Share：
Share this content in WeChat
X
[Chinese][PDF]1661
• Special Focus •
Application value of high-resolution single-shot fast spin-echo ovarian MRI based on deep learning reconstruction in follicle counting
YANG Renjie  ZOU Yujie  LIU Weiyin  LIU Changsheng  WEN Zhi  LI Liang  ZHA Yunfei 

Cite this article as: YANG R J, ZOU Y J, LIU W Y, et al. Application value of high-resolution single-shot fast spin-echo ovarian MRI based on deep learning reconstruction in follicle counting[J]. Chin J Magn Reson Imaging, 2024, 15(10): 50-55, 61. DOI:10.12015/issn.1674-8034.2024.10.009.


[Abstract] Objective To investigate the application value of high-resolution single-shot fast spin-echo (SSFSE) acquisition with deep learning reconstruction (DLR) in follicle counting compared to transvaginal ultrasonography (TVUS), conventional reconstruction (CR) SSFSE and periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) images.Materials and Methods Participants with clinically confirmed or suspected polycystic ovary syndrome (PCOS) were prospectively recruited and underwent ovarian MRI. Those with no history of sexual activity also underwent ovarian TVUS. High-resolution PROPELLER and SSFSE T2-weighted sequences were obtained on three matched planes. The SSFSE sequences implemented both DLR and CR, generating SSFSE-DLR and SSFSE-CR images respectively. Qualitative indicators including blurring artifacts, subjective noise, and conspicuity of follicles were compared using Wilcoxon signed-rank tests. Follicle counting was performed by two observers, with repeatability assessed using intraclass correlation coefficient (ICC) and Bland-Altman method. Absolute values of intra-observer and inter-observer differences were compared using a paired t-test. Follicle count between SSFSE-DL and TVUS was also compared using a paired t-test.Results Twenty-four participants underwent MRI, with 18 of them also undergoing TVUS. Observer 1 assigned higher subjective scores to SSFSE-DLR in comparison to SSFSE-CR and PROPELLER (P<0.05), despite the similar subjective noise observed between SSFSE-DLR and PROPELLER (P>0.05). Observer 2 also rated SSFSE-DLR higher than SSFSE-CR and PROPELLER (P<0.05). Furthermore, SSFSE-DLR demonstrated the best repeatability for follicle counting, achieving the highest ICC, narrowest 95% limits of agreement, and the lowest absolute values of intra-observer and inter-observer differences (P<0.05). Moreover, SSFSE-DL detected more follicles than TVUS (P<0.001).Conclusions SSFSE-DLR images significantly improved the display of ovarian morphology and the repeatability of follicle counting, thereby fortifying the reliability of future polycystic ovary morphology determinations.
[Keywords] polycystic ovary syndrome；polycystic ovary；follicle count；deep learning reconstruction；single-shot fast spin-echo；magnetic resonance imaging

YANG Renjie1   ZOU Yujie2   LIU Weiyin3   LIU Changsheng1   WEN Zhi1   LI Liang1   ZHA Yunfei1*  

1 Department of Radiology, Renmin Hospital of Wuhan University, Wuhan 430060, China

2 Reproductive Medicine Center, Renmin Hospital of Wuhan University, Wuhan 430060, China

3 GE Healthcare (China) Co., Ltd., Beijing 100176, China

Corresponding author: ZHA Y F, E-mail: zhayunfei999@126.com

Conflicts of interest   None.

Received  2024-01-18
Accepted  2024-05-31
DOI: 10.12015/issn.1674-8034.2024.10.009
Cite this article as: YANG R J, ZOU Y J, LIU W Y, et al. Application value of high-resolution single-shot fast spin-echo ovarian MRI based on deep learning reconstruction in follicle counting[J]. Chin J Magn Reson Imaging, 2024, 15(10): 50-55, 61. DOI:10.12015/issn.1674-8034.2024.10.009.

