Five-Class Classification of Cervical Pap Smear Images: A Study of CNN-Error-Correcting SVM Models
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Objectives: Different complex strategies of fusing handcrafted descriptors and features from convolutional neural network(CNN) models have been studied, mainly for two-class Papanicolaou (Pap) smear image classification. This paper explores asimplified system using combined binary coding for a five-class version of this problem. Methods: This system extracted featuresfrom transfer learning of AlexNet, VGG19, and ResNet50 networks before reducing this problem into multiple binarysub-problems using error-correcting coding. The learners were trained using the support vector machine (SVM) method. The outputs of these classifiers were combined and compared to the true class codes for the final prediction. Results: Despitethe superior performance of VGG19-SVM, with mean ± standard deviation accuracy and sensitivity of 80.68% ± 2.00% and80.86% ± 0.45%, respectively, this model required a long training time. There were also false-negative cases using both theVGGNet-SVM and ResNet-SVM models. AlexNet-SVM was more efficient in terms of running speed and prediction consistency. Our findings also showed good diagnostic ability, with an area under the curve of approximately 0.95. Further investigationalso showed good agreement between our research outcomes and that of the state-of-the-art methods, with specificityranging from 93% to 100%. Conclusions: We believe that the AlexNet-SVM model can be conveniently applied for clinicaluse. Further research could include the implementation of an optimization algorithm for hyperparameter tuning, as well asan appropriate selection of experimental design to improve the efficiency of Pap smear image classification.
Abstract
Objectives: Different complex strategies of fusing handcrafted descriptors and features from convolutional neural network(CNN) models have been studied, mainly for two-class Papanicolaou (Pap) smear image classification. This paper explores asimplified system using combined binary coding for a five-class version of this problem. Methods: This system extracted featuresfrom transfer learning of AlexNet, VGG19, and ResNet50 networks before reducing this problem into multiple binarysub-problems using error-correcting coding. The learners were trained using the support vector machine (SVM) method. The outputs of these classifiers were combined and compared to the true class codes for the final prediction. Results: Despitethe superior performance of VGG19-SVM, with mean ± standard deviation accuracy and sensitivity of 80.68% ± 2.00% and80.86% ± 0.45%, respectively, this model required a long training time. There were also false-negative cases using both theVGGNet-SVM and ResNet-SVM models. AlexNet-SVM was more efficient in terms of running speed and prediction consistency. Our findings also showed good diagnostic ability, with an area under the curve of approximately 0.95. Further investigationalso showed good agreement between our research outcomes and that of the state-of-the-art methods, with specificityranging from 93% to 100%. Conclusions: We believe that the AlexNet-SVM model can be conveniently applied for clinicaluse. Further research could include the implementation of an optimization algorithm for hyperparameter tuning, as well asan appropriate selection of experimental design to improve the efficiency of Pap smear image classification.
참고문헌 (30)
- Pelkofski E, "Cervical cancer in women aged 35 years and younger", Clin Ther, 38(3), pp.459~466 (2016)
- Yoruk S, "Determination of knowledge levels, attitude and behaviors of female university students concerning cervical cancer, human papiloma virus and its vaccine", BMC Womens Health, 16, pp.51~ (2016)
- Watson M, "National assessment of HPV and Pap tests : changes in cervical cancer screening, National Health Interview Survey", Prev Med, 100, pp.243~247 (2017)
- MacLaughlin KL, "Trends over time in pap and pap-HPV cotesting for cervical cancer screening", J Womens Health(Larchmt), 28(2), pp.244~249 (2019)
- Akbarzadeh MA, "Implications for cancer care in Iran during COVID-19 pandemic", Radiother Oncol, 148, pp.211~212 (2020)
- Tsang-Wright F, "Breast cancer surgery after the COVID-19 pandemic", Future Oncol, 16(33), pp.2687~2690 (2020)
- Price GJ, "Computerized diagnostic decision support system for the classification of preinvasive cervical squamous lesions", Hum Pathol, 34(11), pp.1193~1203 (2003)
- Garrison A, "Cancer concepts: a guidebook for the non-oncologist" (2015), University of Massachusetts Medical School
- William W, "A review of image analysis and machine learning techniques for automated cervical cancer screening from pap-smear images", Comput Methods Programs Biomed, 164, pp.15~22 (2018)
- Selvathi D, "Classification in BioApps", pp.295~322 (2008), Springer
- Win KY, "Computer aided diagnosis system for detection of cancer cells on cytological pleural effusion images", Biomed Res Int, 2018, pp.6456724~ (2018)
- Fekri-Ershad S, "Pap smear classification using combination of global significant value, texture statistical features and time series features", Multimed Tools Appl, 78(22), pp.31121~31136 (2019)
- Cervantes J, "A comprehensive survey on support vector machine classification : applications, challenges and trends", Neurocomputing, 408, pp.189~215 (2020)
- Chauhan VK, "Problem formulations and solvers in linear SVM : a review", Artif Intell Rev, 52(2), pp.803~855 (2019)
- Aoyagi K, "Simple method to construct process maps for additive manufacturing using a support vector machine", Addit Manuf, 27, pp.353~362 (2019)
- Taha B, "Medical image understanding and analysis", pp.261~272 (2017), Springer
- Khamparia A, "Internet of health things-driven deep learning system for detection and classification of cervical cells using transfer learning", J Supercomput, 76, pp.8590~8608 (2020)
- Jia AD, "Detection of cervical cancer cells based on strong feature CNN-SVM network", Neurocomputing, 411, pp.112~127 (2020)
- Chen W, "Improving computeraided cervical cells classification using transfer learning based snapshot ensemble", Appl Sci, 10(20), pp.7292~ (2020)
- Kurnianingsih, "Segmentation and classification of cervical cells using deep learning", IEEE Access, 7, pp.116925~116941 (2019)
- Ghoneim A, "Cervical cancer classification using convolutional neural networks and extreme learning machines", Future Gener Comput Syst, 102, pp.643~649 (2020)
- Zhang L, "DeepPap : deep convolutional networks for cervical cell classification", IEEE J Biomed Health Inform, 21(6), pp.1633~1643 (2017)
- Wang P, "Adaptive pruning of transfer learned deep convolutional neural network for classification of cervical pap smear images", IEEE Access, 8, pp.50674~50683 (2020)
- Hussain E, "A comprehensive study on the multi-class cervical cancer diagnostic prediction on pap smear images using a fusionbased decision from ensemble deep convolutional neural network", Tissue Cell, 65, pp.101347~ (2020)
- AlMubarak HA, "A hybrid deep learning and handcrafted feature approach for cervical cancer digital histology image classification", Int J Healthc Inf Syst Inform, 14(2), pp.66~87 (2019)
- Johnson DB, "Diagnosis and treatment of cervical intraepithelial neoplasia in general practice", BMJ, 299(6707), pp.1083~1086 (1989)
- Mutombo AB, "Efficacy of commercially available biological agents for the topical treatment of cervical intraepithelial neoplasia : a systematic review", Syst Rev, 8(1), pp.132~ (2019)
- Smith CM, "MicroRNAs, development of Barrett's esophagus, and progression to esophageal adenocarcinoma", World J Gastroenterol, 16(5), pp.531~537 (2010)
- Pham DS, "Joint learning and dictionary construction for pattern recognition", Proceedings of 2008 IEEE Conference on Computer Vision and Pattern Recognition, pp.1~8 (2018)
- Wang Q, "Benchmarking deep learning techniques for face recognition", J Vis Commun Image Represent, 65, pp.102663~ (2019)