QUANTITATIVE COMPARISON OF SEGMENTATION METHODS FOR IN-BODY IMAGES

• Tipo: Artigo em Livro de Atas de Conferência Internacional
• Data: 2009

Autores

Farhan Riaz (Autor) Outra; Mario Dinis Ribeiro (Autor) Outra; Miguel Tavares Coimbra (Autor) FCUP

Ata de Conferência Internacional

Título: 2009 ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY, VOLS 1-20

Páginas: 5785-5788

Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society

Minneapolis, MN, SEP 03-06, 2009

Indexação

ISI Web of Knowledge - 5 Citações

Scopus - 9 Citações

Classificação Científica

FOS: Ciências da engenharia e tecnologias > Biotecnologia ambiental

Outras Informações

ID Authenticus: P-003-PS6

DOI: 10.1109/iembs.2009.5332540

Abstract (EN): In this paper, we present a numerical comparison of how well segmentation algorithms approximate the manual segmentation of gastroenterologists for a set of endoscopic images. Different areas in these images demand different levels of analysis by a clinician and some provide critical information about the patient. Our objective is thus to segment endoscopic images so that the results mimic as closely as possible the areas that were considered relevant by doctors. We focus on a detailed quantitative comparison of two popular segmentation algorithms, mean shift and normalized cuts, when applied to in-body images, most specifically for vital-stained magnification endoscopy. Segmentation results are compared with the manual annotations of the same images performed by two specialist clinicians. Results show that if we simply consider the most relevant segmented patch, normalized cuts performs better. However, if we allow the annotated area to be represented by multiple patches, mean shift is clearly a better choice, although automatic ways to determine its kernel's bandwidth are highly desirable.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 4