AUTOMATIC ORGAN DELINEATION OF COMPUTED TOMOGRAPHY IMAGES FOR RADIOTHERAPY PLANNING IN PROSTATE CANCER An Overview

• Type: Article in International Conference Proceedings Book
• Date: 2011

Authors

Celeste Marques Oliveira (Author) Other; Pedro Pereira Rodrigues (Author) FMUP

Conference proceedings International

Title: HEALTHINF 2011: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON HEALTH INFORMATICS

Pages: 482-485

4th International Conference on Health Informatics (HEALTHINF 2011)

Rome, ITALY, JAN 26-29, 2011

Indexing

ISI Web of Knowledge - 0 Citations

Scopus - 0 Citations

Scientific classification

FOS: Medical and Health sciences > Health sciences

Other information

Authenticus ID: P-002-YNE

Abstract (EN): Prostate cancer is a common cancer worldwide and a leading cause of death. Radiotherapy is usually the first-line treatment for patients with slow-growing cancer that is confined to the prostate. In Radiation Therapy Planning (RTP), the recognition and outlining of clinical volumes in computed tomography (CT) images are one of the most time-consuming steps carried out by human experts. The aim of this review is to identify and summarize evidence of the use of automatic organ delineation of CT images for radiotherapy planning in prostate cancer. From the literature search, a total of seven studies, reported between 1994 and 2009, were selected. We associate the selected studies in order to compare results, in spite of their differences in methodology and outcome evaluators. Most of the studies conclude that the automatic approach is faster, while having equivalent accuracy to manual method. Concerning the observer's variability, automatic segmentation reaches significant gains in reproducibility. As future directions, it is recommended the improvement of the segmentation algorithms in the delineation of problematic soft tissues and future validation studies with large scale trials and possible studies of meta-analysis in the specific problems.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 4