Deep Convolutional Neural Network for Dedicated Regions-of-Interest Based Multi-Parameter Quantitative Ultrashort Echo Time (UTE) Magnetic Resonance Imaging of the Knee Joint

Overview

Journal J Imaging Inform Med

Publisher Springer Nature

Specialty Radiology

Date 2024 Mar 29

PMID 38548992

Authors

Xing Lu

Yajun Ma

Eric Y Chang

Jiyo Athertya

Hyungseok Jang

Saeed Jerban

Dana C Covey

Susan Bukata

Christine B Chung

Jiang Du

Affiliations

Soon will be listed here.

Abstract

We proposed an end-to-end deep learning convolutional neural network (DCNN) for region-of-interest based multi-parameter quantification (RMQ-Net) to accelerate quantitative ultrashort echo time (UTE) MRI of the knee joint with automatic multi-tissue segmentation and relaxometry mapping. The study involved UTE-based T1 (UTE-T1) and Adiabatic T1ρ (UTE-AdiabT1ρ) mapping of the knee joint of 65 human subjects, including 20 normal controls, 29 with doubtful-minimal osteoarthritis (OA), and 16 with moderate-severe OA. Comparison studies were performed on UTE-T1 and UTE-AdiabT1ρ measurements using 100%, 43%, 26%, and 18% UTE MRI data as the inputs and the effects on the prediction quality of the RMQ-Net. The RMQ-net was modified and retrained accordingly with different combinations of inputs. Both ROI-based and voxel-based Pearson correlation analyses were performed. High Pearson correlation coefficients were achieved between the RMQ-Net predicted UTE-T1 and UTE-AdiabT1ρ results and the ground truth for segmented cartilage with acceleration factors ranging from 2.3 to 5.7. With an acceleration factor of 5.7, the Pearson r-value achieved 0.908 (ROI-based) and 0.945 (voxel-based) for UTE-T1, and 0.733 (ROI-based) and 0.895 (voxel-based) for UTE-AdiabT1ρ, correspondingly. The results demonstrated that RMQ-net can significantly accelerate quantitative UTE imaging with automated segmentation of articular cartilage in the knee joint.

References

Klein S, Staring M, Murphy K, Viergever M, Pluim J . elastix: a toolbox for intensity-based medical image registration. IEEE Trans Med Imaging. 2009; 29(1):196-205. DOI: 10.1109/TMI.2009.2035616. View

Keegan J, Slabaugh G, Arridge S, Ye X, Guo Y, Yu S . DAGAN: Deep De-Aliasing Generative Adversarial Networks for Fast Compressed Sensing MRI Reconstruction. IEEE Trans Med Imaging. 2018; 37(6):1310-1321. DOI: 10.1109/TMI.2017.2785879. View

Ma Y, Lu X, Carl M, Zhu Y, Szeverenyi N, Bydder G . Accurate T mapping of short T tissues using a three-dimensional ultrashort echo time cones actual flip angle imaging-variable repetition time (3D UTE-Cones AFI-VTR) method. Magn Reson Med. 2018; 80(2):598-608. PMC: 5912804. DOI: 10.1002/mrm.27066. View

De Moura H, Menon R, Zibetti M, Regatte R . Optimization of spin-lock times for T mapping of human knee cartilage with bi- and stretched-exponential models. Sci Rep. 2022; 12(1):16829. PMC: 9546896. DOI: 10.1038/s41598-022-21269-2. View

Hayashi D, Guermazi A, Hunter D . Osteoarthritis year 2010 in review: imaging. Osteoarthritis Cartilage. 2011; 19(4):354-60. DOI: 10.1016/j.joca.2011.02.003. View

Ma Y, Carl M, Searleman A, Lu X, Chang E, Du J . 3D adiabatic T prepared ultrashort echo time cones sequence for whole knee imaging. Magn Reson Med. 2018; 80(4):1429-1439. PMC: 6097905. DOI: 10.1002/mrm.27131. View

