Medyan Filtre Tabanlı Önişlem Kullanılarak Geliştirilmiş Bir Derin Öğrenme Temelli Rahim Ağzı Kanseri Tespiti

Öz

Rahim ağzı kanseri, teşhis edildikten sonra diğer kanser türlerine kıyasla tedavi başarısı en yüksek olanıdır ve kadınlar arasında en sık görülen kanser türlerinden biridir. Rahim ağzı kanserinin otomatik sınıflandırılması, tedavi sürecini hızlandırmak ve hastaların hayatta kalma oranlarını artırmak için çok önemlidir. Yetersiz farkındalık, tıbbi imkânların yetersizliği ve pahalı tarama prosedürleri ölüm oranlarını artırmaktadır. Bu yaygın kanser sıklıkla Pap smear, servikografi ve kolposkopi gibi çeşitli görüntüleme testleri ile taranır. Kararlar bu testler yardımıyla verilir, ancak servikal hücrelerin yapısal karmaşıklıkları kararı zorlaştırabilir. Sinir ağlarındaki son gelişmeler, hastalık teşhisinde dikkate değer başarılar göstermektedir. Ayrıca, transfer öğrenme avantajlarından dolayı çoğu araştırmacının dikkatini çekmektedir. Bu makale erken teşhis için transfer öğrenme tabanlı bir serviks kanseri tespit yöntemi sunmaktadır. Pap smear görüntüleri, daha iyi sınıflandırma için, görüntülerden gürültüyü çıkarmak amacıyla derin öğrenme modelinin eğitiminden önce medyan filtresi kullanılarak ön işleme tabi tutulmuştur. Kanserli ve kanserli olmayan servikal hücreler, önceden eğitilmiş ağlar aracılığıyla ayırt edilir. SqueezeNet, VGG-19, AlexNet, ResNet-50 ve InceptionV3 isimli beş popüler önceden eğitilmiş ağ kullanılmış ve problem için karşılaştırılmıştır. SqueezeNet diğer sinirsel yapılarla karşılaştırıldığında en iyi test doğruluğunu (%96.90) elde etmiştir ve bu performans önerilen yöntemi serviks kanseri teşhisi için literatürdeki diğer denetimsiz yaklaşımlar arasında en iyisi yapmaktadır. Ek deneyler, Parabasal ve Metaplastik hücreler olmak üzere iki benzer sınıfın sınıflandırılması için önerilen modelin başarısını da kanıtladı. Sonuçlar, önerilen yaklaşımın rahim ağzı kanseri teşhisi için güvenilir, ucuz ve hızlı bir karar destek sistemi sağlayabileceğini göstermektedir.

Anahtar Kelimeler

• Rahim ağzı kanseri teşhisi
• Evrişimli sinir ağları
• Transfer öğrenme
• Pap smear görüntüleri

An Improved Deep Learning Based Cervical Cancer Detection Using a Median Filter Based Preprocessing

Abstract

Cervical cancer is one of the prevalent type of cancer among women although its treatment success is the highest when compared to other types of cancer once diagnosed. Automatic classification of cervical cancer is essential to accelerate the treatment process and increase the survival rate of the patients. Inadequate awareness, deficiency of medical opportunities, and expensive screening procedures increase the death rates. This common cancer is frequently screened by several imaging tests including Pap smear, cervicography and colposcopy. The decisions are made by the help of these tests, but structural complexities of cervical cells may complicate the decision. Recent developments in neural networks show remarkable achievements in disease diagnosis. Also, transfer learning draws the attention of most of the researchers because of its advantages. This paper presents a transfer learning based cervical cancer detection method for early diagnosis. Pap smear images were preprocessed using median filter before training the deep learning model in order to remove noise from the images for better classification. Cancerous and non-cancerous cervical cells are distinguished through pre-trained networks. Five popular pre-trained networks which are SqueezeNet, VGG-19, AlexNet, ResNet-50 and InceptionV3 have been utilized and compared for the problem. SqueezeNet achieved the best validation accuracy (96.90\%) when compared to other neural structures and this performance makes the proposed method the best among other unsupervised approaches in the literature for cervical cancer diagnosis. Additional experiments also proved the success of the proposed model for the classification of two similar classes, namely Parabasal and Metaplastic cells. The results demonstrate that the proposed approach can provide a confidential, cheap, and fast decision support system for cervical cancer diagnosis.

