RTU Research Information System
Latviešu English

Publikācija: Development of a New Segmentation Method for Medical Images

Publication Type Publications in RTU scientific journal
Funding for basic activity Unknown
Defending: ,
Publication language Latvian (lv)
Title in original language Jaunas medicīnas attēlu segmentācijas metodes izstrādāšana
Title in English Development of a New Segmentation Method for Medical Images
Field of research 2. Engineering and technology
Sub-field of research 2.2 Electrical engineering, Electronic engineering, Information and communication engineering
Authors Aleksandrs Glazs
Katrīna Krečetova
Keywords attēlu apstrāde, medicīnas attēli, segmentācija, segmentācijas metode
Abstract This work describes the problem of medical images segmentation. The goal of this work was to develop a new semi-automatic segmentation method that could process medical images. In this work a new segmentation method is proposed. It uses a combination of three segmentation algorithms and two image enhancement methods, of which two histogram-based segmentation algorithms and one image enhancement algorithm were developed by the authors. This work describes the method and developed algorithms. The proposed method is semi-automatic. It automatically enhances the image and divides the image into segments. For now, the choice of segment that contains the pathology zone is user-dependant, but the proposed method makes the search for pathology zones easier. The proposed method was tested on medical images, and after the study of results, the conclusion was made, that the method can be used as an extra tool for image analysis, to make the diagnostics of medical image easier.
Abstract in English This work describes the problem of medical images segmentation. The goal of this work was to develop a new semi-automatic segmentation method that could process medical images. In this work a new segmentation method is proposed. It uses a combination of three segmentation algorithms and two image enhancement methods, of which two histogram-based segmentation algorithms and one image enhancement algorithm were developed by the authors. This work describes the method and developed algorithms. The proposed method is semi-automatic. It automatically enhances the image and divides the image into segments. For now, the choice of segment that contains the pathology zone is user-dependant, but the proposed method makes the search for pathology zones easier. The proposed method was tested on medical images, and after the study of results, the conclusion was made, that the method can be used as an extra tool for image analysis, to make the diagnostics of medical image easier.
Reference Glazs, A., Krečetova, K. Development of a New Segmentation Method for Medical Images. Technologies of Computer Control. Vol.32, 2007, pp.14-21. ISSN 1407-7493.
Full-text Full-text
ID 3432