Utility of Deep Learning Super-resolution in the Context of Osteoarthritis MRI Biomarkers

Overview

Journal J Magn Reson Imaging

Date 2019 Jul 18

PMID 31313397

Citations 30

Authors

Akshay S Chaudhari

Kathryn J Stevens

Jeff P Wood

Amit K Chakraborty

Eric K Gibbons

Zhongnan Fang

Arjun D Desai

Jin Hyung Lee

Garry E Gold

Brian A Hargreaves

Affiliations

Soon will be listed here.

Abstract

Background: Super-resolution is an emerging method for enhancing MRI resolution; however, its impact on image quality is still unknown.

Purpose: To evaluate MRI super-resolution using quantitative and qualitative metrics of cartilage morphometry, osteophyte detection, and global image blurring.

Study Type: Retrospective.

Population: In all, 176 MRI studies of subjects at varying stages of osteoarthritis.

Field Strength/sequence: Original-resolution 3D double-echo steady-state (DESS) and DESS with 3× thicker slices retrospectively enhanced using super-resolution and tricubic interpolation (TCI) at 3T.

Assessment: A quantitative comparison of femoral cartilage morphometry was performed for the original-resolution DESS, the super-resolution, and the TCI scans in 17 subjects. A reader study by three musculoskeletal radiologists assessed cartilage image quality, overall image sharpness, and osteophytes incidence in all three sets of scans. A referenceless blurring metric evaluated blurring in all three image dimensions for the three sets of scans.

Statistical Tests: Mann-Whitney U-tests compared Dice coefficients (DC) of segmentation accuracy for the DESS, super-resolution, and TCI images, along with the image quality readings and blurring metrics. Sensitivity, specificity, and diagnostic odds ratio (DOR) with 95% confidence intervals compared osteophyte detection for the super-resolution and TCI images, with the original-resolution as a reference.

Results: DC for the original-resolution (90.2 ± 1.7%) and super-resolution (89.6 ± 2.0%) were significantly higher (P < 0.001) than TCI (86.3 ± 5.6%). Segmentation overlap of super-resolution with the original-resolution (DC = 97.6 ± 0.7%) was significantly higher (P < 0.0001) than TCI overlap (DC = 95.0 ± 1.1%). Cartilage image quality for sharpness and contrast levels, and the through-plane quantitative blur factor for super-resolution images, was significantly (P < 0.001) better than TCI. Super-resolution osteophyte detection sensitivity of 80% (76-82%), specificity of 93% (92-94%), and DOR of 32 (22-46) was significantly higher (P < 0.001) than TCI sensitivity of 73% (69-76%), specificity of 90% (89-91%), and DOR of 17 (13-22).

Data Conclusion: Super-resolution appears to consistently outperform naïve interpolation and may improve image quality without biasing quantitative biomarkers.

Level Of Evidence: 2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2020;51:768-779.

Citing Articles

Clinical validation of enhanced CT imaging for distal radius fractures through conditional Generative Adversarial Networks (cGAN).

Kim H, Ryu S, Keum J, Oh S, Kim K, Shin Y PLoS One. 2024; 19(8):e0308346.

PMID: 39150966 PMC: 11329132. DOI: 10.1371/journal.pone.0308346.

Super-resolution of clinical CT: Revealing microarchitecture in whole bone clinical CT image data.

Frazer L, Louis N, Zbijewski W, Vaishnav J, Clark K, Nicolella D Bone. 2024; 185:117115.

PMID: 38740120 PMC: 11176006. DOI: 10.1016/j.bone.2024.117115.

Clinical Super-Resolution Computed Tomography of Bone Microstructure: Application in Musculoskeletal and Dental Imaging.

Rytky S, Tiulpin A, Finnila M, Karhula S, Sipola A, Kurttila V Ann Biomed Eng. 2024; 52(5):1255-1269.

PMID: 38361137 PMC: 10995025. DOI: 10.1007/s10439-024-03450-y.

3D-DESS MRI with CAIPIRINHA two- and fourfold acceleration for quantitatively assessing knee cartilage morphology.

Wen D, Zhou X, Hou B, Zhang Q, Raithel E, Wang Y Skeletal Radiol. 2024; 53(8):1481-1494.

PMID: 38347270 DOI: 10.1007/s00256-024-04605-7.

Exploring the impact of super-resolution deep learning on MR angiography image quality.

Hokamura M, Uetani H, Nakaura T, Matsuo K, Morita K, Nagayama Y Neuroradiology. 2023; 66(2):217-226.

PMID: 38148334 DOI: 10.1007/s00234-023-03271-1.

References

Felson D, Niu J, Guermazi A, Sack B, Aliabadi P . Defining radiographic incidence and progression of knee osteoarthritis: suggested modifications of the Kellgren and Lawrence scale. Ann Rheum Dis. 2011; 70(11):1884-6. PMC: 3653624. DOI: 10.1136/ard.2011.155119. View

Wallace I, Worthington S, Felson D, Jurmain R, Wren K, Maijanen H . Knee osteoarthritis has doubled in prevalence since the mid-20th century. Proc Natl Acad Sci U S A. 2017; 114(35):9332-9336. PMC: 5584421. DOI: 10.1073/pnas.1703856114. View

Vos T, Flaxman A, Naghavi M, Lozano R, Michaud C, Ezzati M . Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012; 380(9859):2163-96. PMC: 6350784. DOI: 10.1016/S0140-6736(12)61729-2. View

