RESNET101 VE GOOGLENET DERİN ÖĞRENME MODELLERİ: SAĞLIK SEKTÖRÜNDE BAŞARI DÜZEYLERİNİN KARŞILAŞTIRILMASI

Year 2024, , 390 - 409, 28.06.2024

Muhammed Akif Yenikaya

https://doi.org/10.36543/kauiibfd.2024.015

Abstract

Sağlık sektöründe yapay zekâ (YZ) uygulamaları, tıbbi teşhis ve tedavide önemli bir devrim niteliği taşımaktadır. Bu alandaki ilerlemeler, hastalıkların erken teşhis edilmesi ve sağlık hizmetlerinin verimliliğinin artırılması gibi birçok avantaj sağlamaktadır. Bu çalışmada, tüberküloz (TB) tespiti için derin öğrenme modellerinin kullanılabilirliğini araştırmak maksadıyla, ResNet101 ve GoogLeNet gibi derin öğrenme modellerinin sağlık sektöründe TB tespit potansiyeli bağlamında doğruluk oranları karşılaştırılmıştır. Yapılan analizlerden elde edilen bulgular, derin öğrenme ağlarının TB’li ve bu hastalığı bulundurmayan akciğer röntgen görüntüleri sınıflandırmasında başarılı olduğunu ortaya koymuştur. Ayrıca, başarı seviyeleri incelendiğinde ResNet101 derin öğrenme ağının %99.3 başarı oranı ile araştırmada ele alınan diğer derin öğrenme modeli olan GoogLeNet’e (%98.2) göre daha yüksek bir skor ortaya koyduğu tespit edilmiştir. Araştırma kapsamında elde edilen söz konusu bu bulgular, teşhis doğruluk oranlarının arttırılabilmesi için YZ uygulamalarının önem ve işlevselliğini ortaya koyar mahiyettedir.

Keywords

Sağlık Hizmetleri, Yapay Zekâ, Derin Öğrenme, Göğüs Röntgeni, Hastalık Tespiti

RESNET101 AND GOOGLENET DEEP LEARNING MODELS: COMPARING SUCCESS LEVELS IN THE HEALTH SECTOR

Abstract

Artificial intelligence (AI) applications in the healthcare sector have revolutionized medical diagnosis and treatment. Advances in this field provide many advantages such as early detection of diseases and increasing the efficiency of healthcare services. In this study, in order to investigate the usability of deep learning models for tuberculosis (TB) detection, the accuracy rates of deep learning models such as ResNet101 and GoogLeNet are compared in terms of TB detection potential in the healthcare sector. The results of the analyses revealed that deep learning networks are successful in classifying chest X-ray images with and without TB. In addition, when the success levels were analyzed, it was determined that the ResNet101 deep learning network, with a success rate of 99.3%, showed a higher score than the other deep learning model considered in the study, GoogLeNet (98.2%). These findings obtained within the scope of the research reveal the importance and functionality of AI applications in order to increase diagnostic accuracy rates.

Keywords

Healthcare, Artificial intelligence, deep learning, Chest X-ray, disease detection.