Keywords

• Cervical cancer diagnosis
• Convolutional neural networks
• Transfer learning
• Pap smear images

References

1. Arishanapally, S. C. (2019). Building VGG19 with Keras. Medium. https://medium.com/@saicharanars/building-vgg19-with-keras-f516101c24cf
2. Arya, M., Mittal, N., & Singh, G. (2016). Cervical cancer detection using segmentation on pap smear images. ACM International Conference Proceeding Series, 25-26-Augu, 1–5. https://doi.org/10.1145/2980258.2980311
3. Bhowmik, M. K., Roy, S. D., Nath, N., & Datta, A. (2018). Nucleus region segmentation towards cervical cancer screening using AGMC-TU Pap-smear dataset. ACM International Conference Proceeding Series, 44–53. https://doi.org/10.1145/3243250.3243258
4. Cearley, D. W., Burke, B., Searle, S., & Walker, M. J. (n.d.). Top 10 Strategic Technology Trends for 2018. In brilliantdude.com. Retrieved May 6, 2020, from http://brilliantdude.com/solves/content/GartnerTrends2018.pdf
5. Çevik, K. K., & Dandıl, E. (2019). Classification of Lung Nodules Using Convolutional Neural Networks on CT Images. 2nd International Turkish World Engineering and Science Congress, 27–35.
6. D, N. D. P., Zhao, L., D, C. H. W. P., & Chang, J. F. (2020). Inception v3 based cervical cell classification combined with artificially extracted features. Applied Soft Computing Journal, 93, 1–8. https://doi.org/10.1016/j.asoc.2020.106311
7. Dong, D., Fang, M.-J., Tang, L., Shan, X.-H., Gao, J.-B., Giganti, F., Wang, R.-P., Chen, X., Wang, X.-X., Palumbo, D., Fu, J., Li, W.-C., Li, J., Zhong, L.-Z., De Cobelli, F., Ji, J.-F., Liu, Z.-Y., & Tian, J. (2020). Deep learning radiomic nomogram can predict the number of lymph node metastasis in locally advanced gastric cancer: an international multi-center study. Annals of Oncology. https://doi.org/10.1016/j.annonc.2020.04.003
8. Dongyao Jia, A., Zhengyi Li, B., & Chuanwang Zhang, C. (2020). Detection of cervical cancer cells based on strong feature CNN-SVM network. Neurocomputing, 411, 112–127. https://doi.org/10.1016/j.neucom.2020.06.006

