本研究探討心肌T1映射技術中常見的問題，即呼吸引起MOLLI影像之間的錯位，進而影響T1映射的準確性，需要進行運動校正。我們提出了一種解決方案，透過使用生成對抗網絡來生成一致心臟位置的虛擬MOLLI影像，並將其用作影像對位算法的目標，解決多對比度運動校正問題。我們先在轉移學習資料集上進行預訓練，再微調該模型以適應MOLLI資料集。我們以運動校正影像的擬合質量指標作為評估效能的指標，比較使用四種不同方法進行運動校正的效果。結果顯示，我們提出的方法優於供應商提供的方法，並具有提高心肌T1映射的潛力。此外，我們提供了一個開源軟體，讓研究人員可以根據自己的需求使用此軟體。該方法為MOLLI資料集的回顧性對位提供了一個實用的解決方案，為患有嚴重疾病或呼吸困難的患者產生穩健而準確的原生T1映射。

Among different myocardial T1 mapping techniques, the modified Look-Locker inversion recovery (MOLLI) sequence is the most widely used. However, respiratory-induced misregistration between MOLLI images can cause inaccuracies in T1 maps, necessitating motion correction procedures. In this study, we introduce an approach to address this challenge by employing a generative adversarial network to generate virtual MOLLI images with consistent heart positions. The model was pretrained on a transfer-learning database and then fine-tuned for MOLLI datasets. The generated images were used as targets for the image registration algorithm, tackling the multi-contrast motion correction problem. The performance of the method was evaluated by comparing the fitting quality index (FQI) of motion-corrected images using four different methods. The results showed that the proposed method outperformed the vendor-equipped method, demonstrating its potential for improving myocardial T1 mapping. Our study provides an open-source repository, enabling researchers to adapt the software to their specific needs. This method offers a practical solution for retrospective alignment of MOLLI datasets, producing robust and accurate native T1 mapping for patients with severe illnesses or difficulties in breath-holding.

Reference
1. Pagano JJ, Chow K, Paterson DI, et al. Effects of age, gender, and risk-factors for heart failure on native myocardial T1 and extracellular volume fraction using the SASHA sequence at 1.5T. J Magn Reson Imaging. Nov 2018;48(5):1307-1317. doi:10.1002/jmri.26160
2. Panovský R, Doubková M, Holeček T, et al. Myocardial T1 mapping using SMART1Map and MOLLI mapping in asymptomatic patients with recent extracardiac sarcoidosis. NMR in Biomedicine. 2020;33(11):e4388. doi:https://doi.org/10.1002/nbm.4388
3. Schelbert EB, Messroghli DR. State of the Art: Clinical Applications of Cardiac T1 Mapping. Radiology. Mar 2016;278(3):658-76. doi:10.1148/radiol.2016141802
4. Zhang Q, Burrage MK, Lukaschuk E, et al. Toward Replacing Late Gadolinium Enhancement With Artificial Intelligence Virtual Native Enhancement for Gadolinium-Free Cardiovascular Magnetic Resonance Tissue Characterization in Hypertrophic Cardiomyopathy. Circulation. 2021;144(8):589-599. doi:doi:10.1161/CIRCULATIONAHA.121.054432
5. Messroghli DR, Walters K, Plein S, et al. Myocardial T1 mapping: application to patients with acute and chronic myocardial infarction. Magn Reson Med. 2007;58doi:10.1002/mrm.21272
6. Puntmann VO, D’Cruz D, Smith Z, et al. Native myocardial T1 mapping by cardiovascular magnetic resonance imaging in subclinical cardiomyopathy in patients with systemic lupus erythematosus. Circ Cardiovasc Imaging. 2013;6doi:10.1161/circimaging.112.000151
7. Puntmann VO, Voigt T, Chen Z, et al. Native T1 mapping in differentiation of normal myocardium from diffuse disease in hypertrophic and dilated cardiomyopathy. JACC Cardiovasc Imaging. 2013;6doi:10.1016/j.jcmg.2012.08.019
8. Messroghli DR, Radjenovic A, Kozerke S, Higgins DM, Sivananthan MU, Ridgway JP. Modified Look-Locker inversion recovery (MOLLI) for high-resolution T1 mapping of the heart. Magn Reson Med. Jul 2004;52(1):141-6. doi:10.1002/mrm.20110
9. Piechnik SK, Ferreira VM, Dall’Armellina E, et al. Shortened Modified Look-Locker Inversion recovery (ShMOLLI) for clinical myocardial T1-mapping at 1.5 and 3 T within a 9 heartbeat breathhold. J Cardiovasc Magn Reson. 2010;12doi:10.1186/1532-429x-12-69
10. Tsai JM, Huang TY, Tseng YS, Lin YR. Free-breathing MOLLI: application to myocardial T(1) mapping. Med Phys. Dec 2012;39(12):7291-302. doi:10.1118/1.4764915
11. Huang TY, Tseng YS, Tang YW, Lin YR. Optimization of PROPELLER reconstruction for free-breathing T1-weighted cardiac imaging. Med Phys. Aug 2012;39(8):4896-902. doi:10.1118/1.4736977
12. Chow K, Flewitt JA, Green JD, Pagano JJ, Friedrich MG, Thompson RB. Saturation recovery single-shot acquisition (SASHA) for myocardial T1 mapping. Magn Reson Med. 2013.
