Region growing by sector analysis for detection of blue-gray ovoids in basal cell carcinoma.

Blue-gray ovoids (B-GOs) are critical dermoscopic structures in basal cell carcinomas (BCCs) that pose a challenge for automatic detection. Due to variation in size and color, B-GOs can be easily mistaken for similar structures in benign lesions. Analysis of these structures could help further accomplish the goal of automatic BCC detection. This study introduces an efficient sector-based method for segmenting B-GOs. Four modifications of conventional region-growing techniques are presented: (i) employing a seed area rather than a seed point, (ii) utilizing fixed control limits determined from the seed area to eliminate re-calculations of previously-added regions, (iii) determining region growing criteria using logistic regression, and (iv) area analysis and expansion by sectors. Contact dermoscopy images of 68 confirmed BCCs having B-GOs were obtained. A total of 24 color features were analyzed for all B-GO seed areas. Logistic regression analysis determined blue chromaticity, followed by red variance, were the best features for discriminating B-GO edges from surrounding areas. Segmentation of malignant structures obtained an average Pratt's figure of merit of 0.397. The techniques presented here provide a non-recursive, sector-based, region-growing method applicable to any colored structure appearing in digital images. Further research using these techniques could lead to automatic detection of B-GOs in BCCs.

Sector expansion and elliptical modeling of blue-gray ovoids for basal cell carcinoma discrimination in dermoscopy images.

BACKGROUND: Blue-gray ovoids (B-GOs), a critical dermoscopic structure for basal cell carcinoma (BCC), offer an opportunity for automatic detection of BCC. Due to variation in size and color, B-GOs can be easily mistaken for similar structures in benign lesions. Analysis of these structures could afford accurate characterization and automatic recognition of B-GOs, furthering the goal of automatic BCC detection. This study utilizes a novel segmentation method to discriminate B-GOs from their benign mimics. METHODS: Contact dermoscopy images of 68 confirmed BCCs with B-GOs were obtained. Another set of 131 contact dermoscopic images of benign lesions possessing B-GO mimics provided a benign competitive set. A total of 22 B-GO features were analyzed for all structures: 21 color features and one size feature. Regarding segmentation, this study utilized a novel sector-based, non-recursive segmentation method to expand the masks applied to the B-GOs and mimicking structures. RESULTS: Logistic regression analysis determined that blue chromaticity was the best feature for discriminating true B-GOs in BCC from benign, mimicking structures. Discrimination of malignant structures was optimal when the final B-GO border was approximated by a best-fit ellipse. Using this optimal configuration, logistic regression analysis discriminated the expanded and fitted malignant structures from similar benign structures with a classification rate as high as 96.5%. CONCLUSIONS: Experimental results show that color features allow accurate expansion and localization of structures from seed areas. Modeling these structures as ellipses allows high discrimination of B-GOs in BCCs from similar structures in benign images.