as to if the dysplastic nevus represents a true precursor lesion of melanoma, a separate distinct type of nevus, or a diagnostic dilemma. Currently, the concept of melanocytic dysplasia remains subject to confusing definitions at all levels of the diagnostic process, i.e. clinical appearance, dermatohistopathology, and molecular biology. In this review, we collect evidence that nevi fulfilling Clark and Elder's classic histological criteria mostly represent "endpoints" of nevocytic evolution, whereas a minority of "dysplastic nevi" represent true melanoma precursors. The unsolved dilemma is that neither clinical, histopathological nor molecular criteria exist to make a distinction between dysplastic nevi and early melanomas. Our analysis of the current knowledge on dysplastic nevi shows that dermatoscopy remains the only quantifiable, easily applicable and reproducible diagnostic tool to approach the problem. Due to a "quantum leap" in optical resolution, objective scores can be established, e.g. the total dermatoscopy score (TDS) according to the ABCD rule, and documentation of changes over time are possible by digital image storage devices. Although dermatoscopy does not solve the dilemma of discriminating early, basically feature-less melanomas from dysplastic nevi, and it does not prove that dysplastic nevus is a distinct entity, it helps make melanocytic tumours with unclear malignant potential a manageable disease.     

Dermatoscopy for melanoma and pigmented lesions

This article presents an overview of the history and development of dermatoscopy over the last 2 decades. The common dermatoscopic diagnostic algorithms are discussed, including classic pattern analysis, the ABCD rule (asymmetry, border, color, and dermatoscopic structures), 7-point checklist, and Menzies method, as well as a new method by the authors (ASAP: a simple and practical approach). In addition, evidence on the clinical impact and challenges of dermatoscopy for the diagnosis and management of pigmented lesions and the importance of training are reviewed.     

Melanoma detection. A prospective study comparing diagnosis with the naked eye,dermatoscopy and telespectrophotometry

BACKGROUND: Successful treatment of melanoma depends directly on early diagnosis. Such a diagnosis is based on clinical examination and dermatoscopy. Recently, automated instruments for melanoma detection are under development. OBJECTIVE: To prospectively evaluate the diagnostic possibilities provided by clinical and dermatoscopic examinations and by a computerized telespectrophotometric system (TS). METHODS: The study involves a consecutive series of 298 patients with 313 cutaneous pigmented lesions (66 melanomas and 247 non-melanoma lesions). Each lesion was subjected to the triple diagnostic evaluation, before surgery. Results were expressed in terms of sensitivity and specificity of each kind of evaluation. RESULTS: Clinical evaluation had sensitivity and specificity values of 86 and 77%, respectively, whereas dermatoscopy gave corresponding values of 91 and 74%. TS assessment resulted in a sensitivity of 80% and a specificity of 49%. Differences between clinical and dermatoscopic diagnoses lacked statistical significance (p = 0.22), whereas there was a significant difference comparing both clinical and TS evaluations (p < 0.01) and dermatoscopic and TS evaluations (p < 0.01). Combining clinical and dermatoscopic evaluations, a sensitivity of 97% was achieved. Addition of TS has not changed this figure. CONCLUSIONS: Results of this study confirm and stress the importance of dermatoscopy in the diagnosis of melanoma. Clinical evaluation coupled with dermatoscopy can be considered the cornerstone of such a diagnosis. Although TS is able to achieve interesting results, at present it cannot significantly compete with any of the other tested methods.     

Adding dermatoscopy to naked eye examination of equivocal melanocytic skin lesions: effect on intention to excise by general dermatologists

