Research Article

Evaluation of Spatial Attentive Deep Learning for Automatic Placental Segmentation on Longitudinal MRI

Corresponding Author

Yongkai Liu PhD

[email protected]

orcid.org/0000-0003-4478-4006

Department of Radiological Sciences, University of California, Los Angeles, California, USA

Physics and Biology in Medicine IDP, David Geffen School of Medicine, University of California, Los Angeles, California, USA

Address reprint requests to: Y.L., Department of Radiological Sciences, 300 UCLA Medical Plaza, Suite B114, Los Angeles, California 90095, USA. E-mail: [email protected]

Search for more papers by this author

Fatemeh Zabihollahy PhD,

Fatemeh Zabihollahy PhD

Department of Radiological Sciences, University of California, Los Angeles, California, USA

Search for more papers by this author

Ran Yan MS,

Ran Yan MS

orcid.org/0000-0001-8204-8654

Department of Radiological Sciences, University of California, Los Angeles, California, USA

Department of Bioengineering, Henry Samueli School of Engineering, University of California, Los Angeles, California, USA

Search for more papers by this author

Brian Lee MD,

Brian Lee MD

Department of Pediatrics, David Geffen School of Medicine, University of California, Los Angeles, California, USA

Search for more papers by this author

Carla Janzen MD, PhD,

Carla Janzen MD, PhD

Department of Obstetrics and Gynecology, David Geffen School of Medicine, University of California, Los Angeles, California, USA

Search for more papers by this author

Sherin U. Devaskar MD,

Sherin U. Devaskar MD

Department of Pediatrics, David Geffen School of Medicine, University of California, Los Angeles, California, USA

Search for more papers by this author

Kyunghyun Sung PhD,

Kyunghyun Sung PhD

orcid.org/0000-0003-4175-5322

Department of Radiological Sciences, University of California, Los Angeles, California, USA

Physics and Biology in Medicine IDP, David Geffen School of Medicine, University of California, Los Angeles, California, USA

Department of Bioengineering, Henry Samueli School of Engineering, University of California, Los Angeles, California, USA

Search for more papers by this author

Yongkai Liu PhD,

Corresponding Author

Yongkai Liu PhD

[email protected]

orcid.org/0000-0003-4478-4006

Department of Radiological Sciences, University of California, Los Angeles, California, USA

Physics and Biology in Medicine IDP, David Geffen School of Medicine, University of California, Los Angeles, California, USA

Address reprint requests to: Y.L., Department of Radiological Sciences, 300 UCLA Medical Plaza, Suite B114, Los Angeles, California 90095, USA. E-mail: [email protected]

Search for more papers by this author

Fatemeh Zabihollahy PhD,

Fatemeh Zabihollahy PhD

Department of Radiological Sciences, University of California, Los Angeles, California, USA

Search for more papers by this author

Ran Yan MS,

Ran Yan MS

orcid.org/0000-0001-8204-8654

Department of Radiological Sciences, University of California, Los Angeles, California, USA

Department of Bioengineering, Henry Samueli School of Engineering, University of California, Los Angeles, California, USA

Search for more papers by this author

Brian Lee MD,

Brian Lee MD

Department of Pediatrics, David Geffen School of Medicine, University of California, Los Angeles, California, USA

Search for more papers by this author

Carla Janzen MD, PhD,

Carla Janzen MD, PhD

Department of Obstetrics and Gynecology, David Geffen School of Medicine, University of California, Los Angeles, California, USA

Search for more papers by this author

Sherin U. Devaskar MD,

Sherin U. Devaskar MD

Department of Pediatrics, David Geffen School of Medicine, University of California, Los Angeles, California, USA

Search for more papers by this author

Kyunghyun Sung PhD,

Kyunghyun Sung PhD

orcid.org/0000-0003-4175-5322

Department of Radiological Sciences, University of California, Los Angeles, California, USA

Physics and Biology in Medicine IDP, David Geffen School of Medicine, University of California, Los Angeles, California, USA

Department of Bioengineering, Henry Samueli School of Engineering, University of California, Los Angeles, California, USA

Search for more papers by this author

First published: 16 August 2022

https://doi.org/10.1002/jmri.28403

Citations: 1

Grant Sponsors: NIH U01-HD087221

Share a link

Email
Wechat
Bluesky

Abstract

Background

Automated segmentation of the placenta by MRI in early pregnancy may help predict normal and aberrant placenta function, which could improve the efficiency of placental assessment and the prediction of pregnancy outcomes. An automated segmentation method that works at one gestational age may not transfer effectively to other gestational ages.

Purpose

To evaluate a spatial attentive deep learning method (SADL) for automated placental segmentation on longitudinal placental MRI scans.

Study type

Prospective, single-center.

Subjects

A total of 154 pregnant women who underwent MRI scans at both 14–18 weeks of gestation and at 19–24 weeks of gestation, divided into training (N = 108), validation (N = 15), and independent testing datasets (N = 31).

Field Strength/Sequence

A 3 T, T2-weighted half Fourier single-shot turbo spin-echo (T2-HASTE) sequence.

Assessment

The reference standard of placental segmentation was manual delineation on T2-HASTE by a third-year neonatology clinical fellow (B.L.) under the supervision of an experienced maternal-fetal medicine specialist (C.J. with 20 years of experience) and an MRI scientist (K.S. with 19 years of experience).

Statistical Tests

The three-dimensional Dice similarity coefficient (DSC) was used to measure the automated segmentation performance compared to the manual placental segmentation. A paired t-test was used to compare the DSCs between SADL and U-Net methods. A Bland–Altman plot was used to analyze the agreement between manual and automated placental volume measurements. A P value < 0.05 was considered statistically significant.

Results

In the testing dataset, SADL achieved average DSCs of 0.83 ± 0.06 and 0.84 ± 0.05 in the first and second MRI, which were significantly higher than those achieved by U-Net (0.77 ± 0.08 and 0.76 ± 0.10, respectively). A total of 6 out of 62 MRI scans (9.6%) had volume measurement differences between the SADL-based automated and manual volume measurements that were out of 95% limits of agreement.

Data Conclusions

SADL can automatically detect and segment the placenta with high performance in MRI at two different gestational ages.

Level of Evidence

Technical Efficacy Stage

Conflicts of Interest

The authors disclose no potential conflicts of interest.

References

1Shahedi M, Dormer JD, Anusha Devi TT, et al. Segmentation of uterus and placenta in MR images using a fully convolutional neural network. Proc SPIE Int Soc Opt Eng 2020; 11314:113141R.
PubMed Google Scholar
2Liu D, Shao X, Danyalov A, et al. Human placenta blood flow during early gestation with pseudocontinuous arterial spin labeling MRI. J Magn Reson Imaging 2020; 51: 1247-1257.
10.1002/jmri.26944
PubMed Web of Science® Google Scholar
3Sorensen AV, Hutter JM, Grant EP, Seed M, Gowland P. T2* weighted placental MRI: Basic research tool or an emerging clinical test of placental dysfunction? Ultrasound Obstet Gynecol 2019; 55: 293-302.
10.1002/uog.20855
Web of Science® Google Scholar
4Wenger E, Mårtensson J, Noack H, et al. Comparing manual and automatic segmentation of hippocampal volumes: Reliability and validity issues in younger and older brains. Hum Brain Mapp 2014; 35: 4236-4248.
10.1002/hbm.22473
PubMed Web of Science® Google Scholar
5Maldjian C, Adam R, Pelosi M, Pelosi M, Rudelli RD, Maldjian J. MRI appearance of placenta percreta and placenta accreta. Magn Reson Imaging 1999; 17: 965-971.
10.1016/S0730-725X(99)00035-1
CAS PubMed Web of Science® Google Scholar
6Dahdouh S, Andescavage N, Yewale S, et al. In vivo placental MRI shape and textural features predict fetal growth restriction and postnatal outcome. J Magn Reson Imaging 2018; 47: 449-458.
10.1002/jmri.25806
PubMed Web of Science® Google Scholar
7Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. Med Image Comput Comput Assit Interv 2015; 9351: 234-241.
Google Scholar
8Wang G, Li W, Zuluaga MA, et al. Interactive medical image segmentation using deep learning with image-specific fine tuning. IEEE Trans Med Imaging 2018; 37: 1562-1573.
10.1109/TMI.2018.2791721
PubMed Web of Science® Google Scholar
9Han M, Bao Y, Sun Z, et al. Automatic segmentation of human placenta images with U-net. IEEE Access 2019; 7: 180083-180092.
10.1109/ACCESS.2019.2958133
Web of Science® Google Scholar
10Milletari F, Navab N, Ahmadi SA. V-Net: Fully convolutional neural networks for volumetric medical image segmentation. Proc. -2016 4th Int. Conf. 3D Vision, 3DV 2016. New York; Institute of Electrical and Electronics Engineers Inc; 2016. p 565-571.
10.1109/3DV.2016.79
Google Scholar
11Havaei M, Davy A, Warde-Farley D, et al. Brain tumor segmentation with deep neural networks. Med Image Anal 2017; 35: 18-31.
10.1016/j.media.2016.05.004
PubMed Web of Science® Google Scholar
12Liu Y, Yang G, Mirak SA, et al. Automatic prostate zonal segmentation using fully convolutional network with feature pyramid attention. IEEE Access 2019; 7: 163626-163632.
10.1109/ACCESS.2019.2952534
Web of Science® Google Scholar
13Liu Y, Yang G, Hosseiny M, et al. Exploring uncertainty measures in Bayesian deep attentive neural networks for prostate zonal segmentation. IEEE Access 2020; 8: 151817-151828.
10.1109/ACCESS.2020.3017168
PubMed Web of Science® Google Scholar
14Looney P, Stevenson GN, Nicolaides KH, et al. Fully automated, real-time 3D ultrasound segmentation to estimate first trimester placental volume using deep learning. JCI Insight 2018; 3:e120178. https://doi.org/10.1172/jci.insight.120178.
10.1172/jci.insight.120178
PubMed Web of Science® Google Scholar
