Kulak İçi Hastalıklarının Derin Öğrenme Mimarileriyle Sınıflandırılması ve Karşılaştırılması

Year 2023, Issue: 51, 75 - 85, 31.08.2023

Furkancan Demircan , Murat Ekinci , Zafer Cömert

https://doi.org/10.31590/ejosat.1224070

Abstract

Otitis media (OM), kulak zarı içerisinde oluşan akıntılı, enfeksiyonel hastalıkları tanımlamaktadır. Kulak mumu (earwax), kulak zarı içerisinde bakteri oluşumunu önleyen savunma mekanizmasının aşırı birikimi sonucunda kulakta işitme kaybı oluşmasına neden olan hastalıktır. Kulak zarı içerisinde kalsiyum birikimi sonucunda saydamlığını ve esnekliğini kaybetmesine miringoskleroz denmektedir. Bu hastalıkların tanısı Kulak Burun Boğaz (KBB) uzmanları tarafından kulak zarının otoskopla incelenmesi sonucunda koyulmaktadır ve hataya açıktır. Bu çalışmada, bu problemin çözümüne katkı sağlamak ve bir karar destek sistemi sunmak amacıyla derin öğrenme modelleriyle kulak zarı hastalıklarına ait görüntüler sınıflandırılmıştır. Veri seti olarak 4 sınıf ve 880 görüntünün bulunduğu Ear Imagery veri seti seçilmiştir. Sınıflandırma işlemi için AlexNet, ResNet50, ResNet101, ResNet50V2, ResNet101V2, InceptionV3, Xception ve InceptionResNetV2 derin öğrenme modelleri seçilmiştir. En yüksek başarı değeri %94 ile InceptionResNetV2 mimarisinden ve en hızlı sonuç 438 saniye ile AlexNet mimarisinden elde edilmiştir. Bu yaklaşımla kulak zarına ait hastalıkların potansiyel uzman hatalarından arındırılarak otonom bir sistem ile gerçekleştirilebileceği gösterilmiştir. Gelecekte klinik alanda böyle bir sistemin kullanılması; uzmanların karar verme sürecini destekleyebilir ve hataya açık olan değerlendirme sürecinin daha objektif ve tekrar edilebilir bir şekilde yönetilmesini sağlayabilir.

Keywords

Biyomedikal işaret işleme, Otitis media, Derin Öğrenme, Sınıflandırma

Classification and comparison of ear diseases with deep learning architectures

Abstract

Otitis Media (OM) is infectious disease with discharge in the eardrum. Earwax is a disease that causes hearing loss in the ear as a result of excessive accumulation of the defense mechanism that prevents the formation of bacteria in the eardrum. The loss of transparency and flexibility as a result of calcium accumulation in the eardrum is called myringosclerosis. The diagnosis of these diseases is made by otolaryngologists using an otoscopy examination of the eardrum and this process is prone to error. In this study, otoscopy images were classified with deep learning models to solve this problem. The Ear Imagery dataset with 4 classes and 880 images was chosen as the dataset. AlexNet, ResNet50, ResNet101, ResNet50V2, ResNet101V2, InceptionV3, Xception and InceptionResNetV2 deep learning models were selected for classification. The highest success value was obtained from InceptionResNetV2 architecture with 94% and the fastest result was obtained from AlexNet architecture with 438 seconds. With this approach, it has been shown that diseases of the eardrum can be treated with an autonomous system, freeing from expert error. In the future, such a system in the clinical field will be able to reduce labor and error.

