Computer-aided classification of melanocytic lesions using dermoscopic images

Ferris, Laura K.; Harkes, Jan; Gilbert, Benjamin; Winger, Daniel G.; Golubets, Kseniya; Akilov, Oleg E.; Satyanarayanan, Mahadev

doi:10.1016/j.jaad.2015.07.028

Cited by 94 publications

(63 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Additionally, random forests (RF) learning methods have also been applied to dermatoscopy images for both melanoma and BCC classification. Ferris et al constructed a model of 1000 decision trees and a threshold for malignant diagnosis of 0.4, with sensitivity results higher than physicians and specificity lower, while Kharazmi et al explored the use of vascular features for basal cell carcinoma automatic detection with 100 trees. No reference is made to the reasons considered for tree number selection.…”

Section: Resultsmentioning

confidence: 99%

The role of AI classifiers in skin cancer images

Magalhães

Mendes

Vardasca

2019

Skin Research and Technology

View full text Add to dashboard Cite

Background The use of different imaging modalities to assist in skin cancer diagnosis is a common practice in clinical scenarios. Different features representative of the lesion under evaluation can be retrieved from image analysis and processing. However, the integration and understanding of these additional parameters can be a challenging task for physicians, so artificial intelligence (AI) methods can be implemented to assist in this process. This bibliographic research was performed with the goal of assessing the current applications of AI algorithms as an assistive tool in skin cancer diagnosis, based on information retrieved from different imaging modalities. Materials and methods The bibliography databases ISI Web of Science, PubMed and Scopus were used for the literature search, with the combination of keywords: skin cancer, skin neoplasm, imaging and classification methods. Results The search resulted in 526 publications, which underwent a screening process, considering the established eligibility criteria. After screening, only 65 were qualified for revision. Conclusion Different imaging modalities have already been coupled with AI methods, particularly dermoscopy for melanoma recognition. Learners based on support vector machines seem to be the preferred option. Future work should focus on image analysis, processing stages and image fusion assuring the best possible classification outcome.

show abstract

Section: Resultsmentioning

confidence: 99%

The role of AI classifiers in skin cancer images

Magalhães

Mendes

Vardasca

2019

Skin Research and Technology

View full text Add to dashboard Cite

show abstract

“…Dermoscopy is an imaging modality that uses a simple hand-held device that eliminates surface glare and magnifies structures invisible to the naked eye. Although dermoscopy significantly improves diagnostic accuracy compared to naked-eye examination alone [5,6], average dermoscopic melanoma sensitivity for dermatologists and others seeing dermatology patients in three recent reader studies was only 71-85% and specificity 54-71% [7][8][9]. Non-specialist sensitivity and specificity is even lower; when compared to dermatologists, non-specialists were found to have a sensitivity for melanoma of 54% and a specificity of 73% [10].…”

Section: Introductionmentioning

confidence: 93%

“…These descriptors are among the most significant of all analytic descriptors of melanoma. Ferris et al, found that the top three statistical features for melanoma were all color feature characterizing color histograms and color asymmetry [7]. Analysis of various analytic and clinical features by Rubegni et al, showed that the single most significant feature was red asymmetry [15].…”

Section: Introductionmentioning

confidence: 99%

Fuzzy Color Clustering for Melanoma Diagnosis in Dermoscopy Images

et al. 2017

View full text Add to dashboard Cite

A fuzzy logic-based color histogram analysis technique is presented for discriminating benign skin lesions from malignant melanomas in dermoscopy images. The approach extends previous research for utilizing a fuzzy set for skin lesion color for a specified class of skin lesions, using alpha-cut and support set cardinality for quantifying a fuzzy ratio skin lesion color feature. Skin lesion discrimination results are reported for the fuzzy clustering ratio over different regions of the lesion over a data set of 517 dermoscopy images consisting of 175 invasive melanomas and 342 benign lesions. Experimental results show that the fuzzy clustering ratio applied over an eight-connected neighborhood on the outer 25% of the skin lesion with an alpha-cut of 0.08 can recognize 92.6% of melanomas with approximately 13.5% false positive lesions. These results show the critical importance of colors in the lesion periphery. Our fuzzy logic-based description of lesion colors offers relevance to clinical descriptions of malignant melanoma.

show abstract

“…Using dermoscopy imaging, melanoma is fully visible at the earliest stage, when it is fully curable . Over a billion dollars per year is spent on biopsies of lesions that turn out to be benign, and even then, cases of melanoma are missed by domain experts in dermoscopy . Dermoscopy increases the diagnostic accuracy over clinical visual inspection, but only after significant training .…”

Section: Introductionmentioning

confidence: 99%

Automatic lesion border selection in dermoscopy images using morphology and color features

Mishra

Kaur

Kasmi

et al. 2019

Skin Research and Technology

View full text Add to dashboard Cite

Purpose We present a classifier for automatically selecting a lesion border for dermoscopy skin lesion images, to aid in computer‐aided diagnosis of melanoma. Variation in photographic technique of dermoscopy images makes segmentation of skin lesions a difficult problem. No single algorithm provides an acceptable lesion border to allow further processing of skin lesions. Methods We present a random forests border classifier model to select a lesion border from 12 segmentation algorithm borders, graded on a “good‐enough” border basis. Morphology and color features inside and outside the automatic border are used to build the model. Results For a random forests classifier applied to an 802‐lesion test set, the model predicts a satisfactory border in 96.38% of cases, in comparison to the best single border algorithm, which detects a satisfactory border in 85.91% of cases. Conclusion The performance of the classifier‐based automatic skin lesion finder is found to be better than any single algorithm used in this research.

show abstract

Computer-aided classification of melanocytic lesions using dermoscopic images

Cited by 94 publications

References 9 publications

The role of AI classifiers in skin cancer images

The role of AI classifiers in skin cancer images

Fuzzy Color Clustering for Melanoma Diagnosis in Dermoscopy Images

Automatic lesion border selection in dermoscopy images using morphology and color features

Contact Info

Product

Resources

About