Skip to main content

Deep learning for the determination of myometrial invasion depth and automatic lesion identification in endometrial cancer MR imaging: a preliminary study in a single institution

Abstract

Objective

To determine the diagnostic performance of a deep learning (DL) model in evaluating myometrial invasion (MI) depth on T2-weighted imaging (T2WI)–based endometrial cancer (EC) MR imaging (ECM).

Methods

We retrospectively enrolled 530 patients with pathologically proven EC at our institution between January 1, 2013, and December 31, 2017. All imaging data were reviewed on picture archiving and communication systems (PACS) server. Both sagittal and coronal T2WI-based MR images were used for lesion area determination. All MR images were divided into two groups: deep (more than 50%) and shallow (less than 50%) MI based on their pathological diagnosis. We trained a detection model based on YOLOv3 algorithm to locate the lesion area on ECM. Then, the detected regions were fed into a classification model based on DL network to identify MI depth automatically.

Results

In the testing dataset, the trained model detected lesion regions with an average precision rate of 77.14% and 86.67% in both sagittal and coronal images, respectively. The classification model yielded an accuracy of 84.78%, a sensitivity of 66.67%, a specificity of 87.50%, a positive predictive value of 44.44%, and a negative predictive value of 94.59% in determining deep MI. The radiologists and trained network model together yielded an accuracy of 86.2%, a sensitivity of 77.8%, a specificity of 87.5%, a positive predictive value of 48.3%, and a negative predictive value of 96.3%.

Conclusion

In this study, the DL network model derived from MR imaging provided a competitive, time-efficient diagnostic performance in MI depth identification.

Key Points

• The models established with the deep learning method could help improve the diagnostic confidence and performance of MI identification based on endometrial cancer MR imaging.

• The models enabled the classification of endometrial cancer MR images to the two categories with a sensitivity of 0.67, a specificity of 0.88, and an accuracy of 0.85.

• Using the detected lesion region to evaluate myometrial invasion depth could remove redundant information in the image and provide more effective features.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Abbreviations

ACC:

Accuracy

AUC:

Area under the ROC curve

CI:

Confidence interval

CNN:

Convolutional neural network

DL:

Deep learning

DWI:

Diffusion-weighted imaging

EC:

Endometrial cancer

ECM:

Endometrial cancer MR imaging

EP2D:

Echo-planar imaging two-dimensional

FS TW2I:

Fat-saturated T2-weighted imaging

LNM:

Lymph node metastases

MD:

Maximum diameter

MI:

Myometrial invasion

ML:

Machine learning

MRI:

Magnetic resonance imaging

NPV:

Negative predictive value

PACS:

Picture archiving and communication systems

PPV:

Positive predictive value

PR:

Precision-recall

ROC:

Receiver operating characteristic curve

SEN:

Sensitivity

SPE:

Specificity

T1WI:

T1-weighted imaging

TW2:

T2-weighted imaging

TSE:

Turbo spin-echo

References

  1. 1.

    Morice P, Leary A, Creutzberg C, Abu-Rustum N, Darai E (2016) Endometrial cancer. Lancet 387:1094–1108. https://doi.org/10.1016/S0140-6736(15)00130-0

  2. 2.

    Siegel RL, Miller KD, Jemal A (2019) Cancer statistics, 2019. CA Cancer J Clin 69:7–34. https://doi.org/10.3322/caac.21551

    Article  PubMed  PubMed Central  Google Scholar 

  3. 3.

    Nougaret S, Horta M, Sala E et al (2019) Endometrial cancer MRI staging: updated guidelines of the European Society of Urogenital Radiology. Eur Radiol 29:792–805. https://doi.org/10.1007/s00330-018-5515-y

    Article  PubMed  Google Scholar 

  4. 4.

    Lavaud P, Fedida B, Canlorbe G, Bendifallah S, Darai E, Thomassin-Naggara I (2018) Preoperative MR imaging for ESMO-ESGO-ESTRO classification of endometrial cancer. Diagn Interv Imaging 99:387–396. https://doi.org/10.1016/j.diii.2018.01.010

  5. 5.

    Amant F, Mirza MR, Koskas M, Creutzberg CL (2018) Cancer of the corpus uteri. Int J Gynecol Obstet 143(S2):37–50

    Article  Google Scholar 

  6. 6.

    Colombo N, Preti E, Landoni F et al (2013) Endometrial cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol 24:vi33–vvi8. https://doi.org/10.1093/annonc/mdt353

    Article  PubMed  Google Scholar 

  7. 7.

