Akciğer Kanser Tipi Tespitinde Etkili Bir Görüntü Çoğullama Tekniği

Son yıllarda derin öğrenme mimarilerinin sınıflama ve tahmin üzerine yüksek başarımlara sahip olması bu alanlara ilgiyi artırmıştır. Özellikle medikal alanlarda hastalık tanısında bilgisayar tabanlı karar destek sistemlerinin yaygınlaşması ile veri setlerinin önemi ve paylaşılması da ön plana çıkmıştır. Ancak oluşturulan veri setlerinin derin mimariler için yeterli veri sayısına sahip olmaması sınıflama performansı açısından sorun olabilmektedir. Veri miktarının artırılması ise çoğu zaman maliyetli, zaman alıcı ve ilgili uzmanın her zaman bulunamaması sebebiyle mümkün olamamaktadır. Bahsedilen durumlar veri çoğullama yöntemlerinin devreye girmesini ve bu alana yönelmeyi gerektirmiştir. Bu çalışmada Dalgacık aktivasyon fonksiyonlu Aşırı Öğrenme Makinası Oto Kodlayıcı (D-AÖM-OK) tabanlı veri artırma yöntemi önerilmiştir. Önerilen yöntem dünyadaki kanser oranının en büyük yüzdesini içeren akciğer kanser sınıflaması üzerinde test edilmiştir. Çoğullanan eğitim veri seti GoogLeNet mimarisine giriş olarak uygulanmıştır. D-AÖM-OK’ın performansı çoğullanmamış ve geleneksel çoğullama yöntemleri ile karşılaştırılmıştır. Önerilen yöntem çoğullanmamış duruma kıyasla %11,12, klasik yöntemlerle çoğullanmış veri setine göre ise %2,55 oranında daha yüksek başarım göstermektedir.

An Effective Image Augmenting Technique in Detection of Lung Cancer Types

In recent years, the high performance of deep learning architectures on classification and prediction has increased the interest in these areas. The importance and sharing of data sets has come to the fore with the widespread use of computer-based decision support systems in the diagnosis of disease, especially in medical fields. However, the fact that the generated data sets are not sufficient for deep architectures can be a problem in terms of classification performance. Increasing the amount of data is often not possible because it is costly, time consuming and the relevant specialist is not always available. The mentioned situations necessitated the introduction of data augmenting methods and tending to this area. In this study, an Extreme Learning Machine Auto-Encoder (W-ELM-AE) based data augmentation method with Wavelet activation function is proposed. The proposed method has been tested on the lung cancer classification, which includes the largest percentage of cancer rates in the world. The augmented training dataset is applied as an input to the GoogLeNet architecture. The performance of W-ELM-AE has been compared with non-augmented and traditional augmenting methods. The proposed method shows a higher performance of 11.12% compared to the unaugmented case, and 2.55% higher than the dataset augmented with classical methods.

