Automatic Telangiectasia Analysis in Dermoscopy Images Using Adaptive Critic Design
Abstract
Background: Telangiectasia, tiny skin vessels, are important dermoscopy structures used to discriminate basal cell carcinoma (BCC) from benign skin lesions. This research builds off of previously developed image analysis techniques to identify vessels automatically to discriminate benign lesions from BCCs. Methods: A biologically inspired reinforcement learning approach is investigated in an adaptive critic design framework to apply action-dependent heuristic dynamic programming (ADHDP) for discrimination based on computed features using different skin lesion contrast variations to promote the discrimination process. Lesion discrimination results for ADHDP are compared with multilayer perception backpropagation artificial neural networks. Results: This study uses a data set of 498 dermoscopy skin lesion images of 263 BCCs and 226 competitive benign images as the input sets. This data set is extended from previous research [Cheng et al., Skin Research and Technology, 2011, 17: 278]. Experimental results yielded a diagnostic accuracy as high as 84.6% using the ADHDP approach, providing an 8.03% improvement over a standard multilayer perception method. Conclusion: We have chosen BCC detection rather than vessel detection as the endpoint. Although vessel detection is inherently easier, BCC detection has potential direct clinical applications. Small BCCs are detectable early by dermoscopy and potentially detectable by the automated methods described in this research.
Recommended Citation
B. Cheng et al., "Automatic Telangiectasia Analysis in Dermoscopy Images Using Adaptive Critic Design," Skin Research and Technology, vol. 18, no. 4, pp. 389 - 396, John Wiley & Sons, Dec 2011.
The definitive version is available at https://doi.org/10.1111/j.1600-0846.2011.00584.x
Department(s)
Electrical and Computer Engineering
Second Department
Chemistry
Keywords and Phrases
Adaptive critic designs; ADHDP; Automated methods; Back propagation artificial neural network (BPANN); Basal cell carcinoma; Benign lesion; Biologically inspired; Clinical application; Contrast variation; Data sets; Dermoscopy; Dermoscopy images; Diagnostic accuracy; Heuristic dynamic programming; Image analysis techniques; Input set; Lesion discrimination; Multi-layer perception; Reinforcement learning approach; Skin lesion; Skin lesion images; Skin research; Telangiectasia; Vessel detection; Backpropagation; Crystal structure; Dermatology; Heuristic methods; Image processing; Neural networks; Reinforcement learning; Research; Diagnosis; action dependent heuristic dynamic programming; algorithm; artificial neural network; automation; clinical feature; computer program; controlled study; diagnostic test accuracy study; epiluminescence microscopy; human; image analysis; major clinical study; skin defect; Artificial Intelligence; Carcinoma; Basal Cell; Differential; Humans; Image Interpretation; Computer-Assisted; Pattern Recognition; Automated; Reproducibility of Results; Sensitivity and Specificity; Skin Neoplasms; Telangiectasis; Adaptive critic design
International Standard Serial Number (ISSN)
0909-752X
Document Type
Article - Journal
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2011 John Wiley & Sons, All rights reserved.
Publication Date
01 Dec 2011