Development of diagnostic algorithms for image analysis of skin lesions

Yova, DM; Delibasis, AK; Papaodysseus, CN; Koukoutsis, E; Vasilopoulos, P

dc.contributor.author	Yova, DM	en
dc.contributor.author	Delibasis, AK	en
dc.contributor.author	Papaodysseus, CN	en
dc.contributor.author	Koukoutsis, E	en
dc.contributor.author	Vasilopoulos, P	en
dc.date.accessioned	2014-03-01T02:48:30Z
dc.date.available	2014-03-01T02:48:30Z
dc.date.issued	1997	en
dc.identifier.issn	0277786X	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/33874
dc.subject	Computers	en
dc.subject	Dermatology	en
dc.subject	Diagnostic algorithms	en
dc.subject	Image analysis	en
dc.subject	Image processing	en
dc.subject	Melanoma	en
dc.subject	Skin cancer	en
dc.subject.other	Bubbles (in fluids)	en
dc.subject.other	Cameras	en
dc.subject.other	CCD cameras	en
dc.subject.other	Charge coupled devices	en
dc.subject.other	Dermatology	en
dc.subject.other	Digital cameras	en
dc.subject.other	Digital image storage	en
dc.subject.other	Image analysis	en
dc.subject.other	Image enhancement	en
dc.subject.other	Image processing	en
dc.subject.other	Imaging systems	en
dc.subject.other	Light	en
dc.subject.other	Light sources	en
dc.subject.other	Lighting	en
dc.subject.other	Optical data processing	en
dc.subject.other	Optoelectronic devices	en
dc.subject.other	System program documentation	en
dc.subject.other	And detections	en
dc.subject.other	Color tables	en
dc.subject.other	Diagnostic algorithms	en
dc.subject.other	Digital imaging	en
dc.subject.other	Digital imaging systems	en
dc.subject.other	Frame grabbers	en
dc.subject.other	Malignant melanomata	en
dc.subject.other	Melanoma	en
dc.subject.other	Pentium	en
dc.subject.other	Prototype systems	en
dc.subject.other	Region growing	en
dc.subject.other	Skin cancer	en
dc.subject.other	Skin lesions	en
dc.subject.other	Thresholding	en
dc.subject.other	Skin	en
dc.title	Development of diagnostic algorithms for image analysis of skin lesions	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1117/12.274329	en
heal.identifier.secondary	http://dx.doi.org/10.1117/12.274329	en
heal.publicationDate	1997	en
heal.abstract	The crucial step in the diagnostic treatment of skin cancer is the initial examination and detection of any unusual change of a skin lesion. Digital imaging permits the documentation of the size, shape and color of lesions and their later comparison, so the last years its key role as an adjunct to early malignant melanoma diagnosis has arisen. In this work, a novel approach to diagnosis is presented by developing a digital imaging system for capturing and processing the images of individual lesions. It was used a 12 bit CCD camera, connected with a Pentium PC equipped with a Matrox frame grabber and convenient software. Images were collected using a zoom close up lens and a light source, attached to the front of the CCD camera. The format of the images was 640 X 480 pixels with 8 bit color table. The border of the lesion was found using a region growing based algorithm combined with the Gradient Operator. It was calculated the border irregularity and the asymmetry of the lesion, using the compactness formula. There was developed algorithms based on thresholding and region growing to determine the borders of the lesion. Compactness has been calculated as well. The pre-prototype system has been placed at the National Cancer Hospital to support melanoma diagnosis. ©2004 Copyright SPIE - The International Society for Optical Engineering.	en
heal.journalName	Proceedings of SPIE - The International Society for Optical Engineering	en
dc.identifier.doi	10.1117/12.274329	en
dc.identifier.volume	2983	en
dc.identifier.spage	115	en
dc.identifier.epage	122	en