___

  • Shorten, C., & Khoshgoftaar, T. M. (2019). A survey on image data augmentation for deep learning. Journal of big data, 6(1), 1-48.
  • Sambasivan, N., Kapania, S., Highfill, H., Akrong, D., Paritosh, P., & Aroyo, L. M. (2021, May). “Everyone wants to do the model work, not the data work”: Data Cascades in High-Stakes AI. In proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (pp. 1-15).
  • Zhang, Y., Choon, N. H., Lin, H., Abd Yusof, N. F., Zhang, Y., & Wang, X. (2022). An Overview of Analysis of Medical Images Using Data Visualization and Deep Learning Applications. Forest Chemicals Review, 2321-2332.
  • Ingle, K., Chaskar, U., & Rathod, S. (2021, July). Lung Cancer Types Prediction Using Machine Learning Approach. In 2021 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT) (pp. 01-06). IEEE.
  • Wang, F., Zhong, S. H., Peng, J., Jiang, J., & Liu, Y. (2018, February). Data augmentation for EEG-based emotion recognition with deep convolutional neural networks. In International Conference on Multimedia Modeling (pp. 82-93).
  • S. Hussein, R. Gillies, K. Cao, Q. Song, and U. Bagci, "Tumornet:Lung nodule characterization using multi-view convolutional neural network with gaussian process," in 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017), April 2017, pp. 1007–1010.
  • Nishizaki, H. (2017, December). Data augmentation and feature extraction using variational autoencoder for acoustic modeling. In 2017 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) (pp. 1222-1227). IEEE.
  • Frid-Adar, M., Klang, E., Amitai, M., Goldberger, J., & Greenspan, H. (2018, April). Synthetic data augmentation using GAN for improved liver lesion classification. In 2018 IEEE 15th international symposium on biomedical imaging (ISBI 2018) (pp. 289-293). IEEE.
  • Ferreira, J., Ferro, M., Fernandes, B., Valenca, M., Bastos-Filho, C., & Barros, P. (2017, November). Extreme learning machine autoencoder for data augmentation. In 2017 IEEE Latin American Conference on Computational Intelligence (LA-CCI)(pp. 1-6). IEEE.
  • K. Munir, H. Elahi, A. Ayub, F. Frezza, and A. Rizzi, (2019) Cancer diagnosis using deep learning: A bibliographic review, Cancers (Basel)., 11(9): 1–36, doi: 10.3390/cancers11091235.
  • Manikandan, T., Devi, B., & Helanvidhya, T. A (2019) Computer-Aided Diagnosis System for Lung Cancer Detection with Automatic Region Growing, Multistage Feature Selection and Neural Network Classifier.
  • Cifci, M. Derin Öğrenme Metodu Kullanarak BT Görüntülerinden Akciğer Kanseri Teşhisi. Dokuz Eylül Üniversitesi Mühendislik Fakültesi Fen ve Mühendislik Dergisi, 24(71), 487-500.
  • Mohammed SH (2021), Detection Of Cancer Area In Lung Images With The Help Of Deep Learning Algorithm, Elazığ Chest CT-Scan Images Dataset (2020) Hany M. https://www.kaggle.com/mohamedhanyyy/chest-ctscan-images.
  • Wei, K., Li, T., Huang, F., Chen, J., & He, Z. (2022). Cancer classification with data augmentation based on generative adversarial networks. Frontiers of Computer Science, 16(2), 1-11. SPIE-AAPM-NCI BreastPathQ: (2021) http://breastpathq.grand-challenge.org/ April
  • Polat M. (2021) Göğüs x-ray görüntülerinde derin öğrenme algoritmaları ile akciğer bölütlemesi, Erzurum
  • Kandel, I., Castelli, M., & Manzoni, L. (2022). Brightness as an Augmentation Technique for Image Classification. Emerging Science Journal, 6(4), 881-892.
  • Francisco JM-B, Fiammetta S, Jose MJ, Daniel U, Leonardo F. Forward noise adjustment scheme for data augmentation. arXiv preprints. 2018.
  • Huang, G. B., Zhu, Q. Y., and Siew, C. K. (2004) Extreme learning machine: a new learning scheme of feedforward neural networks, IEEE International Joint Conference on Neural Networks, Budapest, 25-29 July, 2, 985-990.
  • Arı, B. , Alçin, Ö. F. & Şengür, A. (2022). A Lung Sound Classification System Based on Data Augmenting Using ELM-Wavelet-AE . Turkish Journal of Science and Technology , 17 (1) , 79-88 . DOI: 10.55525/tjst.1063039
  • Güner, A., Alçin, Ö. F., & Şengür, A. (2019). Automatic digital modulation classification using extreme learning machine with local binary pattern histogram features. Measurement, 145, 214-225.
  • Ari, B., Siddique, K., Alçin, Ö. F., Aslan, M., Şengür, A., & Mehmood, R. M. (2022). Wavelet ELM-AE Based Data Augmentation and Deep Learning for Efficient Emotion Recognition Using EEG Recordings. IEEE Access.
  • Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., ... & Rabinovich, A. (2015). Going deeper with convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1-9)
  • Toğaçar, M., & Ergen, B. (2019). Biyomedikal Görüntülerde Derin Öğrenme ile Mevcut Yöntemlerin Kıyaslanması. Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 31(1), 109-121.
  • Chicco, D., & Jurman, G. (2020). The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC genomics, 21(1), 1-13.
  • Rehman, A., Kashif, M., Abunadi, I., & Ayesha, N. (2021). Lung cancer detection and classification from chest CT scans using machine learning techniques. In 2021 1st International Conference on Artificial Intelligence and Data Analytics (CAIDA) (pp. 101-104). IEEE.
  • Sadhu A, Mehra A, Kulshrestha A, Goyal V. (2022) Cancer detection from medical images using deep convolution neural networks. Int J Adv Res Comput Commun Eng. ;11(3):70–81.