Title :
Fully automated segmentation of lungs and large cancerous areas
Author :
Ozsavas, Emin Emrah ; Telatar, Z. ; Dirican, Bahar ; Sager, Omer
Author_Institution :
Elektrik-Elektron. Muhendisligi Bolumu, Aankara Univ., Ankara, Turkey
Abstract :
In the field of modern radiotherapy, in order to determine the critical organs and cancerous areas accurately various automated segmentation algorithms have been proposed. Segmentation of the lungs from computed tomography (CT) scans is an indispensable part of radiation treatment planning (RTP). Conventional lung segmentation algorithms may fail due to the low contrast between the lungs and surrounding structures in case of large cancerous areas. In this study, to be used in RTP, a fully automated method that can segment the lungs from CT scans accurately and in the sequel detect large cancerous areas is proposed. The proposed method that consists of 8 steps in total and includes our new algorithms along with the well-known image processing algorithms was applied to the CT scans of 20 patients with lung cancer. The obtained results show that this concise and effective method avoiding heavy computational load and offering expedited segmentation may be used in RTP.
Keywords :
cancer; computerised tomography; image segmentation; lung; medical image processing; radiation therapy; computed tomography; fully automated image segmentation; image processing algorithm; large cancerous areas; lung image; radiation treatment planning; Cancer; Computed tomography; Conferences; Image segmentation; Lungs; Planning; Signal processing; Computed Tomography (CT); Radiation Treatment Planning (RTP); cancer; lung; segmentation;
Conference_Titel :
Signal Processing and Communications Applications Conference (SIU), 2014 22nd
Conference_Location :
Trabzon
DOI :
10.1109/SIU.2014.6830302