Simultaneous cosegmentation of tumors in PET-CT images using deep fully convolutional networks

Zisha Zhong; Yusung Kim; Kristin Plichta; Bryan G Allen; Leixin Zhou; John Buatti; Xiaodong Wu

doi:10.1002/mp.13331

Back

Simultaneous cosegmentation of tumors in PET-CT images using deep fully convolutional networks

Journal article

Open access

Peer reviewed

Simultaneous cosegmentation of tumors in PET-CT images using deep fully convolutional networks

Zisha Zhong, Yusung Kim, Kristin Plichta, Bryan G Allen, Leixin Zhou, John Buatti and Xiaodong Wu

Medical physics (Lancaster), Vol.46(2), pp.619-633

02/2019

DOI: 10.1002/mp.13331

PMCID: PMC6527327

PMID: 30537103

Files and links (1)

url

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6527327View

Published (Version of record) Open Access

Abstract

To investigate the use and efficiency of 3-D deep learning, fully convolutional networks (DFCN) for simultaneous tumor cosegmentation on dual-modality nonsmall cell lung cancer (NSCLC) and positron emission tomography (PET)-computed tomography (CT) images. We used DFCN cosegmentation for NSCLC tumors in PET-CT images, considering both the CT and PET information. The proposed DFCN-based cosegmentation method consists of two coupled three-dimensional (3D)-UNets with an encoder-decoder architecture, which can communicate with the other in order to share complementary information between PET and CT. The weighted average sensitivity and positive predictive values denoted as Scores, dice similarity coefficients (DSCs), and the average symmetric surface distances were used to assess the performance of the proposed approach on 60 pairs of PET/CTs. A Simultaneous Truth and Performance Level Estimation Algorithm (STAPLE) of 3 expert physicians' delineations were used as a reference. The proposed DFCN framework was compared to 3 graph-based cosegmentation methods. Strong agreement was observed when using the STAPLE references for the proposed DFCN cosegmentation on the PET-CT images. The average DSCs on CT and PET are 0.861 ± 0.037 and 0.828 ± 0.087, respectively, using DFCN, compared to 0.638 ± 0.165 and 0.643 ± 0.141, respectively, when using the graph-based cosegmentation method. The proposed DFCN cosegmentation using both PET and CT also outperforms the deep learning method using either PET or CT alone. The proposed DFCN cosegmentation is able to outperform existing graph-based segmentation methods. The proposed DFCN cosegmentation shows promise for further integration with quantitative multimodality imaging tools in clinical trials.

Positron Emission Tomography Computed Tomography

Neural Networks, Computer

Time Factors

Humans

Lung Neoplasms - diagnostic imaging

Whole Body Imaging

Carcinoma, Non-Small-Cell Lung - diagnostic imaging

Image Processing, Computer-Assisted - methods

Details

Title: Subtitle: Simultaneous cosegmentation of tumors in PET-CT images using deep fully convolutional networks
Creators: Zisha Zhong - Department of Radiation Oncology, University of Iowa Hospitals and Clinics, Iowa City, IA, 52242, USA
Yusung Kim - Department of Radiation Oncology, University of Iowa Hospitals and Clinics, Iowa City, IA, 52242, USA
Kristin Plichta - Department of Radiation Oncology, University of Iowa Hospitals and Clinics, Iowa City, IA, 52242, USA
Bryan G Allen - Department of Radiation Oncology, University of Iowa Hospitals and Clinics, Iowa City, IA, 52242, USA
Leixin Zhou - Department of Radiation Oncology, University of Iowa Hospitals and Clinics, Iowa City, IA, 52242, USA
John Buatti - Department of Radiation Oncology, University of Iowa Hospitals and Clinics, Iowa City, IA, 52242, USA
Xiaodong Wu - Department of Radiation Oncology, University of Iowa Hospitals and Clinics, Iowa City, IA, 52242, USA
Resource Type: Journal article
Publication Details: Medical physics (Lancaster), Vol.46(2), pp.619-633
DOI: 10.1002/mp.13331
PMID: 30537103
PMCID: PMC6527327
NLM abbreviation: Med Phys
ISSN: 0094-2405
eISSN: 2473-4209
Publisher: United States
Grant note: U01CA140206 / NCI NIH HHS R01 EB004640 / NIBIB NIH HHS R21 CA209874 / NCI NIH HHS UL1 TR002537 / NCATS NIH HHS UL1TR002537 / NCI NIH HHS P01 CA217797 / NCI NIH HHS 1R21CA209874 / NCI NIH HHS P30 CA086862 / NCI NIH HHS U01 CA140206 / NCI NIH HHS
Language: English
Date published: 02/2019
Academic Unit: Electrical and Computer Engineering; Radiation Oncology; Neurosurgery; Otolaryngology
Record Identifier: 9984046812102771

Metrics

23 Record Views

70 Times Cited - Web of Science