Tumor Delineation in PET to generate Gross Tumor Volume- Comparison and Analysis of Four Image Segmentation

Christy Esther S , Devakumar Devadhas

PET. Image Segmentation. Tumor delineation. Gross tumor volume. Radiotherapy treatment planning

Numerous image segmentation methods have been developed for the segmentation of tumor in PET. In this article, four image segmentation methods have been compared and evaluated to determine the near accurate method. A series of phantom studies were performed, to determine an most accurate and uniformly applicable method to define the Gross tumor volume (GTV) with PET. The obtained results were affirmative; hence the methods were performed on 10 patient data for accurate delineation of GTV in PET images. Four image segmentation techniques 1) 25% threshold, 2) threshold based Schaefer method, 3) statistical segmentation method, and 4) region growing method, were first tested by a phantom study to determine the GTV in PET images. The methods were then evaluated using patient data in which the segmentation results were compared with the Gross tumor volume (GTV) obtained by manual method. All the four methods showed good results for the phantom study. On the other hand, clinical studies show that, the 25% threshold method gave good segmentation results. The statistical method and region growing method gave consistent good results with a dice similarity coefficient about 96%, while the Schaefer method underestimated the tumor volume.

Published Paper   E-Certificate

"Tumor Delineation in PET to generate Gross Tumor Volume- Comparison and Analysis of Four Image Segmentation ", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.5, Issue 1, page no.64 - 69, January-2020, Available :https://ijsdr.org/papers/IJSDR2001011.pdf

Volume 5 Issue 1, January-2020

Pages : 64 - 69

Paper Reg. ID: IJSDR_191213

Published Paper Id: IJSDR2001011

Downloads: 000347036

Research Area: Engineering

Country: Coimbatore, Tamilnadu, India

Published Paper PDF: https://ijsdr.org/papers/IJSDR2001011

Published Paper URL: https://ijsdr.org/viewpaperforall?paper=IJSDR2001011

DOI: http://doi.one/10.1729/Journal.23036

ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016

An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator

Publisher: IJSDR(IJ Publication) Janvi Wave

This work is licensed under a Creative Commons Attribution 4.0 International License and The Open Definition