Keywords

Biomedical signal processing, Otitis media, Deep Learning, Classification

References

• Alake, R. (2020, 22 Aralık). Deep Learning: Understanding The Inception Module. Erişim adresi: https://towardsdatascience.com/deep-learning-understand-the-inception-module-56146866e652
• Alhudhaif, A., Cömert, Z. ve Polat, K. 2021. “Otitis media detection using tympanic membrane images with a novel multi-class machine learning algorithm”, PeerJ Computer Science, 7, 405
• Başaran, E., Cömert, Z. ve Çelik, Y. 2020. “Convolutional neural network approach for automatic tympanic membrane detection and classification”. Biomedical Signal Processing and Control, 56, 101734.
• Boesh, G. (2022). Deep Residual Networks (ResNet, ResNet50) – Guide in 2022. Erişim adresi: https://viso.ai/deep-learning/resnet-residual-neural-network/
• Dash, A. B. (2021, 30 Haziran). Top 10 Activation Function's Advantages & Disadvantages. Erişim adresi: https://www.linkedin.com/pulse/top-10-activation-functions-advantages-disadvantages-dash
• Fabien, M. (2019, 20 Mart). Xception Model and Depthwise Separable Convolutions. Erişim adresi: https://maelfabien.github.io/deeplearning/xception/#
• Google Developer. (2022, 18 Temmuz). Classification: ROC Curve and AUC. Erişim adresi: https://developers.google.com/machine-learning/crash-course/classification/roc-and-auc
• Hayes, K. (2022, 11 Nisan). An Overview of Myringosclerosis and Tympanosclerosis. Erişim adresi: https://www.verywellhealth.com/myringosclerosis-and-tympanosclerosis-1191943#:~:text=Myringosclerosis%20and%20tympanosclerosis%20are%20similar,due%20to%20accumulated%20calcium%20deposits.
• He, K., Zhang, X., Ren, S. ve Sun, J. 2015. “Delving deep into rectifiers: Surpassing human-level performance on imagenet classification”. In Proceedings of the IEEE international conference on computer vision, 1026-1034
• He, K., Zhang, X., Ren, S. ve Sun, J. 2016. “Deep residual learning for image recognition”. In Proceedings of the IEEE conference on computer vision and pattern recognition, 770-778
• IBM Cloud. (2020, 1 Mayıs). Deep Learning. Erişim adresi: https://www.ibm.com/cloud/learn/deep-learning#:~:text=Deep%20learning%20is%20a%20subset,from%20large%20amounts%20of%20data.
• Jacob, T. (2022, 25 Şubat). Vanishing Gradient Problem, Explained. Erişim adresi: https://www.kdnuggets.com/2022/02/vanishing-gradient-problem.html#:~:text=When%20there%20are%20more%20layers,this%20the%20vanishing%20gradient%20probl em.
• Mayo Clinic (2022, 12 Temmuz). Earwax Blockage. Erişim adresi: https://www.mayoclinic.org/diseases-conditions/earwax-blockage/symptoms-causes/syc-20353004
• Mohammed, K. K., Hassanien, A. E. ve Afify, H. M. 2022. “Classification of Ear Imagery Database using Bayesian Optimization based on CNN-LSTM Architecture”. Journal of Digital Imaging, 1-15
• Raj, B. (2018, 29 Mayıs). A Simple Guide to the Versions of the Inception Network. Erişim adresi: https://towardsdatascience.com/a-simple-guide-to-the-versions-of-the-inception-network-7fc52b863202
• Sahling, M., Benders, T., König, O., Boll-Avetisyan, N. 2021. Building a Phoneme Inventory through Blocked Ears: The Effects of Prior Otitis Media with Effusion on Children's Phoneme Discrimination.
• Serdar Yegualp. (2022, 3 Haziran). What is TensorFlow? The machine learning library explained. Erişim adresi: https://www.infoworld.com/article/3278008/what-is-tensorflow-the-machine-learning-library-explained.html
• Sundgaard, J. V., Harte, J., Bray, P., Laugesen, S., Kamide, Y., Tanaka, C. ve Christensen, A. N. 2021. “Deep metric learning for otitis media classification”. Medical Image Analysis, 71, 102034.
• Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J. ve Wojna, Z. 2016. “Rethinking the inception architecture for computer vision”. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2818-2826
• Tran, T. T., Fang, T. Y., Pham, V. T., Lin, C., Wang, P. C. ve Lo, M. T. 2018. “Development of an automatic diagnostic algorithm for pediatric otitis media”. Otology & Neurotology, 39, 1060-1065.
• Uçar, M., Akyol, K., Atila, Ü. M. İ. T. ve Uçar, E. 2021. “Classification of different tympanic membrane conditions using fused deep hypercolumn features and bidirectional LSTM”. IRBM.
• Viscaino, M., Maass, J. C., Delano, P. H., Torrente, M., Stott, C. ve Auat Cheein, F. 2020. “Computer-aided diagnosis of external and middle ear conditions: A machine learning approach”. Plos one, 15, 0229226.
• Wu, Z., Lin, Z., Li, L., Pan, H., Chen, G., Fu, Y. ve Qiu, Q. 2021. “Deep learning for classification of pediatric otitis media”. The Laryngoscope, 131, 2344-2351.
• Yavuz, H ve arkadaşları. (2000, Ekim). Otitis Media. Erişim adresi: https://www.ttb.org.tr/STED/sted1000/2.html#top