9. Ekici, S., & Jawzal, H. (2020). Breast cancer diagnosis using thermography and convolutional neural networks. Medical Hypotheses, 137, 109542. https://doi.org/10.1016/j.mehy.2019.109542
10. Feng, X., Jiang, Y., Yang, X., Du, M., & Li, X. (2019). Computer vision algorithms and hardware implementations: A survey . Integration, the VLSI Journal, 69, 309–320. https://reader.elsevier.com/reader/sd/pii/S0167926019301762?token=53D2C5EBAAF7604A55486ED6D40948D4F9C78179B7E62A07E9B73DC972CA30E58A32D661873FA500AAEDA26F64AA61B7
11. Gatys, L. A., Ecker, A. S., & Bethge, M. (2017). Texture and art with deep neural networks. Current Opinion in Neurobiology, 46, 178–186. https://doi.org/10.1016/j.conb.2017.08.019
12. Gautam, A., & Raman, B. (2021). Towards effective classification of brain hemorrhagic and ischemic stroke using CNN. Biomedical Signal Processing and Control, 63, 102178. https://doi.org/10.1016/j.bspc.2020.102178
13. Guo, Y., Shang, X., & Li, Z. (2019). Identification of cancer subtypes by integrating multiple types of transcriptomics data with deep learning in breast cancer. Neurocomputing, 324, 20–30. https://doi.org/10.1016/j.neucom.2018.03.072
14. Haryanto, T., Sitanggang, I. S., Agmalaro, M. A., & Rulaningtyas, R. (2020). The Utilization of Padding Scheme on Convolutional Neural Network for Cervical Cell Images Classification. International Conference on Computer Engineering, Network, and Intelligent Multimedia (CENIM 2020), 34–38. https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9297895
15. Hashimoto, Y., Watanabe, Y., Takano, H., Setsuda, A., Ohno, I., Imaoka, H., Sasaki, M., Watanabe, K., Umemoto, K., Kimura, G., Shibuki, T., Kan, M., Mitsunaga, S., & Ikeda, M. (2019). 590 – High Diagnostic Yield Using Advanced Artificial Intelligence in Cytology of Pancreatic Cancer by Eus-Fna. Gastroenterology, 156(6), S-115. https://doi.org/10.1016/s0016-5085(19)37081-7 Immunohistochemistry. (2020). Wikipedia.
16. Indrabayu, Fatmasari, A. R., & Nurtanio, I. (2017). A colour space based detection for cervical cancer using fuzzy C-means clustering. ACM International Conference Proceeding Series, 137–141. https://doi.org/10.1145/3121138.3121196
17. Ke, J., Deng, J., & Lu, Y. (2019). Noise Reduction with Image Inpainting : An Application in Clinical Data Diagnosis. SIGGRAPH’19.
18. Ke, J., Jiang, Z., Liu, C., Bednarz, T., Sowmya, A., & Liang, X. (2019). Selective Detection and Segmentation of Cervical Cells. ICBBT’19. https://doi.org/https://doi.org/10.1145/3340074.3340081
19. Li, C., Xue, D., Zhou, X., Zhang, J., Zhang, H., Yao, Y., Kong, F., Zhang, L., & Sun, H. (2019). Transfer Learning Based Classification of Cervical Cancer Immunohistochemistry Images. ISICDM’19, 24–26. https://doi.org/https://doi.org/10.1145/3364836.3364857
20. Li, H., Galperin-Aizenberg, M., Pryma, D., Simone, C. B., & Fan, Y. (2018). Unsupervised machine learning of radiomic features for predicting treatment response and overall survival of early stage non-small cell lung cancer patients treated with stereotactic body radiation therapy. Radiotherapy and Oncology, 129(2), 218–226. https://doi.org/10.1016/j.radonc.2018.06.025
21. Lundervold, A. S., & Lundervold, A. (2019). An overview of deep learning in medical imaging focusing on MRI. Zeitschrift Für Medizinische Physik, 29(2), 102–127. https://doi.org/10.1016/J.ZEMEDI.2018.11.002
22. Martínez-Más, J., Bueno-Crespo, A., Imbernón, B., Cecilia, J. M., Martínez-España, R., Remezal-Solano, M., Sánchez-Espinosa, A., Ortiz-Reina, S., & Martínez-Cendán, J. P. (2018). Deep learning approach for classifying papanicolau cervical smears. ACM International Conference Proceeding Series. https://doi.org/10.1145/3229710.3229732