Background. According to the literature, dermatoscopy can improve diagnostic accuracy for melanoma. However, a weak point of the studies in the literature is that most were carried out in a ‘privileged’ setting of dermatologists who are expert in dermato‐oncology, and who work in departments specializing in screening pigmented lesions. This study was set up to specifically evaluate whether the use of dermatoscopy by general dermatologists would also improve accuracy. Aim. To analyse the effect on intention to excise lesions (intervention yes/no) of adding either dermatoscopy (20 years after the advent of the method) or detailed lesion classification (melanoma yes/no) to clinical examination by the naked eye. More specifically, we evaluated whether the current practice of general dermatologists using dermatoscopy improves the sensitivity and specificity values, and thus the diagnostic accuracy. Methods. Eight general dermatologists examined separately clinical images and combined (clinical and dermatoscopic) images of 200 melanocytic lesions that had been excised (64 melanomas and 136 melanocytic naevi). Results. Focusing on intention to excise (intervention yes/no), addition of dermatoscopy to naked eye examination resulted in an increase in sensitivity for all observers (average gain + 4.5%) but an overall nonsignificant reduction in specificity (?4.5%, P = 0.10). Diagnostic accuracy, which increased when examination was focused on melanoma (yes/no) classification (+ 4.1%, P < 0.05) remained unchanged (?1.62%; P = 0.36). Conclusions. The effect of adding dermatoscopy to naked eye examination of melanocytic lesions on ‘general dermatologists’ changes according to the aim of the examination. Dealing with the intention to excise, the increase of sensitivity associated with dermatoscopy (lower risk of leaving a melanoma unexcised) is obtained at the expense of specificity (higher number of melanocytic naevi excised) without improving overall diagnostic accuracy.     

Poppyfield bleeding: a new dermatoscopic sign and its histopathological background

Dermatoscopy increases the accuracy of diagnosis of melanoma. An atypical vascular pattern may be an indicator of cutaneous malignant melanoma (CMM). During dermatoscopy of certain CMMs numerous ruby droplets of blood appear when the dermatoscope is pressed firmly against the lesion. The aim of this paper was to examine the histopathological background for this observation. CMMs from 8 patients showing the poppyfield sign, i.e. squirts of ruby blood droplets, were paired with 8 CMMs of equal Breslow thickness not showing this sign. The 16 CMMs were placed in an unsystematic sequence and presented to two dermato-pathologists who assessed the lesions independently for confirmation of Breslow thickness, Clark level, ulceration and presence of dilated tumour vessels. There was no disagreement between the pathologists' assessments. Age of the patients and Breslow thickness of the cutaneous malignant melanoma were similar in the two groups. All 8 poppyfield CMMs had dilated tumour vessels compared with 25% (2/8) of the non-poppyfield CMMs (p< 0.007). Histological ulceration was observed in all poppyfield CMMs and none of the non-poppyfield CMMs (p< 0.001). The poppyfield bleeding sign is a dermatoscopic clue to dilated tumour vessels. It may be a dermatoscopic reflection of increased vascular density described in primary CMMs compared with adjacent skin and may also reflect the presence of primitive vessels in CMMs displaying increased fragility.     

Amelanotic metastatic cutaneous melanoma

Dermatoscopy of melanocytic lesions has guided the decision of when or not to biopsy a lesion. The use of this tool has increased clinical examination''s sensitivity and specificity in 89% and 96% respectively. However, dermatoscopic evaluation of amelanotic or hypomelanotic melanomas, as well as metastases, can be difficult. There is still no standardization for the analysis of these pathologies, which relies mostly on their vascular pattern. We describe the dermatoscopy of acral metastatic amelanotic melanoma.     

Clinical and dermatoscopic diagnosis of malignant melanoma. Assessed by expert and non-expert groups.

We investigated the nosographic and diagnostic probabilities and likelihood ratios of dermatoscopy in order to evaluate the method's role in decision-making regarding melanoma. Clinical slides and dermatoscopic photos were obtained from 232 patients referred for dermatoscopy. Four dermatoscopy "experts" and 5 "non-experts" assessed the slides. Diagnoses were compared with histopathology. Sensitivity of the clinical assessments was 0.78 vs. 0.69 ("experts" vs. "non-experts"), sensitivity of dermatoscopy assessment was 0.83 vs. 0.69 (p = 0.04). The expert group demonstrated increased specificity (from 0.89 to 0.94) when applying dermatoscopy compared with clinical assessment alone (p=0.03). Positive likelihood ratios were doubled in the "expert group" and the negative likelihood ratios improved 25% with dermatoscopy compared with clinical assessment.     