Chu C, Williams A, West R, Qian Y, Fu F, Do B . Quantitative Magnetic Resonance Imaging UTE-T2* Mapping of Cartilage and Meniscus Healing After Anatomic Anterior Cruciate Ligament Reconstruction. Am J Sports Med. 2014; 42(8):1847-56. PMC: 5278879. DOI: 10.1177/0363546514532227. View

Ma Y, Carl M, Shao H, Tadros A, Chang E, Du J . Three-dimensional ultrashort echo time cones T (3D UTE-cones-T ) imaging. NMR Biomed. 2017; 30(6). PMC: 5505275. DOI: 10.1002/nbm.3709. View

Williams A, Qian Y, Bear D, Chu C . Assessing degeneration of human articular cartilage with ultra-short echo time (UTE) T2* mapping. Osteoarthritis Cartilage. 2010; 18(4):539-46. PMC: 2846232. DOI: 10.1016/j.joca.2010.02.001. View

10.

Ma Y, Jang H, Jerban S, Chang E, Chung C, Bydder G . Making the invisible visible-ultrashort echo time magnetic resonance imaging: Technical developments and applications. Appl Phys Rev. 2022; 9(4):041303. PMC: 9677812. DOI: 10.1063/5.0086459. View

11.

Brandt K, Radin E, Dieppe P, van de Putte L . Yet more evidence that osteoarthritis is not a cartilage disease. Ann Rheum Dis. 2006; 65(10):1261-4. PMC: 1798332. DOI: 10.1136/ard.2006.058347. View

12.

Liu F, Zhou Z, Jang H, Samsonov A, Zhao G, Kijowski R . Deep convolutional neural network and 3D deformable approach for tissue segmentation in musculoskeletal magnetic resonance imaging. Magn Reson Med. 2017; 79(4):2379-2391. PMC: 6271435. DOI: 10.1002/mrm.26841. View

13.

Liu F, Feng L, Kijowski R . MANTIS: Model-Augmented Neural neTwork with Incoherent k-space Sampling for efficient MR parameter mapping. Magn Reson Med. 2019; 82(1):174-188. PMC: 7144418. DOI: 10.1002/mrm.27707. View

14.

Sui B, Lv J, Tong X, Li Y, Wang C . Simultaneous image reconstruction and lesion segmentation in accelerated MRI using multitasking learning. Med Phys. 2021; 48(11):7189-7198. DOI: 10.1002/mp.15213. View

15.

Ma Y, Zhao W, Wan L, Guo T, Searleman A, Jang H . Whole knee joint T values measured in vivo at 3T by combined 3D ultrashort echo time cones actual flip angle and variable flip angle methods. Magn Reson Med. 2018; 81(3):1634-1644. PMC: 6347520. DOI: 10.1002/mrm.27510. View

16.

Chang E, Du J, Chung C . UTE imaging in the musculoskeletal system. J Magn Reson Imaging. 2014; 41(4):870-83. PMC: 4297256. DOI: 10.1002/jmri.24713. View

17.

Wang X, Jiang L, Li L, Xu M, Deng X, Dai L . Joint Learning of 3D Lesion Segmentation and Classification for Explainable COVID-19 Diagnosis. IEEE Trans Med Imaging. 2021; 40(9):2463-2476. PMC: 8544955. DOI: 10.1109/TMI.2021.3079709. View

18.

Abramson R, Su P, Shyr Y . Quantitative metrics in clinical radiology reporting: a snapshot perspective from a single mixed academic-community practice. Magn Reson Imaging. 2012; 30(9):1357-66. PMC: 3466403. DOI: 10.1016/j.mri.2012.04.018. View

19.

Ma Y, Chang E, Carl M, Du J . Quantitative magnetization transfer ultrashort echo time imaging using a time-efficient 3D multispoke Cones sequence. Magn Reson Med. 2017; 79(2):692-700. PMC: 5670020. DOI: 10.1002/mrm.26716. View

20.

Virtanen P, Gommers R, Oliphant T, Haberland M, Reddy T, Cournapeau D . SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods. 2020; 17(3):261-272. PMC: 7056644. DOI: 10.1038/s41592-019-0686-2. View