Title :
Automatic learning of spatial patterns for diagnosis of skin lesions
Author :
Zortea, Maciel ; Skrøvseth, Stein Olav ; Godtliebsen, Fred
Author_Institution :
Dept. of Math. & Stat., Univ. of Tromso, Tromsø, Norway
fDate :
Aug. 31 2010-Sept. 4 2010
Abstract :
We present a technique for automatic diagnosis of malignant melanoma based exclusively on local pattern analysis. The technique relies on local binary patterns in small sections in the image, and automatically selects the relevant texture features from those that discriminate best between benign and malignant skin lesions. The classification is performed using support vector machines, and the feature vectors are clustered using K-means clustering. The effects of K and window size are investigated. Reported average specificity and sensitivity are 73% for optimal parameter choice, indicating that the procedure is a useful part of a diagnostic system.
Keywords :
cancer; feature extraction; image classification; image texture; medical image processing; pattern clustering; skin; support vector machines; K-means clustering; automatic diagnosis; automatic learning; classification; feature vectors; local binary patterns; malignant melanoma; sensitivity; skin lesions; spatial patterns; specificity; support vector machines; texture features; Cancer; Kernel; Lesions; Malignant tumors; Skin; Support vector machines; Training; Algorithms; Humans; Learning; Melanoma; Pattern Recognition, Automated; Sensitivity and Specificity; Skin Neoplasms;
Conference_Titel :
Engineering in Medicine and Biology Society (EMBC), 2010 Annual International Conference of the IEEE
Conference_Location :
Buenos Aires
Print_ISBN :
978-1-4244-4123-5
DOI :
10.1109/IEMBS.2010.5626801