Automatic segmentation of organs at risk in thoracic CT scans by combining 2D and 3D convolutional neural networks

Louis D. van Harten, Julia M.H. Noothout, Joost J.C. Verhoeff, Jelmer M. Wolterink, Ivana Išgum

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)
95 Downloads (Pure)

Abstract

Segmentation of organs at risk (OARs) in medical images is an important step in treatment planning for patients undergoing radiotherapy (RT). Manual segmentation of OARs is often time-consuming and tedious. Therefore, we propose a method for automatic segmentation of OARs in thoracic RT treatment planning CT scans of patients diagnosed with lung, breast or esophageal cancer. The method consists of a combination of a 2D and a 3D convolutional neural network (CNN), where both networks have substantially different architectures. We analyse the performance for these networks individually and show that a combination of both networks produces the best results. With this combination, we achieve average Dice coefficients of 0.84± 0.05, 0.94± 0.02, 0.91± 0.02, and 0.93± 0.01 for the esophagus, heart, trachea, and aorta, respectively. These results demonstrate potential for automating segmentation of organs at risk in routine radiotherapy treatment planning.

Original languageEnglish
Title of host publicationSegTHOR 2019
Subtitle of host publicationProceedings of the 2019 Challenge on Segmentation of THoracic Organs at Risk in CT Images (SegTHOR2019)
EditorsCaroline Petitjean, Su Ruan, Zoé Lamber, Bernard Dubray
PublisherCEUR
Chapter12
Number of pages4
Publication statusPublished - 2019
Externally publishedYes
Event2019 SegTHOR Challenge: Segmentation of THoracic Organs at Risk in CT Images - Hilton Molino Stucky, Venice, Italy
Duration: 8 Apr 20198 Apr 2019

Publication series

NameCEUR workshop proceedings
PublisherRheinisch Westfälische Technische Hochschule
Volume2349
ISSN (Print)1613-0073

Conference

Conference2019 SegTHOR Challenge
CountryItaly
CityVenice
Period8/04/198/04/19

Fingerprint

Dive into the research topics of 'Automatic segmentation of organs at risk in thoracic CT scans by combining 2D and 3D convolutional neural networks'. Together they form a unique fingerprint.

Cite this