[1]
HUDDLESTON H G, DOKRAS A. Diagnosis and treatment of polycystic ovary syndrome[J]. JAMA, 2022, 327(3): 274-275. DOI: 10.1001/jama.2021.23769.
[2]
WALTER K. What is polycystic ovary syndrome?[J]. JAMA, 2022, 327(3): 294-298. DOI: 10.1001/jama.2021.19776.
[3]
KOSTOPOULOU E, ANAGNOSTIS P, BOSDOU J K, et al. Polycystic ovary Syndrome in Adolescents: pitfalls in Diagnosis and Management[J]. Curr Obes Rep, 2020, 9(3): 193-203. DOI: 10.1007/s13679-020-00388-9.
[4]
ROTTERDAM ESHRE/ASRM-SPONSORED PCOS CONSENSUS WORKSHOP GROUP. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS)[J]. Hum Reprod, 2004, 19(1): 41-47. DOI: 10.1093/humrep/deh098.
[5]
TEEDE H J, MISSO M L, COSTELLO M F, et al. Recommendations from the international evidence-based guideline for the assessment and management of polycystic ovary syndrome[J]. Fertil Steril, 2018, 110(3): 364-379. DOI: 10.1016/j.fertnstert.2018.05.004.
[6]
TEEDE H J, TAY C T, LAVEN J J E, et al. Recommendations from the 2023 international evidence-based guideline for the assessment and management of polycystic ovary syndrome[J]. J Clin Endocrinol Metab, 2023, 108(10): 2447-2469. DOI: 10.1210/clinem/dgad463.
[7]
KENIGSBERG L E, AGARWAL C, SIN S, et al. Clinical utility of magnetic resonance imaging and ultrasonography for diagnosis of polycystic ovary syndrome in adolescent girls[J]. Fertil Steril, 2015, 104(5): 1302-1309. DOI: 10.1016/j.fertnstert.2015.08.002.
[8]
PEREIRA-ESHRAGHI C F, TAO R, CHIUZAN C C, et al. Ovarian follicle count by magnetic resonance imaging is greater in adolescents and young adults with polycystic ovary syndrome than in controls[J]. F S Rep, 2022, 3(2): 102-109. DOI: 10.1016/j.xfre.2022.01.008.
[9]
BARBER T M, ALVEY C, GREENSLADE T, et al. Patterns of ovarian morphology in polycystic ovary syndrome: a study utilising magnetic resonance imaging[J]. Eur Radiol, 2010, 20(5): 1207-1213. DOI: 10.1007/s00330-009-1643-8.
[10]
SADOWSKI E A, MATUREN K E, ROCKALL A, et al. Ovary: mri characterisation and o-rads mri[J/OL]. Br J Radiol, 2021, 94(1125): 20210157 [2023-11-07]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9327753/. DOI: 10.1259/bjr.20210157.
[11]
FUJIMOTO K, KOYAMA T, KEN T M, et al. BLADE acquisition method improves T2-weighted MR images of the female pelvis compared with a standard fast spin-echo sequence[J]. Eur J Radiol, 2011, 80(3): 796-801. DOI: 10.1016/j.ejrad.2010.08.002.
[12]
SAKALA M D, SHAMPAIN K L, WASNIK A P. Advances in MR imaging of the female pelvis[J]. Magn Reson Imaging Clin N Am, 2020, 28(3): 415-431. DOI: 10.1016/j.mric.2020.03.007.
[13]
MISAKA T, ASATO N, ONO Y, et al. Image quality improvement of single-shot turbo spin-echo magnetic resonance imaging of female pelvis using a convolutional neural network[J/OL]. Medicine, 2020, 99(47): e23138 [2023-11-07]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7676607/. DOI: 10.1097/MD.0000000000023138.
[14]
CHAUDHARI A S, FANG Z N, KOGAN F, et al. Super-resolution musculoskeletal MRI using deep learning[J]. Magn Reson Med, 2018, 80(5): 2139-2154. DOI: 10.1002/mrm.27178.
[15]
RECHT M P, ZBONTAR J, SODICKSON D K, et al. Using deep learning to accelerate knee MRI at 3 T: results of an interchangeability study[J]. AJR Am J Roentgenol, 2020, 215(6): 1421-1429. DOI: 10.2214/AJR.20.23313.
[16]
KIM M, KIM H S, PARK J E, et al. Thin-slice pituitary MRI with deep learning-based reconstruction for preoperative prediction of cavernous sinus invasion by pituitary adenoma: a prospective study[J]. AJNR Am J Neuroradiol, 2022, 43(2): 280-285. DOI: 10.3174/ajnr.A7387.
[17]
LEE K L, KESSLER D A, DEZONIE S, et al. Assessment of deep learning-based reconstruction on T2-weighted and diffusion-weighted prostate MRI image quality[J/OL]. Eur J Radiol, 2023, 166: 111017 [2023-11-08]. https://www.ejradiology.com/article/S0720-048X(23)00331-5/fulltext. DOI: 10.1016/j.ejrad.2023.111017.
[18]
TAKENAKA D, OZAWA Y, YAMAMOTO K, et al. Deep learning reconstruction to improve the quality of MR imaging: evaluating the best sequence for T-category assessment in non-small cell lung cancer patients[J/OL]. Magn Reson Med Sci, 2023 [2023-11-08]. https://www.jstage.jst.go.jp/article/mrms/advpub/0/advpub_mp.2023-0068/_article. DOI: 10.2463/mrms.mp.2023-0068.
[19]
HOKAMURA M, UETANI H, NAKAURA T, et al. Exploring the impact of super-resolution deep learning on MR angiography image quality[J]. Neuroradiology, 2024, 66(2): 217-226. DOI: 10.1007/s00234-023-03271-1.
[20]