    Ahmed M, Al-Khafaji JF, Class CA et al (2018) Can MRI help assess aggressiveness of endometrial cancer? Clin Radiol 73:833.e11–833.e18. https://doi.org/10.1016/j.crad.2018.05.002

    CAS  Article  Google Scholar 

  8. 8.

    Sala E, Rockall AG, Freeman SJ et al (2013) The added role of MR imaging in treatment stratification of patients with gynecologic malignancies: what the radiologist needs to know. Radiology 266:717–740

    Article  Google Scholar 

  9. 9.

    Zhang H, Mao Y, Chen X et al (2019) Magnetic resonance imaging radiomics in categorizing ovarian masses and predicting clinical outcome: a preliminary study. Eur Radiol 29:3358–3371. https://doi.org/10.1007/s00330-019-06124-9

    Article  PubMed  Google Scholar 

  10. 10.

    Sahin H, Sarioglu FC, Bagci M, Karadeniz T, Uluer H, Sanci M. (2018) Preoperative magnetic resonance volumetry in predicting myometrial invasion, lymphovascular space invasion, and tumor grade: is it valuable in International Federation of Gynecology and Obstetrics Stage I Endometrial Cancer? Int J Gynecol Cancer 28:666–674. https://doi.org/10.1097/IGC.0000000000001208

  11. 11.

    Qu JR, Qin L, Li X et al (2018) Predicting parametrial invasion in cervical carcinoma (stages IB1, IB2, and IIA): diagnostic accuracy of T2-weighted imaging combined with DWI at 3 T. Am J Roentgenol 210:677–684. https://doi.org/10.2214/AJR.17.18104

    Article  Google Scholar 

  12. 12.

    Nakamura K, Nakayama K, Ishikawa N et al (2018) Preoperative tumor size is associated with deep myometrial invasion and lymph node metastases and is a negative prognostic indicator for patients with endometrial carcinoma. Oncotarget 9:23164–23172. https://doi.org/10.18632/oncotarget.25248

    Article  PubMed  PubMed Central  Google Scholar 

  13. 13.

    Nougaret S, Reinhold C, Alsharif SS et al (2015) Endometrial cancer: combined MR volumetry and diffusion-weighted imaging for assessment of myometrial and lymphovascular invasion and tumor grade. Radiology 276:797–808. https://doi.org/10.1148/radiol.15141212

    Article  PubMed  PubMed Central  Google Scholar 

  14. 14.

    Ueno Y, Forghani B, Forghani R et al (2017) Endometrial carcinoma: MR imaging-based texture model for preoperative risk stratification - a preliminary analysis1. Radiology 284:748–757. https://doi.org/10.1148/radiol.2017161950

    Article  PubMed  Google Scholar 

  15. 15.

    Tanaka T, Terai Y, Ono YJ et al (2015) Preoperative MRI and intraoperative frozen section diagnosis of myometrial invasion in patients with endometrial cancer. Int J Gynecol Cancer 25:879–883. https://doi.org/10.1097/IGC.0000000000000470

    Article  PubMed  Google Scholar 

  16. 16.

    Litjens G, Kooi T, Bejnordi BE et al (2017) A survey on deep learning in medical image analysis. Med Image Anal 42:60–88. https://doi.org/10.1016/j.media.2017.07.005

    Article  PubMed  PubMed Central  Google Scholar 

  17. 17.

    Shen D, Wu G, Suk H-I (2017) Deep learning in medical image analysis. Annu Rev Biomed Eng 19:221–248. https://doi.org/10.1146/annurev-bioeng-071516-044442

  18. 18.

    Shen L, Margolies LR, Rothstein JH, Fluder E, McBride R, Sieh W (2019) Deep learning to improve breast cancer detection on screening mammography. Sci Rep 9. https://doi.org/10.1038/s41598-019-48995-4

  19. 19.

    Du Y, Zhang R, Zargari A et al (2018) Classification of tumor epithelium and stroma by exploiting image features learned by deep convolutional neural networks. Ann Biomed Eng 46:1988–1999. https://doi.org/10.1007/s10439-018-2095-6

    Article  PubMed  PubMed Central  Google Scholar 

  20. 20.

    Wang S, Liu Z, Rong Y et al (2016) Deep learning provides a new computed tomography-based prognostic biomarker for recurrence prediction in high-grade serous ovarian cancer. Radiother Oncol 171–177. https://doi.org/10.1016/j.radonc.2018.10.019

  21. 21.