15Hu R, Singla R, Yan R, Mayer C, Rohling RN. Automated placenta segmentation with a convolutional neural network weighted by acoustic shadow detection. 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). Berlin, Germany: IEEE; 2019. p 6718-6723.
10.1109/EMBC.2019.8857448
Google Scholar
16Alansary A, Kamnitsas K, Davidson A, et al. Fast fully automatic segmentation of the human placenta from motion corrupted MRI. International Conference on Medical Image Computing and Computer-Assisted Intervention Lecture Notes in Computer Science (Including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics). New York: Springer Verlag; 2016. p 589-597.
10.1007/978-3-319-46723-8_68
Google Scholar
17Wang G, Zuluaga MA, Li W, et al. DeepIGeoS: A deep interactive geodesic framework for medical image segmentation. IEEE Trans Pattern Anal Mach Intell 2019; 41: 1559-1572.
10.1109/TPAMI.2018.2840695
PubMed Web of Science® Google Scholar
18Wang G, Zuluaga MA, Pratt R, et al. Slic-seg: Slice-by-slice segmentation propagation of the placenta in fetal MRI using one-plane scribbles and online learning. International Conference on Medical Image Computing and Computer-Assisted Intervention Lecture Notes in Computer Science (Including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics). New York: Springer Verlag; 2015. p 29-37.
10.1007/978-3-319-24574-4_4
Google Scholar
19Wang G, Zuluaga MA, Pratt R, et al. Slic-Seg: A minimally interactive segmentation of the placenta from sparse and motion-corrupted fetal MRI in multiple views. Med Image Anal 2016; 34: 137-147.
10.1016/j.media.2016.04.009
PubMed Web of Science® Google Scholar
20Patel MR, Klufas RA, Alberico RA, Edelman RR. Half-fourier acquisition single-shot turbo spin-echo (HASTE) MR: Comparison with fast spin-echo MR in diseases of the brain. Am J Neuroradiol 1997; 18: 1635-1640.
CAS PubMed Web of Science® Google Scholar
21Huang Z, Wang X, Huang L, Huang C, Wei Y, Liu W. Ccnet: Criss-cross attention for semantic segmentation. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). New York: IEEE; 2019. p 603-612.
10.1109/ICCV.2019.00069
Google Scholar
22He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Proc IEEE Conf Comput Vis Pattern Recognit 2016; 770-778.
Google Scholar
23Li H, Xiong P, An J, Wang L. Pyramid attention network for semantic segmentation. 2018.
Google Scholar
24Ghafoorian M, Karssemeijer N, Heskes T, et al. Location sensitive deep convolutional neural networks for segmentation of white matter hyperintensities. Sci Rep 2017; 7: 1-12.
10.1038/s41598-017-05300-5
CAS PubMed Web of Science® Google Scholar
25Van Griethuysen JJM, Fedorov A, Parmar C, et al. Computational radiomics system to decode the radiographic phenotype. Cancer Res 2017; 77: e104-e107.
10.1158/0008-5472.CAN-17-0339
CAS PubMed Web of Science® Google Scholar
26Dice LR. Measures of the amount of ecologic association between species. Ecology 1945; 26: 297-302.
10.2307/1932409
Web of Science® Google Scholar
27Giavarina D. Understanding Bland Altman analysis. Biochem Med 2015; 25: 141-151.
10.11613/BM.2015.015
PubMed Web of Science® Google Scholar
28Leatherbarrow RJ. Using linear and non-linear regression to fit biochemical data. Trends Biochem Sci 1990; 15: 455-458.
10.1016/0968-0004(90)90295-M
PubMed Web of Science® Google Scholar
29Metzenbauer M, Hafner E, Hoefinger D, et al. Three-dimensional ultrasound measurement of the placental volume in early pregnancy: Method and correlation with biochemical placenta parameters. Placenta 2001; 22: 602-605.
10.1053/plac.2001.0684
CAS PubMed Web of Science® Google Scholar
30Adams T, Yeh C, Bennett-Kunzier N, Kinzler WL. Long-term maternal morbidity and mortality associated with ischemic placental disease. Semin Perinatol 2014; 38: 146-150.
10.1053/j.semperi.2014.03.003
PubMed Web of Science® Google Scholar

Citing Literature

Volume57, Issue5

May 2023

Pages 1533-1540

This article also appears in:

Placental and Fetal MRI

Evaluation of Spatial Attentive Deep Learning for Automatic Placental Segmentation on Longitudinal MRI

Abstract

Background

Purpose

Study type

Subjects

Field Strength/Sequence

Assessment

Statistical Tests

Results

Data Conclusions

Level of Evidence

Technical Efficacy Stage

Conflicts of Interest

References

Citing Literature

References

Information

About Wiley Online Library

Help & Support

Opportunities

Connect with Wiley

Evaluation of Spatial Attentive Deep Learning for Automatic Placental Segmentation on Longitudinal MRI

Abstract

Background

Purpose

Study type

Subjects

Field Strength/Sequence

Assessment

Statistical Tests

Results

Data Conclusions

Level of Evidence

Technical Efficacy Stage

Conflicts of Interest

References

Citing Literature

References

Related

Information