SNEAG D B, QUELER S C, CAMPBELL G, et al. Optimized 3D brachial plexus MR neurography using deep learning reconstruction[J]. Skeletal Radiol, 2024, 53(4): 779-789. DOI: 10.1007/s00256-023-04484-4.
[21]
LEBEL R M. Performance characterization of a novel deeplearning-based MR image reconstruction pipeline[EB/OL]. [2022-11-07]. https://arxiv.org/ftp/arxiv/papers/2008/2008.06559.pdf.
[22]
CHEN Z L, PAWAR K, EKANAYAKE M, et al. Deep learning for image enhancement and correction in magnetic resonance imaging-state-of-the-art and challenges[J]. J Digit Imaging, 2023, 36(1): 204-230. DOI: 10.1007/s10278-022-00721-9.
[23]
YANG R J, ZOU Y J, LIU W V, et al. High-resolution single-shot fast spin-echo MR imaging with deep learning reconstruction algorithm can improve repeatability and reproducibility of follicle counting[J]. J Clin Med, 2023, 12(9): 3234-3241. DOI: 10.3390/jcm12093234.
[24]
BROWN M, PARK A S, SHAYYA R F, et al. Ovarian imaging by magnetic resonance in adolescent girls with polycystic ovary syndrome and age-matched controls[J]. J Magn Reson Imaging, 2013, 38(3): 689-693. DOI: 10.1002/jmri.23992.
[25]
LANE B F, VANDERMEER F Q, OZ R C, et al. Comparison of sagittal T2-weighted BLADE and fast spin-echo MRI of the female pelvis for motion artifact and lesion detection[J]. AJR Am J Roentgenol, 2011, 197(2): W307-W313. DOI: 10.2214/AJR.10.5918.
[26]
TSUBOYAMA T, ONISHI H, NAKAMOTO A, et al. Impact of deep learning reconstruction combined with a sharpening filter on single-shot fast spin-echo T2-weighted magnetic resonance imaging of the uterus[J]. Invest Radiol, 2022, 57(6): 379-386. DOI: 10.1097/RLI.0000000000000847.
[27]
ZAND K R, REINHOLD C, HAIDER M A, et al. Artifacts and pitfalls in MR imaging of the pelvis[J]. J Magn Reson Imaging, 2007, 26(3): 480-497. DOI: 10.1002/jmri.20996.
[28]
SHANBHOGUE K, TONG A, SMEREKA P, et al. Accelerated single-shot T2-weighted fat-suppressed (FS) MRI of the liver with deep learning-based image reconstruction: qualitative and quantitative comparison of image quality with conventional T2-weighted FS sequence[J]. Eur Radiol, 2021, 31(11): 8447-8457. DOI: 10.1007/s00330-021-08008-3.
[29]
WARY P, HOSSU G, AMBARKI K, et al. Deep learning HASTE sequence compared with T2-weighted BLADE sequence for liver MRI at 3 Tesla: a qualitative and quantitative prospective study[J]. Eur Radiol, 2023, 33(10): 6817-6827. DOI: 10.1007/s00330-023-09693-y.
[30]
KISO K, TSUBOYAMA T, ONISHI H, et al. Effect of deep learning reconstruction on respiratory-triggered T2-weighted MR imaging of the liver: a comparison between the single-shot fast spin-echo and fast spin-echo sequences[J]. Magn Reson Med Sci, 2024, 23(2): 214-224. DOI: 10.2463/mrms.mp.2022-0111.
[31]
CHEN F Y, CHENG J Y, TAVIANI V, et al. Data-driven self-calibration and reconstruction for non-cartesian wave-encoded single-shot fast spin echo using deep learning[J]. J Magn Reson Imaging, 2020, 51(3): 841-853. DOI: 10.1002/jmri.26871.
[32]
DEWAILLY D, LUJAN M E, CARMINA E, et al. Definition and significance of polycystic ovarian morphology: a task force report from the Androgen Excess and Polycystic Ovary Syndrome Society[J]. Hum Reprod Update, 2014, 20(3): 334-352. DOI: 10.1093/humupd/dmt061.
[33]
FONDIN M, RACHAS A, HUYNH V, et al. Polycystic ovary syndrome in adolescents: which MR imaging-based diagnostic criteria?[J]. Radiology, 2017, 285(3): 961-970. DOI: 10.1148/radiol.2017161513.
[34]
LEONHARDT H, GULL B, STENER-VICTORIN E, et al. Ovarian volume and antral follicle count assessed by MRI and transvaginal ultrasonography: a methodological study[J]. Acta Radiol, 2014, 55(2): 248-256. DOI: 10.1177/0284185113495835.
[35]
NYLANDER M, FRØSSING S, BJERRE A H, et al. Ovarian morphology in polycystic ovary syndrome: estimates from 2D and 3D ultrasound and magnetic resonance imaging and their correlation to anti-Müllerian hormone[J]. Acta Radiol, 2017, 58(8): 997-1004. DOI: 10.1177/0284185116676656.
[36]
WANG S J, ZHANG M M, DUAN N, et al. Using transvaginal ultrasonography and MRI to evaluate ovarian volume and follicle count of infertile women: a comparative study[J]. Clin Radiol, 2022, 77(8): 621-627. DOI: 10.1016/j.crad.2022.04.016.

PREV Application of deep learning reconstruction techniques in optimizing breast MRI image quality
NEXT Imaging study of real-time fMRI neurofeedback training based on the nucleus ambiguus to improve obesity
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn