Advertisement

European Radiology

, Volume 29, Issue 11, pp 6152–6162 | Cite as

Identification of suspicious invasive placentation based on clinical MRI data using textural features and automated machine learning

  • Huaiqiang Sun
  • Haibo Qu
  • Lu Chen
  • Wei Wang
  • Yi Liao
  • Ling Zou
  • Ziyi Zhou
  • Xiaodong Wang
  • Shu ZhouEmail author
Imaging Informatics and Artificial Intelligence

Abstract

Objective

The aim of this study was to investigate whether intraplacental texture features from routine placental MRI can objectively and accurately predict invasive placentation.

Material and methods

This retrospective study includes 99 pregnant women with pathologically confirmed placental invasion and 56 pregnant women with simple placenta previa. All participants underwent magnetic resonance imaging after 24 gestational weeks. The placenta was segmented in sagittal images from both turbo spin echo (TSE) and balanced turbo field echo (bTFE) sequences. Textural features were extracted from the both original and Laplacian of Gaussian (LoG)-filtered MRI images. An automated machine learning algorithm was applied to the extracted feature sets to obtain the optimal preprocessing steps, classification algorithm, and corresponding hyper-parameters.

Results

A gradient boosting classifier using all textual features from original and LoG-filtered TSE images and bTFE images identified by the automated machine learning algorithm achieved the optimal performance with sensitivity, specificity, accuracy, and area under ROC curve (AUC) of 100%, 88.5%, 95.2%, and 0.98 in the prediction of placental invasion. In addition, textural features that contributed to the prediction of placental invasion differ from the features significantly affected by normal placenta maturation.

Conclusions

Quantifying intraplacental heterogeneity using LoG filtration and texture analysis highlights the different heterogeneous appearance caused by abnormal placentation relative to normal maturation. The predictive model derived from automated machine learning yielded good performance, indicating the proposed radiomic analysis pipeline can accurately predict placental invasion and facilitate clinical decision-making for pregnant women with suspicious placental invasion.

Key Points

The intraplacental texture features have high efficiency in prediction of invasive placentation after 24 gestational weeks.

The features with dominated predictive power did not overlap with the features significantly affected by gestational age.

Keywords

Magnetic resonance imaging Placenta accreta Radiomics Computer-assisted image analysis Machine learning 

Abbreviations

AUC

Area under the ROC curve

bTFE

Balanced turbo field echo

LoG

Laplacian of Gaussian

PI

Placental invasion

SPP

Simple placenta previa

TSE

Turbo spin echo

Notes

Funding

This study was supported by the National Natural Science Foundation of China (81601458), the Ministry of Science and Technology of China (2016YFC0100803, 2018YFC1004603), the Health and Family Planning Commission of Sichuan Province (16ZD017), and Bureau of Science and Technology of Chengdu City (2014-HM01-00314-SF).

Compliance with ethical standards

Guarantor

The scientific guarantor of this publication is S. Zhou.

Conflict of interest

The authors declare that they have no conflict of interest.

Statistics and biometry

One of the authors (H. Sun) has significant statistical expertise.

Informed consent

Written informed consent was waived by the Institutional Review Board.

Ethical approval

Institutional Review Board approval was obtained.

Methodology

• retrospective

• cross-sectional study

• performed at one institution

Supplementary material

330_2019_6372_MOESM1_ESM.docx (30.6 mb)
ESM 1 (DOCX 31312 kb)