23. Mousser, W., & Ouadfel, S. (2019). Deep Feature Extraction for Pap-Smear Image Classification : A Comparative Study. ICCTA 2019. https://doi.org/https://doi.org/10.1145/3323933.3324060
24. Nitta, S., Tsutsumi, M., Sakka, S., Endo, T., Hashimoto, K., Hasegawa, M., Hayashi, T., Kawai, K., & Nishiyama, H. (2019). Machine learning methods can more efficiently predict prostate cancer compared with prostate-specific antigen density and prostate-specific antigen velocity. Prostate International, 7(3), 114–118. https://doi.org/10.1016/j.prnil.2019.01.001
25. Pitas, I., & Venetsanopoulos, A. N. (1990). Median Filters. In Nonlinear Digital Filters. Springer US. https://doi.org/10.1007/978-1-4757-6017-0
26. Plissiti, M. E., Dimitrakopoulos, P., Sfikas, G., Nikou, C., Krikoni, O., & Charchanti, A. (2018). SIPAKMED : A NEW DATASET FOR FEATURE AND IMAGE BASED CLASSIFICATION OF NORMAL AND PATHOLOGICAL CERVICAL CELLS IN PAP SMEAR IMAGES Dept . of Computer Science & Engineering , University of Ioannina , Greece Dept . of Anatomy-Histology and Embryology , Facul. 25th IEEE International Conference on Image Processing (ICIP), 3144–3148. https://doi.org/10.1109/ICIP.2018.8451588
27. Senturk, Z. K., & Kara, R. (2014). Breast Cancer Diagnosis Via Data Mining: Performance Analysis of Seven Different Algorithms. Computer Science & Engineering: An International Journal, 4(1), 35–46. https://doi.org/10.5121/cseij.2014.4104
28. Shi, Y., Dai, D., Liu, C., & Yan, H. (2009). Sparse discriminant analysis for breast cancer biomarker identification and classification. Progress in Natural Science, 19(11), 1635–1641. https://doi.org/10.1016/j.pnsc.2009.04.013
29. Singh, S. K., & Goyal, A. (2020). Three stage cervical cancer classifier based on hybrid ensemble learning with modified binary PSO using pretrained neural networks. The Imaging Science Journal, 68(1), 41–55. https://doi.org/10.1080/13682199.2020.1734306
30. Thomas, S. M., Lefevre, J. G., Baxter, G., & Hamilton, N. A. (2021). Interpretable deep learning systems for multi-class segmentation and classification of non-melanoma skin cancer. Medical Image Analysis, 68, 101915. https://doi.org/10.1016/j.media.2020.101915
31. Tsang, S.-H. (2018). Inception-v3 — 1st Runner Up (Image Classification) in ILSVRC 2015. Medium.
32. Win, K. P., Kitjaidure, Y., Hamamoto, K., & Myo Aung, T. (2020). Computer-Assisted Screening for Cervical Cancer Using Digital Image Processing of Pap Smear Images. Applied Sciences, 10(5), 1800. https://doi.org/10.3390/app10051800
33. Win, K. P., Kitjaidure, Y., Phyu, M., & Hamamoto, K. (2019). Cervical Cancer Detection and Classification from Pap Smear Images. ICBSP ’19, 47–54. https://doi.org/https://doi.org/10.1145/3366174.3366178
34. World Health Organization. (n.d.). Cancer. Retrieved April 21, 2020, from https://www.who.int/health-topics/cancer#tab=tab_1
35. Xue, Z., Antani, S., Long, L. R., & Thoma, G. R. (2010). An online segmentation tool for cervicographic image analysis. IHI’10 - Proceedings of the 1st ACM International Health Informatics Symposium, 425–429. https://doi.org/10.1145/1882992.1883056
36. Yang, W., Gou, X., Xu, T., Yi, X., & Jiang, M. (2019). Cervical Cancer Risk Prediction Model and Analysis of Risk Factors based on Machine Learning. ICBBT’19.
37. Yang, Z., Yi, D., & Shen, J. (2019). Computer-aided cervical cancer screening method based on multi-spectral narrow-band imaging. ICBIP ’19, 62–66. https://doi.org/10.1145/3354031.3354037
38. Zhang, L., Lu, L., Nogues, I., Summers, R. M., Liu, S., & Yao, J. (2017). DeepPap: Deep convolutional networks for cervical cell classification. IEEE Journal of Biomedical and Health Informatics, 21(6), 1633–1643. https://doi.org/10.1109/JBHI.2017.2705583