Correlation between clinical, dermatoscopic, and histopathologic variables in atypical melanocytic nevi

IntroductionAtypical melanocytic nevi are acquired melanocytic lesions that were described for the first time by Clark in studies of melanocytic nevi in patients with melanomas. Today, the use of dermatoscopy has made identification of this type of nevus much easier.ObjectiveOur aim was to study the correlation between the clinical, dermatoscopic, and histopathologic findings of melanocytic nevi and compare our findings with those of previous studies. We also aimed to investigate the value of dermatoscopy for identifying atypical melanocytic nevi.Material and methodsIn this cross-sectional, observational study, 200 melanocytic lesions were analyzed in 166 patients examined between January 1, 2005 and December 31, 2005. We recorded the clinical, dermatoscopic, and histopathologic characteristics of each lesion and established the correlation between the different findings on a case-by-case basis. We then determined the agreement between diagnoses and assessed the value of dermatoscopy for identifying atypical melanocytic melanoma.ResultsThe clinical characteristics associated with atypical histology were a macular component (P < .001), irregular borders, and presence of 3 or more colors. Asymmetry, diameter greater than 5 or 6 mm, and progression were not associated with atypical histopathologic characteristics (P > .05). Agreement between clinical and histologic diagnosis was weak (κp = 0.38), whereas the agreement between dermatoscopic and histologic diagnosis was moderate (κp = 0.52). The area under the receiver operating characteristic curve for the model that included dermatoscopy was larger than that for the model that only included clinical data, and this difference was statistically significant.ConclusionsAtypical clinical features were not found to correspond to atypical histology. Dermatoscopy improved the accuracy of clinical diagnosis of atypical melanocytic nevus.     

Regular clinical dermatoscope use with training improves melanoma diagnosis by dermatologists

Dermatoscopy is not accepted by all dermatologists as a useful diagnostic tool. We set out to test if training followed by regular clinical use of dermatoscopes improved diagnostic accuracy in melanoma diagnosis. Six dermatologists who had not previously used dermatoscopes were studied before and after a 10‐month period of dermatoscope use with training and use of the Modified Pattern Analysis Diagnostic Algorithm. Diagnostic accuracy was assessed using test cards containing clinical and dermatoscopic photographs. The number of melanomas undiagnosed after training fell from 18 to 5 and the number potentially left unexcised fell from 18 to 3. The numbers of benign lesions potentially excised remained unchanged. The study shows that the use of dermatoscopes with training greatly increased the accuracy of diagnosis of melanoma by dermatologists. This practical study supports the use of dermatoscopy in pigmented‐lesion diagnosis and demonstrates how dermatoscopy training could be incorporated into UK specialist training programmes.     

Using 3D differential forms to characterize a pigmented lesion in vivo

Background/purpose: After the formulation of ABCD rules, many new feature extraction methods are emerging to describe the asymmetry, border irregularity, color variation and diameter of malignant melanoma. In this paper, a new research direction orthogonal to ABCD rules that characterizes 3D local disruption of skin surfaces to realize automatic recognition of melanoma is described.
Methods: This paper examines 3D differential forms of skin surfaces to characterize the local geometrical properties of melanoma. Firstly, 3D data of skin surfaces are obtained using a photometric stereo device. Then differential forms of lesion surfaces are determined to describe the geometrical texture patterns involved. Using only these geometrical features, a simple least-squared error-based linear classifier can be constructed to realize the classification of malignant melanomas and benign lesions.
Results: As with the 3D data of 35 melanoma and 66 benign lesion samples collected from local pigmented lesion clinics, the optimal sensitivity and specificity of the constructed linear classifier are 71.4% and 86.4%, respectively. The total area enclosed by the corresponding receiver operating characteristics curve is 0.823.
Conclusion: This study indicates that differential forms obtained from 3D data are very promising in characterizing melanoma. Combining these features with other skin features such as border irregularity and color variation might further improve the accuracy and reliability of the automatic diagnosis of melanoma.