References

  1. 1.
    Oyelese Y, Smulian JC (2006) Placenta previa, placenta accreta, and vasa previa. Obstet Gynecol 107:927–941.  https://doi.org/10.1097/01.AOG.0000207559.15715.98 CrossRefPubMedPubMedCentralGoogle Scholar
  2. 2.
    Garmi G, Salim R (2012) Epidemiology, etiology, diagnosis, and management of placenta accreta. Obstet Gynecol Int 2012:1–7.  https://doi.org/10.1155/2012/873929 CrossRefGoogle Scholar
  3. 3.
    Baughman WC, Corteville JE, Shah RR (2008) Placenta accreta: spectrum of US and MR imaging findings. Radiographics 28:1905–1916.  https://doi.org/10.1148/rg.287085060 CrossRefPubMedGoogle Scholar
  4. 4.
    Kocher MR, Sheafor DH, Bruner E et al (2017) Diagnosis of abnormally invasive posterior placentation: the role of MR imaging. Radiol Case Rep 12:295–299.  https://doi.org/10.1016/j.radcr.2017.01.014 CrossRefPubMedPubMedCentralGoogle Scholar
  5. 5.
    Delli Pizzi A, Tavoletta A, Narciso R et al (2019) Prenatal planning of placenta previa: diagnostic accuracy of a novel MRI-based prediction model for placenta accreta spectrum (PAS) and clinical outcome. Abdom Radiol 44:1873–1882.  https://doi.org/10.1007/s00261-018-1882-8 CrossRefGoogle Scholar
  6. 6.
    D’Antonio F, Iacovella C, Palacios-Jaraquemada J et al (2014) Prenatal identification of invasive placentation using magnetic resonance imaging: systematic review and meta-analysis. Ultrasound Obstet Gynecol 44:8–16.  https://doi.org/10.1002/uog.13327 CrossRefPubMedGoogle Scholar
  7. 7.
    Alamo L, Anaye A, Rey J et al (2013) Detection of suspected placental invasion by MRI: do the results depend on observer’ experience? Eur J Radiol 82:e51–e57.  https://doi.org/10.1016/j.ejrad.2012.08.022 CrossRefPubMedGoogle Scholar
  8. 8.
    Ueno Y, Kitajima K, Kawakami F et al (2014) Novel MRI finding for diagnosis of invasive placenta praevia: evaluation of findings for 65 patients using clinical and histopathological correlations. Eur Radiol 24:881–888.  https://doi.org/10.1007/s00330-013-3076-7 CrossRefPubMedGoogle Scholar
  9. 9.
    Lax A, Prince MR, Mennitt KW et al (2007) The value of specific MRI features in the evaluation of suspected placental invasion. Magn Reson Imaging 25:87–93.  https://doi.org/10.1016/j.mri.2006.10.007 CrossRefPubMedGoogle Scholar
  10. 10.
    Lambin P, Rios-Velazquez E, Leijenaar R et al (2012) Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer 48:441–446.  https://doi.org/10.1016/j.ejca.2011.11.036 CrossRefPubMedPubMedCentralGoogle Scholar
  11. 11.
    Gillies RJ, Kinahan PE, Hricak H (2016) Radiomics: images are more than pictures, they are data. Radiology 278:563–577.  https://doi.org/10.1148/radiol.2015151169 CrossRefPubMedGoogle Scholar
  12. 12.
    Langs G, Röhrich S, Hofmanninger J et al (2018) Machine learning: from radiomics to discovery and routine. Radiologe 58:1–6.  https://doi.org/10.1007/s00117-018-0407-3 CrossRefPubMedPubMedCentralGoogle Scholar
  13. 13.
    Castellano G, Bonilha L, Li LM, Cendes F (2004) Texture analysis of medical images. Clin Radiol 59:1061–1069.  https://doi.org/10.1016/j.crad.2004.07.008 CrossRefPubMedGoogle Scholar
  14. 14.
    Horowitz JM, Berggruen S, McCarthy RJ et al (2015) When timing is everything: are placental MRI examinations performed before 24 weeks’ gestational age reliable? Am J Roentgenol 205:685–692.  https://doi.org/10.2214/AJR.14.14134 CrossRefGoogle Scholar
  15. 15.
    Wang G, Zuluaga MA, Pratt R et al (2016) Slic-Seg: a minimally interactive segmentation of the placenta from sparse and motion-corrupted fetal MRI in multiple views. Med Image Anal 34:137–147.  https://doi.org/10.1016/j.media.2016.04.009 CrossRefPubMedPubMedCentralGoogle Scholar
  16. 16.
    Yushkevich PA, Piven J, Hazlett HC et al (2006) User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage 31:1116–1128.  https://doi.org/10.1016/j.neuroimage.2006.01.015 CrossRefGoogle Scholar
  17. 17.
