Akshay S Chaudhari1,2, Zhongnan Fang3, Feliks Kogan1, Jeff Wood1, Kathryn J Stevens1,4, Eric K Gibbons5, Jin Hyung Lee2,3,6,7, Garry E Gold1,2,4, Brian A Hargreaves1,2,7. 1. Department of Radiology, Stanford University, Stanford, California. 2. Department of Bioengineering, Stanford University, Stanford, California. 3. LVIS Corporation, Palo Alto, California. 4. Department of Orthopedic Surgery, Stanford University, Stanford, California. 5. Department of Radiology and Imaging Sciences, University of Utah, Salt Lake City, Utah. 6. Department of Neurology and Neurological Sciences, Stanford University, Stanford, California. 7. Department of Electrical Engineering, Stanford University, Stanford, California.
Abstract
PURPOSE: To develop a super-resolution technique using convolutional neural networks for generating thin-slice knee MR images from thicker input slices, and compare this method with alternative through-plane interpolation methods. METHODS: We implemented a 3D convolutional neural network entitled DeepResolve to learn residual-based transformations between high-resolution thin-slice images and lower-resolution thick-slice images at the same center locations. DeepResolve was trained using 124 double echo in steady-state (DESS) data sets with 0.7-mm slice thickness and tested on 17 patients. Ground-truth images were compared with DeepResolve, clinically used tricubic interpolation, and Fourier interpolation methods, along with state-of-the-art single-image sparse-coding super-resolution. Comparisons were performed using structural similarity, peak SNR, and RMS error image quality metrics for a multitude of thin-slice downsampling factors. Two musculoskeletal radiologists ranked the 3 data sets and reviewed the diagnostic quality of the DeepResolve, tricubic interpolation, and ground-truth images for sharpness, contrast, artifacts, SNR, and overall diagnostic quality. Mann-Whitney U tests evaluated differences among the quantitative image metrics, reader scores, and rankings. Cohen's Kappa (κ) evaluated interreader reliability. RESULTS: DeepResolve had significantly better structural similarity, peak SNR, and RMS error than tricubic interpolation, Fourier interpolation, and sparse-coding super-resolution for all downsampling factors (p < .05, except 4 × and 8 × sparse-coding super-resolution downsampling factors). In the reader study, DeepResolve significantly outperformed (p < .01) tricubic interpolation in all image quality categories and overall image ranking. Both readers had substantial scoring agreement (κ = 0.73). CONCLUSION: DeepResolve was capable of resolving high-resolution thin-slice knee MRI from lower-resolution thicker slices, achieving superior quantitative and qualitative diagnostic performance to both conventionally used and state-of-the-art methods.
PURPOSE: To develop a super-resolution technique using convolutional neural networks for generating thin-slice knee MR images from thicker input slices, and compare this method with alternative through-plane interpolation methods. METHODS: We implemented a 3D convolutional neural network entitled DeepResolve to learn residual-based transformations between high-resolution thin-slice images and lower-resolution thick-slice images at the same center locations. DeepResolve was trained using 124 double echo in steady-state (DESS) data sets with 0.7-mm slice thickness and tested on 17 patients. Ground-truth images were compared with DeepResolve, clinically used tricubic interpolation, and Fourier interpolation methods, along with state-of-the-art single-image sparse-coding super-resolution. Comparisons were performed using structural similarity, peak SNR, and RMS error image quality metrics for a multitude of thin-slice downsampling factors. Two musculoskeletal radiologists ranked the 3 data sets and reviewed the diagnostic quality of the DeepResolve, tricubic interpolation, and ground-truth images for sharpness, contrast, artifacts, SNR, and overall diagnostic quality. Mann-Whitney U tests evaluated differences among the quantitative image metrics, reader scores, and rankings. Cohen's Kappa (κ) evaluated interreader reliability. RESULTS: DeepResolve had significantly better structural similarity, peak SNR, and RMS error than tricubic interpolation, Fourier interpolation, and sparse-coding super-resolution for all downsampling factors (p < .05, except 4 × and 8 × sparse-coding super-resolution downsampling factors). In the reader study, DeepResolve significantly outperformed (p < .01) tricubic interpolation in all image quality categories and overall image ranking. Both readers had substantial scoring agreement (κ = 0.73). CONCLUSION: DeepResolve was capable of resolving high-resolution thin-slice knee MRI from lower-resolution thicker slices, achieving superior quantitative and qualitative diagnostic performance to both conventionally used and state-of-the-art methods.
Authors: Akshay S Chaudhari; Bragi Sveinsson; Catherine J Moran; Emily J McWalter; Ethan M Johnson; Tao Zhang; Garry E Gold; Brian A Hargreaves Journal: Magn Reson Med Date: 2017-01-11 Impact factor: 4.668
Authors: Shanshan Bao; Jonathan I Tamir; Jeffrey L Young; Umar Tariq; Martin Uecker; Peng Lai; Weitian Chen; Michael Lustig; Shreyas S Vasanawala Journal: J Magn Reson Imaging Date: 2016-10-11 Impact factor: 4.813
Authors: Kerstin Hammernik; Teresa Klatzer; Erich Kobler; Michael P Recht; Daniel K Sodickson; Thomas Pock; Florian Knoll Journal: Magn Reson Med Date: 2017-11-08 Impact factor: 4.668
Authors: Richard Kijowski; Donna G Blankenbaker; Jessica L Klaers; Kazuhiko Shinki; Arthur A De Smet; Walter F Block Journal: Radiology Date: 2009-02-12 Impact factor: 11.105
Authors: Matthew J Muckley; Benjamin Ades-Aron; Antonios Papaioannou; Gregory Lemberskiy; Eddy Solomon; Yvonne W Lui; Daniel K Sodickson; Els Fieremans; Dmitry S Novikov; Florian Knoll Journal: Magn Reson Med Date: 2020-07-14 Impact factor: 4.668
Authors: Akshay S Chaudhari; Kathryn J Stevens; Bragi Sveinsson; Jeff P Wood; Christopher F Beaulieu; Edwin H G Oei; Jarrett K Rosenberg; Feliks Kogan; Marcus T Alley; Garry E Gold; Brian A Hargreaves Journal: J Magn Reson Imaging Date: 2018-12-23 Impact factor: 4.813
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 Journal: J Magn Reson Imaging Date: 2019-07-16 Impact factor: 4.813
Authors: Indranil Guha; Syed Ahmed Nadeem; Chenyu You; Xiaoliu Zhang; Steven M Levy; Ge Wang; James C Torner; Punam K Saha Journal: Proc SPIE Int Soc Opt Eng Date: 2020-02-28
Authors: Qing Lyu; Hongming Shan; Cole Steber; Corbin Helis; Chris Whitlow; Michael Chan; Ge Wang Journal: IEEE Trans Med Imaging Date: 2020-02-18 Impact factor: 10.048