SavedAdd your journal

Journals / Konya Journal of Engineering Sciences / 2021 / Cilt: 9 Sayı: 1

Classification of Pneumonia Using Pre-Trained Deep Networks with Chest X-Ray Images

ÖNCEDEN EĞİTİLMİŞ DERİN AĞLAR İLE GÖĞÜS RÖNTGENİ GÖRÜNTÜLERİ KULLANARAK PNÖMONİ SINIFLANDIRILMASI

Mehmet Bilal ER
Pages
193–204
Publication Date
2021-03-02
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The highest classification success is obtained from the hybrid feature vector as 98.32%.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Pneumonia is a lung infection that can be caused by bacteria, viruses, or fungi.
📊
Key Findings & Data:Early detection and treatment of pneumonia can significantly reduce mortality.
💡
Conclusion & Impact:The highest classification success is obtained from the hybrid feature vector as 98.32%.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Pnömoni · Evrişimsel sinir ağlar · AlexNet

Abstract

Pneumonia is a lung infection that can be caused by bacteria, viruses, or fungi. The infection causes the lungs to become inflamed and filled with fluid or pus. It can be a serious and life-threatening disease. Many people die every year due to pneumonia worldwide. Early detection and treatment of pneumonia can significantly reduce mortality. For this reason, this research is to propose a method based on pre-trained deep network models using x-ray images to detect pneumonia. Various pre-trained Convolutional Neural Networks were used as feature extractors to classify chest x-ray images into two classes without pneumonia and pneumonia. AlexNet, VGG16, ResNet (ResNet18, ResNet50, ResNet101) models are preferred as pre-trained deep network models. The hybrid feature vector is obtained by combining the features obtained from these models. As the classifier, Support Vector Machines (SVM) and Softmax in the last layer of deep networks are used. Experiments are carried out on the data set commonly used in the literature. The highest classification success is obtained from the hybrid feature vector as 98.32%.

Özet

Pnömoni, bakterilerin, virüslerin veya mantarların neden olabileceği bir akciğer enfeksiyonudur. Enfeksiyon, akciğerlerin hava keselerinin iltihaplanmasına ve sıvı veya irin ile dolmasına neden olur. Ciddi ve hayatı tehdit eden bir hastalık olabilir. Dünya genelinde her yıl pnömoni nedeniyle çok sayıda kişi ölmektedir. Pnömoninin erken tespiti ve tedavisi, ölüm oranlarını önemli ölçüde azaltabilmektedir. Bu nedenle, bu araştırmada pnömoniyi tespit etmek için röntgen görüntüleri kullanarak önceden eğitilmiş derin öğrenme modellerine dayanan yöntem önerilmektir. Göğüs röntgen görüntülerini pnömoni ve pnömoni olmayan iki sınıfta sınıflandırmak için çeşitli önceden eğitilmiş Evrişimsel Sinir Ağları özellik çıkarıcı olarak kullanılmıştır. Önceden eğitilmiş derin öğrenme modelleri olarak AlexNet, VGG16, ResNet (ResNet18, ResNet50, ResNet101) modelleri tercih edilmiştir. Bu modellerden elde edilen özellikler birleştirilerek hibrit özellik vektörü elde edilmiştir. Sınıflandırıcı olarak Destek Vektör Makineleri (DVM) ve derin öğrenme modellerinin son katmanında bulunan Softmax kullanılmıştır. Deneyler literatürde yaygın kullanılan veri seti üzerinde yapılmıştır. En yüksek sınıflandırma başarısı %98,32 olarak hibrit özellik vektöründen elde edilmiştir.

🔖

Cite This Article

ER, M. B. (2021). Classification of Pneumonia Using Pre-Trained Deep Networks with Chest X-Ray Images. Konya Journal of Engineering Sciences, Cilt: 9 Sayı: 1, 193–204. https://ajindex.com/articles/909108

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (10)

  1. Anthimopoulos, M., Christodoulidis, S., Ebner, L., Christe, A., & Mougiakakou, S. (2016). Lung Pattern Classification for Interstitial Lung Diseases Using a Deep Convolutional Neural Network. IEEE Transactions on Medical Imaging, 35(5), 1207–1216. https://doi.org/10.1109/tmi.2016.2535865
  2. Aydoğdu, M., Ozyilmaz, E., Aksoy, H., Gürsel, G., & Ekim, N. (2010). Mortality prediction in community- acquired pneumonia requiring mechanical ventilation; values of pneumonia and intensive care unit severity scores. Tüberküloz ve Toraks, 58, 25–34.
  3. Davies, H. D., Wang, E. E., Manson, D., Babyn, P., & Shuckett, B. (1996). Reliability of the chest radiograph in the diagnosis of lower respiratory infections in young children. The Pediatric Infectious Disease Journal, 15(7), 600–604. https://doi.org/10.1097/00006454-199607000-00008
  4. Er, M. B., & Aydilek, I. B. (2019). Music Emotion Recognition by Using Chroma Spectrogram and Deep Visual Features. International Journal of Computational Intelligence Systems, 12(2), 1622–1634. https://doi.org/https://doi.org/10.2991/ijcis.d.191216.001
  5. Guan, Q., Huang, Y., Zhong, Z., Zheng, Z., Zheng, L., & Yang, Y. (2019). Thorax Disease Classification with Attention Guided Convolutional Neural Network. Pattern Recognition Letters, 131. https://doi.org/10.1016/j.patrec.2019.11.040
  6. Jain, R., Nagrath, P., Kataria, G., Sirish Kaushik, V., & Jude Hemanth, D. (2020). Pneumonia detection in chest X-ray images using convolutional neural networks and transfer learning. Measurement, 165, 108046. https://doi.org/10.1016/j.measurement.2020.108046
  7. Jakhar, K., & Hooda, N. (2018). Big Data Deep Learning Framework using Keras: A Case Study of Pneumonia Prediction. In 2018 4th International Conference on Computing Communication and Automation (ICCCA). IEEE. https://doi.org/10.1109/ccaa.2018.8777571
  8. Kabir, E., Siuly, & Zhang, Y. (2016). Epileptic seizure detection from EEG signals using logistic model trees. Brain Informatics, 3(2), 93–100. https://doi.org/10.1007/s40708-015-0030-2
  9. Kallianos, K., Mongan, J., Antani, S., Henry, T., Taylor, A., Abuya, J., & Kohli, M. (2019). How far have we come? Artificial intelligence for chest radiograph interpretation. Clinical Radiology, 74(5), 338– 345. https://doi.org/10.1016/j.crad.2018.12.015
  10. Kumar Acharya, A., & Satapathy, R. (2020). A Deep Learning Based Approach towards the Automatic Diagnosis of Pneumonia from Chest Radio-Graphs. Biomedical and Pharmacology Journal, 13(1), 449–455. https://doi.org/10.13005/bpj/1905 Mo