Chaudhari A, Fang Z, Kogan F, Wood J, Stevens K, Gibbons E . Super-resolution musculoskeletal MRI using deep learning. Magn Reson Med. 2018; 80(5):2139-2154. PMC: 6107420. DOI: 10.1002/mrm.27178. View

Chaudhari A, Black M, Eijgenraam S, Wirth W, Maschek S, Sveinsson B . Five-minute knee MRI for simultaneous morphometry and T relaxometry of cartilage and meniscus and for semiquantitative radiological assessment using double-echo in steady-state at 3T. J Magn Reson Imaging. 2017; 47(5):1328-1341. PMC: 5899635. DOI: 10.1002/jmri.25883. View

Blaimer M, Breuer F, Mueller M, Heidemann R, Griswold M, Jakob P . SMASH, SENSE, PILS, GRAPPA: how to choose the optimal method. Top Magn Reson Imaging. 2004; 15(4):223-36. DOI: 10.1097/01.rmr.0000136558.09801.dd. View

Peterfy C, Schneider E, Nevitt M . The osteoarthritis initiative: report on the design rationale for the magnetic resonance imaging protocol for the knee. Osteoarthritis Cartilage. 2008; 16(12):1433-41. PMC: 3048821. DOI: 10.1016/j.joca.2008.06.016. View

Hunter D, Guermazi A, Lo G, Grainger A, Conaghan P, Boudreau R . Evolution of semi-quantitative whole joint assessment of knee OA: MOAKS (MRI Osteoarthritis Knee Score). Osteoarthritis Cartilage. 2011; 19(8):990-1002. PMC: 4058435. DOI: 10.1016/j.joca.2011.05.004. View

Ding C, Garnero P, Cicuttini F, Scott F, Cooley H, Jones G . Knee cartilage defects: association with early radiographic osteoarthritis, decreased cartilage volume, increased joint surface area and type II collagen breakdown. Osteoarthritis Cartilage. 2005; 13(3):198-205. DOI: 10.1016/j.joca.2004.11.007. View

10.

Kohl S, Meier S, Ahmad S, Bonel H, Exadaktylos A, Krismer A . Accuracy of cartilage-specific 3-Tesla 3D-DESS magnetic resonance imaging in the diagnosis of chondral lesions: comparison with knee arthroscopy. J Orthop Surg Res. 2015; 10:191. PMC: 4696275. DOI: 10.1186/s13018-015-0326-1. View

11.

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

12.

Lustig M, Donoho D, Pauly J . Sparse MRI: The application of compressed sensing for rapid MR imaging. Magn Reson Med. 2007; 58(6):1182-95. DOI: 10.1002/mrm.21391. View

13.

Eckstein F, Hudelmaier M, Wirth W, Kiefer B, Jackson R, Yu J . Double echo steady state magnetic resonance imaging of knee articular cartilage at 3 Tesla: a pilot study for the Osteoarthritis Initiative. Ann Rheum Dis. 2005; 65(4):433-41. PMC: 1798091. DOI: 10.1136/ard.2005.039370. View

14.

Baum T, Joseph G, Karampinos D, Jungmann P, Link T, Bauer J . Cartilage and meniscal T2 relaxation time as non-invasive biomarker for knee osteoarthritis and cartilage repair procedures. Osteoarthritis Cartilage. 2013; 21(10):1474-84. PMC: 3929642. DOI: 10.1016/j.joca.2013.07.012. View

15.

Eckstein F, Kwoh C, Link T . Imaging research results from the osteoarthritis initiative (OAI): a review and lessons learned 10 years after start of enrolment. Ann Rheum Dis. 2014; 73(7):1289-300. DOI: 10.1136/annrheumdis-2014-205310. View

16.

Busse R, Hariharan H, Vu A, Brittain J . Fast spin echo sequences with very long echo trains: design of variable refocusing flip angle schedules and generation of clinical T2 contrast. Magn Reson Med. 2006; 55(5):1030-7. DOI: 10.1002/mrm.20863. View

17.

Chaudhari A, Stevens K, Sveinsson B, Wood J, Beaulieu C, Oei E . Combined 5-minute double-echo in steady-state with separated echoes and 2-minute proton-density-weighted 2D FSE sequence for comprehensive whole-joint knee MRI assessment. J Magn Reson Imaging. 2018; 49(7):e183-e194. PMC: 7850298. DOI: 10.1002/jmri.26582. View

18.

Mardani M, Gong E, Cheng J, Vasanawala S, Zaharchuk G, Xing L . Deep Generative Adversarial Neural Networks for Compressive Sensing MRI. IEEE Trans Med Imaging. 2018; 38(1):167-179. PMC: 6542360. DOI: 10.1109/TMI.2018.2858752. View

19.

Neogi T, Bowes M, Niu J, de Souza K, Vincent G, Goggins J . Magnetic resonance imaging-based three-dimensional bone shape of the knee predicts onset of knee osteoarthritis: data from the osteoarthritis initiative. Arthritis Rheum. 2013; 65(8):2048-58. PMC: 3729737. DOI: 10.1002/art.37987. View

20.

Heidemann R, Ozsarlak O, Parizel P, Michiels J, Kiefer B, Jellus V . A brief review of parallel magnetic resonance imaging. Eur Radiol. 2003; 13(10):2323-37. DOI: 10.1007/s00330-003-1992-7. View