Application of deep learning to the classification of images from colposcopy

Sato,Masakazu; Horie,Koji; Hara,Aki; Miyamoto,Yuichiro; Kurihara,Kazuko; Tomio,Kensuke; Yokota,Harushige

doi:10.3892/ol.2018.7762

Application of deep learning to the classification of images from colposcopy

Authors:
- Masakazu Sato
- Koji Horie
- Aki Hara
- Yuichiro Miyamoto
- Kazuko Kurihara
- Kensuke Tomio
- Harushige Yokota
View Affiliations

Affiliations: Department of Gynecology, Saitama Cancer Centre, Ina, Saitama 362‑0806, Japan
Published online on: January 10, 2018 https://doi.org/10.3892/ol.2018.7762
Pages: 3518-3523
Copyright: © Sato et al. This is an open access article distributed under the terms of Creative Commons Attribution License.

Metrics: Total Views: 0 (Spandidos Publications: | PMC Statistics: )

Metrics: Total PDF Downloads: 0 (Spandidos Publications: | PMC Statistics: )

Cited By (CrossRef): 0 citations Loading Articles...

Abstract

The objective of the present study was to investigate whether deep learning could be applied successfully to the classification of images from colposcopy. For this purpose, a total of 158 patients who underwent conization were enrolled, and medical records and data from the gynecological oncology database were retrospectively reviewed. Deep learning was performed with the Keras neural network and TensorFlow libraries. Using preoperative images from colposcopy as the input data and deep learning technology, the patients were classified into three groups [severe dysplasia, carcinoma in situ (CIS) and invasive cancer (IC)]. A total of 485 images were obtained for the analysis, of which 142 images were of severe dysplasia (2.9 images/patient), 257 were of CIS (3.3 images/patient), and 86 were of IC (4.1 images/patient). Of these, 233 images were captured with a green filter, and the remaining 252 were captured without a green filter. Following the application of L2 regularization, L1 regularization, dropout and data augmentation, the accuracy of the validation dataset was ~50%. Although the present study is preliminary, the results indicated that deep learning may be applied to classify colposcopy images.

View Figures

View References

1	Schmidhuber J: Deep learning in neural networks: An overview. Neural Netw. 61:85–117. 2015. View Article : Google Scholar : PubMed/NCBI
2	Esteva A, Kuprel B, Novoa RA, Ko J, Swetter SM, Blau HM and Thrun S: Dermatologist-level classification of skin cancer with deep neural networks. Nature. 542:115–118. 2017. View Article : Google Scholar : PubMed/NCBI
3	Janowczyk A and Madabhushi A: Deep learning for digital pathology image analysis: A comprehensive tutorial with selected use cases. J Pathol Inform. 7:292016. View Article : Google Scholar : PubMed/NCBI
4	Rajkomar A, Lingam S, Taylor AG, Blum M and Mongan J: High-throughput classification of radiographs using deep convolutional neural networks. J Digit Imaging. 30:95–101. 2017. View Article : Google Scholar : PubMed/NCBI
5	Yu L, Chen H, Dou Q, Qin J and Heng PA: Automated melanoma recognition in dermoscopy images via very deep residual networks. Ieee Trans Med Imaging. 36:994–1004. 2016. View Article : Google Scholar : PubMed/NCBI
6	Zhang YC and Kagen AC: Machine learning interface for medical image analysis. J Digit Imaging. 30:615–621. 2017. View Article : Google Scholar : PubMed/NCBI
7	Rampasek L and Goldenberg A: Tensorflow: Biology's gateway to deep learning? Cell Syst. 2:12–14. 2016. View Article : Google Scholar : PubMed/NCBI
8	Ginsburg O, Bray F, Coleman MP, Vanderpuye V, Eniu A, Kotha SR, Sarker M, Huong TT, Allemani C, Dvaladze A, et al: The global burden of women's cancers: A grand challenge in global health. Lancet. 389:847–860. 2017. View Article : Google Scholar : PubMed/NCBI
9	Cook DA, Smith LW, Law J, Mei W, van Niekerk DJ, Ceballos K, Gondara L, Franco EL, Coldman AJ, Ogilvie GS, et al: Aptima HPV assay versus hybrid capture^® 2 HPV test for primary cervical cancer screening in the HPV FOCAL trial. J Clin Virol. 87:23–29. 2017. View Article : Google Scholar : PubMed/NCBI
10	Coste J, Cochand-Priollet B, de Cremoux P, Le Galès C, Cartier I, Molinié V, Labbé S, Vacher-Lavenu MC and Vielh P; French Society of Clinical Cytology Study Group, : Cross sectional study of conventional cervical smear, monolayer cytology and human papillomavirus DNA testing for cervical cancer screening. BMJ. 326:7332003. View Article : Google Scholar : PubMed/NCBI
11	den Boon JA, Pyeon D, Wang SS, Horswill M, Schiffman M, Sherman M, Zuna RE, Wang Z, Hewitt SM and Pearson R: Molecular transitions from papillomavirus infection to cervical precancer and cancer: Role of stromal oestrogen receptor signalling. Proc Natl Acad Sci USA. 112:pp. E3255–E3264. 2015; View Article : Google Scholar : PubMed/NCBI
12	Garcia-Arteaga JD, Kybic J and Li W: Automatic colposcopy video tissue classification using higher order entropy-based image registration. Comput Biology Med. 41:960–970. 2011. View Article : Google Scholar
13	Balas C: A novel optical imaging method for the early detection, quantitative grading and mapping of cancerous and precancerous lesions of cervix. IEEE Trans Biomed Eng. 48:96–104. 2001. View Article : Google Scholar : PubMed/NCBI
14	Acosta-Mesa HG, Cruz-Ramirez N and Hernández-Jiménez R: Aceto-white temporal pattern classification using k-NN to identify precancerous cervical lesion in colposcopic images. Comput Biol Med. 39:778–784. 2009. View Article : Google Scholar : PubMed/NCBI
15	Acosta-Mesa HG, Rechy-Ramirez F, Mezura-Montes E, Cruz-Ramirez N and Hernández Jiménez R: Application of time series discretization using evolutionary programming for classification of precancerous cervical lesions. J Biomed Inform. 49:73–83. 2014. View Article : Google Scholar : PubMed/NCBI
16	Simoes PW, Izumi NB, Casagrande RS, Venson R, Veronezi CD, Moretti GP, da Rocha EL, Cechinel C, Ceretta LB, Comunello E, et al: Classification of images acquired with colposcopy using artificial neural networks. Cancer Infor. 13:119–124. 2014. View Article : Google Scholar
17	Medved D, Nugues P and Nilsson J: Predicting the outcome for patients in a heart transplantation queue using deep learning. Conf Proc IEEE Eng Med Biol Soc. 2017:pp. 74–77. 2017; PubMed/NCBI