References

• Bar, Y., Diamant, I., Wolf, L., & Greenspan, H. (2015). Deep learning with non-medical training used for chest pathology identification. Paper presented at the Medical Imaging 2015: Computer-Aided Diagnosis.
• Bengio, Y. (2009). Learning deep architectures for AI. Foundations and Trends in Machine Learning, 2(1), 1–127.
• Codlin, A. J., Dao, T. P., Vo, L. N. Q., Forse, R. J., Van Truong, V., Dang, H. M., ... & Caws, M. (2021). Independent evaluation of 12 artificial intelligence solutions for the detection of tuberculosis. Scientific reports, 11(1), 23895.
• David, P. M., Onno, J., Keshavjee, S., & Khan, F. A. (2022). Conditions required for the artificial-intelligence-based computer-aided detection of tuberculosis to attain its global health potential. The Lancet Digital Health, 4(10), e702-e704.
• Deng, L., & Yu, D. (2014). Deep learning: Methods and applications. Foundations and Trends in Signal Processing, 7(3–4), 197–387.
• Faust, O., Hagiwara, Y., Hong, T. J., Lih, O. S., & Acharya, U. R. (2018). Deep learning for healthcare applications based on physiological signals: A review. Computer Methods and Programs in Biomedicine, 161, 1-13.
• Guo Z., Chen Q., Wu G., Xu Y., Shibasaki R., & Shao X. (2017). Village building identification based on ensemble convolutional neural networks. Sensors, 17(11), 2487.
• He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. Paper presented at the Proceedings of the IEEE conference on computer vision and pattern recognition.
• Hwang, S., Kim, H.-E., Jeong, J., & Kim, H.-J. (2016). A novel approach for tuberculosis screening based on deep convolutional neural networks. Paper presented at the Medical Imaging 2016: Computer-Aided Diagnosis.
• Jaeger, S., Karargyris, A., Candemir, S., Folio, L., Siegelman, J., Callaghan, F., Xue, Z., Palaniappan, K., Singh, R. K., & Antani, S. (2013). Automatic tuberculosis screening using chest radiographs. IEEE Transactions on Medical Imaging, 33(2), 233-245.
• Julia, D. L. F. (2016). "Mocha.jl: Deep Learning in Julia." Retrieved from https://devblogs.nvidia.com/parallelforall/mocha-jl-deep-learning-julia/
• Kaggle (2021). Accessed: January 25, 2024: https://www.kaggle.com/datasets/tawsifurrahman/tuberculosis-tb-chest-xray-dataset
• Khan, M. T., Kaushik, A. C., Ji, L., Malik, S. I., Ali, S., & Wei, D. Q. (2019). Artificial neural networks for prediction of tuberculosis disease. Frontiers in Microbiology, 10, 395.
• Khan, F. A., Majidulla, A., Tavaziva, G., Nazish, A., Abidi, S. K., Benedetti, A., ... & Saeed, S. (2020). Chest x-ray analysis with deep learning-based software as a triage test for pulmonary tuberculosis: a prospective study of diagnostic accuracy for culture-confirmed disease. The Lancet Digital Health, 2(11), e573-e581.
• Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). ImageNet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems (Vol. 25, pp. 1097–1105).
• Lakhani, P., & Sundaram, B. (2017). Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology, 284(2), 574-582.
• Lee, Y., & Nam, S. (2021). Performance comparisons of AlexNet and GoogLeNet in cell growth inhibition IC50 prediction. International Journal of Molecular Sciences, 22(14), 1-12.
• Lopes, U. K., & Valiati, J. F. (2017). Pre-trained convolutional neural networks as feature extractors for tuberculosis detection. Computational Biology and Medicine, 89, 135–143.
• Onno, J., Khan, F. A., Daftary, A., & David, P. M. (2023). Artificial intelligence-based computer aided detection (AI-CAD) in the fight against tuberculosis: Effects of moving health technologies in global health. Social Science & Medicine, 327, 115949.
• Panicker, R. O., Kalmady, K. S., Rajan, J., & Sabu, M. K. (2018). Automatic detection of tuberculosis bacilli from microscopic sputum smear images using deep learning methods. Biocybernetics and Biomedical Engineering, 38(3), 691-699.
• Reid, M. J., Arinaminpathy, N., Bloom, A., Bloom, B. R., Boehme, C., Chaisson, R., Chin, D. P., Churchyard, G., Cox, H., & Ditiu, L. (2019). Building a tuberculosis-free world: The Lancet Commission on tuberculosis. The Lancet, 393(10178), 1331-1384.
• Shen, D., Wu, G., & Suk, H.-I. (2017). Deep learning in medical image analysis. Annual Review of Biomedical Engineering, 19, 221-248.
• Song, H. A., & Lee, S.-Y. (2013). Hierarchical Representation Using NMF. In International Conference on Neural Information Processing (pp. 466–473).
• Targ, S., Almeida, D., & Lyman, K. (2016). Resnet in Resnet: Generalizing residual architectures. arXiv preprint arXiv:1603.08029.
• Williams, F. H. (1907). The use of X-ray examinations in pulmonary tuberculosis. Boston Medical and Surgical Journal, 157, 850–853.
• World Health Organization (2018). Global tuberculosis report 2018. Geneva: World Health Organization.
• Yenikaya, M. A., Kerse, G. (2022). A comparison of accuracy rates of alexnet and googlenet deep learning models in image classification. In Congress Book VII. International European Conference on Social Sciences, Antalya, Türkiye, pp 713-719
• Yenikaya, M. A., & Oktaysoy, O. (2023). The use of artificial intelligence applications in the health sector: Preliminary diagnosis with deep learning method. Sakarya University Graduate School of Business Journal, 5(2), 127-131.
• Yenikaya, M. A., Kerse, G., & Oktaysoy, O. (2024). Artificial intelligence in the healthcare sector: comparison of deep learning networks using chest X-ray images. Frontiers in Public Health, 12, 1386110.