13. Weingartner S, Messner NM, Budjan J, et al. Myocardial T1-mapping at 3T using saturation-recovery: reference values, precision and comparison with MOLLI. J Cardiovasc Magn Reson. Nov 18 2016;18(1):84. doi:10.1186/s12968-016-0302-x
14. Yu CY, Huang TY, Chung HW. Single breath-hold MR T1 mapping in the heart: Hybrid MOLLI combining saturation and inversion recovery. Magn Reson Imaging. Feb 2023;96:85-92. doi:10.1016/j.mri.2022.12.001
15. Kellman P, Wilson JR, Xue H, Ugander M, Arai AE. Extracellular volume fraction mapping in the myocardium, part 1: evaluation of an automated method. J Cardiovasc Magn Reson. Sep 10 2012;14:63. doi:10.1186/1532-429X-14-63
16. Kellman P, Wilson JR, Xue H, et al. Extracellular volume fraction mapping in the myocardium, part 2: initial clinical experience. J Cardiovasc Magn Reson. 2012;14doi:10.1186/1532-429x-14-64
17. Look DC, Locker DR. Time saving in measurement of NMR and EPR relaxation times. Review of Scientific Instruments. 1970;41(2):250-251.
18. Xue H, Shah S, Greiser A, et al. Motion correction for myocardial T1 mapping using image registration with synthetic image estimation. Magn Reson Med. Jun 2012;67(6):1644-55. doi:10.1002/mrm.23153
19. Xue H, Greiser A, Zuehlsdorff S, et al. Phase-sensitive inversion recovery for myocardial T(1) mapping with motion correction and parametric fitting. Magn Reson Med. 2013;69doi:10.1002/mrm.24385
20. Campello VM, Gkontra P, Izquierdo C, et al. Multi-Centre, Multi-Vendor and Multi-Disease Cardiac Segmentation: The M&Ms Challenge. IEEE Transactions on Medical Imaging. 2021;40(12):3543-3554. doi:10.1109/TMI.2021.3090082
21. Bernard O, Lalande A, Zotti C, et al. Deep Learning Techniques for Automatic MRI Cardiac Multi-Structures Segmentation and Diagnosis: Is the Problem Solved? IEEE Trans Med Imaging. Nov 2018;37(11):2514-2525. doi:10.1109/TMI.2018.2837502
22. Wuttke J. lmfit – a C library for Levenberg-Marquardt least-squares minimization and curve fitting. Accessed 2023/4/7, 2023. https://jugit.fz-juelich.de/mlz/lmfit
23. Isola P, Zhu JY, Zhou T, Efros AA. Image-to-Image Translation with Conditional Adversarial Networks. 2017:5967-5976.
24. Armanious K, Jiang C, Fischer M, et al. MedGAN: Medical image translation using GANs. Comput Med Imaging Graph. Jan 2020;79:101684. doi:10.1016/j.compmedimag.2019.101684
25. Avants BB, Tustison NJ, Stauffer M, Song G, Wu B, Gee JC. The Insight ToolKit image registration framework. Front Neuroinform. 2014;8:44. doi:10.3389/fninf.2014.00044
26. Lowekamp BC, Chen DT, Ibáñez L, Blezek D. The design of SimpleITK. Frontiers in neuroinformatics. 2013;7:45.
27. Woo J, Stone M, Prince JL. Multimodal Registration via Mutual Information Incorporating Geometric and Spatial Context. IEEE Transactions on Image Processing. 2015;24(2):757-769. doi:10.1109/TIP.2014.2387019
28. Mattes D, Haynor D, Vesselle H, Lewellyn T, Eubank W. Nonrigid multimodality image registration. vol 4322. Medical Imaging 2001. SPIE; 2001.
29. Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. Springer; 2015:234-241.
30. Raisi-Estabragh Z, Harvey NC, Neubauer S, Petersen SE. Cardiovascular magnetic resonance imaging in the UK Biobank: a major international health research resource. Eur Heart J Cardiovasc Imaging. Feb 22 2021;22(3):251-258. doi:10.1093/ehjci/jeaa297
31. Blaha MJ, DeFilippis AP. Multi-Ethnic Study of Atherosclerosis (MESA): JACC Focus Seminar 5/8. J Am Coll Cardiol. Jun 29 2021;77(25):3195-3216. doi:10.1016/j.jacc.2021.05.006