    Song Y, Zhang YD, Yan X et al (2018) Computer-aided diagnosis of prostate cancer using a deep convolutional neural network from multiparametric MRI. J Magn Reson Imaging 48:1570–1577. https://doi.org/10.1002/jmri.26047

    Article  PubMed  Google Scholar 

  22. 22.

    Razzak MI, Naz S, Zaib A (2017) Deep learning for medical image processing: overview, challenges and future. in: Dey N, Ashour A, Borra S (eds) Classification in BioApps. Lecture Notes in Computational Vision and Biomechanics, vol 26. Springer, Cham

  23. 23.

    Zhou ZH, Liu XY (2006) Training cost-sensitive neural networks with methods addressing the class imbalance problem. IEEE Trans Knowl Data Eng 18:63–77. https://doi.org/10.1109/TKDE.2006.17

    Article  Google Scholar 

  24. 24.

    Redmon J, Farhadi A, Ap C (2018) YOLOv3. Nutr Rev 36:346–348. https://doi.org/10.1111/j.1753-4887.1978.tb03704.x

    Article  Google Scholar 

  25. 25.

    Russakovsky O, Deng J, Su H et al (2015) ImageNet large scale visual recognition challenge. Int J Comput Vis 115:211–252. https://doi.org/10.1007/s11263-015-0816-y

    Article  Google Scholar 

  26. 26.

    He K, Zhang X, Ren S, Sun J (2015) Deep residual learning for image recognition. arXiv preprint arXiv:151203385v1 7:171–180. https://doi.org/10.3389/fpsyg.2013.00124

  27. 27.

    Davis J, Goadrich M (2006) The relationship between precision-recall and ROC curves. Proceedings of the 23rd International Conference on Machine Learning 148:233–240. https://doi.org/10.1145/1143844.1143874

    Article  Google Scholar 

  28. 28.

    Günakan E, Atan S, Haberal AN, Küçükyıldız İA, Gökçe E, Ayhan A (2019) A novel prediction method for lymph node involvement in endometrial cancer: machi learning. Int J Gynecol Cancer 29:320–324. https://doi.org/10.1136/ijgc-2018-000033

  29. 29.

    Downing MJ, Papke DJ, Tyekucheva S, Mutter GL (2019) A new classification of benign, premalignant, and malignant endometrial tissues using machine learning applied to 1413 candidate variables. Int J Gynecol Pathol. https://doi.org/10.1097/PGP.0000000000000615

  30. 30.

    Yasaka K, Akai H, Abe O, Kiryu S (2018) Deep learning with convolutional neural network for differentiation of liver masses at dynamic contrast-enhanced CT: a preliminary study. Radiology 286:887–896. https://doi.org/10.1148/radiol.2017170706

    Article  PubMed  Google Scholar 

  31. 31.

    Ribli D, Horváth A, Unger Z, Pollner P, Csabai I (2018) Detecting and classifying lesions in mammograms with deep learning. Sci Rep 8:4165. https://doi.org/10.1038/s41598-018-22437-z

  32. 32.

    Pandiselvi T, Maheswaran R (2019) Efficient framework for identifying, locating, detecting and classifying MRI brain tumor in MRI images. J Med Syst 43:189. https://doi.org/10.1007/s10916-019-1253-1

    CAS  Article  PubMed  Google Scholar 

Download references

Funding

This study has received funding by National Natural Science Foundation of China (No.61731009, 81,771,816).

Author information

Affiliations

Authors

Corresponding authors

Correspondence to Guang Yang or He Zhang.

Ethics declarations

Guarantor

The scientific guarantor of this publication is Guofu Zhang.

Conflict of interest

The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.

Statistics and biometry

No complex statistical methods were necessary for this paper.

Informed consent

Written informed consent was waived by the Institutional Review Board.

Ethical approval

Institutional Review Board approval was obtained.

Methodology

• Retrospective

• Diagnostic or prognostic study

• Performed at one institution

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary materials

Supplementary Table 1

Details of parameters for 1.5 Tesla MRI imaging protocols (DOCX 18 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Chen, X., Wang, Y., Shen, M. et al. Deep learning for the determination of myometrial invasion depth and automatic lesion identification in endometrial cancer MR imaging: a preliminary study in a single institution. Eur Radiol 30, 4985–4994 (2020). https://doi.org/10.1007/s00330-020-06870-1

Download citation

Keywords

  • Endometrial cancer
  • Magnetic resonance imaging
  • Deep learning
  • Neoplasm staging