    Van Griethuysen JJM, Fedorov A, Parmar C et al (2017) Computational radiomics system to decode the radiographic phenotype. Cancer Res 77:e104–e107.  https://doi.org/10.1158/0008-5472.CAN-17-0339 CrossRefPubMedPubMedCentralGoogle Scholar
  18. 18.
    Yip SSF, Aerts HJWL (2016) Applications and limitations of radiomics. Phys Med Biol 61:R150–R166.  https://doi.org/10.1088/0031-9155/61/13/R150 CrossRefPubMedPubMedCentralGoogle Scholar
  19. 19.
    Fadl S, Moshiri M, Fligner CL et al (2017) Placental imaging: normal appearance with review of pathologic findings. Radiographics 37:979–998.  https://doi.org/10.1148/rg.2017160155 CrossRefPubMedGoogle Scholar
  20. 20.
    Dukart J, Schroeter ML, Mueller K (2011) Age correction in dementia – matching to a healthy brain. PLoS One 6:e22193.  https://doi.org/10.1371/journal.pone.0022193 CrossRefPubMedPubMedCentralGoogle Scholar
  21. 21.
    Olson RS, Bartley N, Urbanowicz RJ, Moore JH (2016) Evaluation of a Tree-based Pipeline Optimization Tool for automating data science. Proc 2016 Genet Evol Comput Conf - GECCO ‘16 485–492.  https://doi.org/10.1145/2908812.2908918
  22. 22.
    Cohen J (1960) A coefficient of agreement for nominal scales. Educ Psychol Meas 20:37–46.  https://doi.org/10.1177/001316446002000104 CrossRefGoogle Scholar
  23. 23.
    Natekin A, Knoll A (2013) Gradient boosting machines, a tutorial. Front Neurorobot 7.  https://doi.org/10.3389/fnbot.2013.00021
  24. 24.
    Wang S, Summers RM (2012) Machine learning and radiology. Med Image Anal 16:933–951.  https://doi.org/10.1016/j.media.2012.02.005 CrossRefPubMedPubMedCentralGoogle Scholar
  25. 25.
    Parmar C, Grossmann P, Bussink J et al (2015) Machine learning methods for quantitative radiomic biomarkers. Sci Rep 5:13087.  https://doi.org/10.1038/srep13087 CrossRefPubMedPubMedCentralGoogle Scholar
  26. 26.
    Larue RTHMHM, Defraene G, De Ruysscher D et al (2017) Quantitative radiomics studies for tissue characterization: a review of technology and methodological procedures. Br J Radiol 90:20160665.  https://doi.org/10.1259/bjr.20160665 CrossRefPubMedPubMedCentralGoogle Scholar
  27. 27.
    Scheffler K, Lehnhardt S (2003) Principles and applications of balanced SSFP techniques. Eur Radiol 13:2409–2418.  https://doi.org/10.1007/s00330-003-1957-x CrossRefPubMedGoogle Scholar
  28. 28.
    Neycenssac F (1993) Contrast enhancement using the Laplacian-of-a-Gaussian filter. CVGIP Graph Model Image Process 55:447–463.  https://doi.org/10.1006/cgip.1993.1034 CrossRefGoogle Scholar
  29. 29.
    Rahaim NSA, Whitby EH (2015) The MRI features of placental adhesion disorder and their diagnostic significance: systematic review. Clin Radiol 70:917–925.  https://doi.org/10.1016/j.crad.2015.04.010 CrossRefPubMedGoogle Scholar
  30. 30.
    Tang X (1998) Texture information in run-length matrices. IEEE Trans Image Process 7:1602–1609.  https://doi.org/10.1109/83.725367 CrossRefPubMedGoogle Scholar
  31. 31.
    Blaicher W, Brugger PC, Mittermayer C et al (2006) Magnetic resonance imaging of the normal placenta. Eur J Radiol 57:256–260.  https://doi.org/10.1016/j.ejrad.2005.11.025 CrossRefPubMedGoogle Scholar
  32. 32.
    Haralick RM, Shanmugam K, Dinstein I (1973) Textural features for image classification. IEEE Trans Syst Man Cybern SMC-3:610–621.  https://doi.org/10.1109/TSMC.1973.4309314 CrossRefGoogle Scholar

Copyright information

© European Society of Radiology 2019

Authors and Affiliations

  1. 1.Huaxi MR Research Center, Department of RadiologyWest China Hospital of Sichuan UniversityChengduChina
  2. 2.Department of RadiologyWest China Second Hospital of Sichuan UniversityChengduChina
  3. 3.Key Laboratory of Birth Defects and Related Diseases of Women and Children (Sichuan University)Ministry of EducationChengduChina
  4. 4.Department of Periodical PressWest China Hospital of Sichuan UniversityChengduChina
  5. 5.Department of PathologyWest China Second Hospital of Sichuan UniversityChengduChina
  6. 6.Department of Obstetrics and GynecologyWest China Second Hospital of Sichuan UniversityChengduChina

Personalised recommendations