PLOSL: Population learning followed by one shot learning pulmonary image registration using tissue volume preserving and vesselness constraints.

TitlePLOSL: Population learning followed by one shot learning pulmonary image registration using tissue volume preserving and vesselness constraints.
Publication TypePublication
Year2022
AuthorsWang D, Pan Y, Durumeric OC, Reinhardt JM, Hoffman EA, Schroeder JD, Christensen GE
JournalMed Image Anal
Volume79
Pagination102434
Date Published2022 07
ISSN1361-8423
KeywordsAlgorithms, Humans, Image Processing, Computer-Assisted, Lipodystrophy, Lung, Osteochondrodysplasias, Subacute Sclerosing Panencephalitis, Tomography, X-Ray Computed
Abstract

This paper presents the Population Learning followed by One Shot Learning (PLOSL) pulmonary image registration method. PLOSL is a fast unsupervised learning-based framework for 3D-CT pulmonary image registration algorithm based on combining population learning (PL) and one-shot learning (OSL). The PLOSL image registration has the advantages of the PL and OSL approaches while reducing their respective drawbacks. The advantages of PLOSL include improved performance over PL, substantially reducing OSL training time and reducing the likelihood of OSL getting stuck in local minima. PLOSL pulmonary image registration uses tissue volume preserving and vesselness constraints for registration of inspiration-to-expiration and expiration-to-inspiration pulmonary CT images. A coarse-to-fine convolution encoder-decoder CNN architecture is used to register large and small shape features. During training, the sum of squared tissue volume difference (SSTVD) compensates for intensity differences between inspiration and expiration computed tomography (CT) images and the sum of squared vesselness measure difference (SSVMD) helps match the lung vessel tree. Results show that the PLOSL (SSTVD+SSVMD) algorithm achieved subvoxel landmark error while preserving pulmonary topology on the SPIROMICS data set, the public DIR-LAB COPDGene and 4DCT data sets.

DOI10.1016/j.media.2022.102434
Alternate JournalMed Image Anal
PubMed ID35430476
Grant ListHHSN268200900019C / HL / NHLBI NIH HHS / United States
P30 ES005605 / ES / NIEHS NIH HHS / United States
HHSN268200900015C / HL / NHLBI NIH HHS / United States
HHSN268200900014C / HL / NHLBI NIH HHS / United States
HHSN268200900018C / HL / NHLBI NIH HHS / United States
U24 HL141762 / HL / NHLBI NIH HHS / United States
HHSN268200900013C / HL / NHLBI NIH HHS / United States
HHSN268200900016C / HL / NHLBI NIH HHS / United States
U01 HL137880 / HL / NHLBI NIH HHS / United States
HHSN268200900020C / HL / NHLBI NIH HHS / United States
HHSN268200900017C / HL / NHLBI NIH HHS / United States
MS#: 
MS261
Manuscript Full Title: 
PLOSL: Population learning followed by one shot learning pulmonary image registration using tissue volume preserving and vesselness constraints.
Manuscript Lead/Corresponding Author Affiliation: 
Reading Center: Radiology (University of Iowa)
ECI: 
Manuscript Status: 
Published and Public