Tuberculosis Diagnosis in X-Ray Images using CvT

• Carlos M. Santos Neto UFMA
• Anderson L. Silva UFMA
• Alexandre C. P. Pessoa UFMA
• Darlan B. P. Quintanilha UFMA
• João D. S. de Almeida UFMA
• Geraldo Braz Junior UFMA
• João O. B. Diniz IFMA

DOI: https://doi.org/10.5753/sbcas.2024.2224

Abstract

Tuberculosis (TB) remains one of the leading causes of death from infectious diseases. In 2022, an estimated 10.6 million people worldwide contracted TB. Chest X-rays, a non-invasive medical examination used to detect pathologies in various areas of the chest, are a crucial tool in TB diagnosis. Recent advancements in the field of computer vision, particularly with the application of deep learning techniques, have led to significant progress in the automated detection of abnormalities in chest X-rays. This has opened the door to machine-aided diagnosis. In this work, we propose a method for diagnosing tuberculosis in radiographic images using the Convolutional Vision Transformers neural network. The results show relevant metrics, with an accuracy of 93.13%, an F1-score of 92.68%, and an AUC-ROC of 97.16%, using the public image databases Shezen and Montgomery County. These results are superior to the state of the art.

References

Akiba, T., Sano, S., Yanase, T., Ohta, T., and Koyama, M. (2019). Optuna: A next-generation hyperparameter optimization framework. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.

Capellán-Martín, D., Gómez-Valverde, J. J., Bermejo-Peláez, D., and Ledesma-Carbayo, M. J. (2023). A lightweight, rapid and efficient deep convolutional network for chest x-ray tuberculosis detection. IEEE International Symposium on Biomedical Imaging.

Chollet, F. et al. (2015). Keras. [link].

Costa, G. S. S., Paiva, A. C., Junior, G. B., and Ferreira, M. M. (2021). Covid-19 automatic diagnosis with ct images using the novel transformer architecture. In Anais do XXI simpósio brasileiro de computação aplicada à saúde, pages 293–301. SBC.

da Silva, G. L., de Oliveira, F. Y., Diniz, J. O., Diniz, P. S., Quintanilha, D. B., Silva, A. C., de Paiva, A. C., and de Cavalcanti, E. A. (2021). An automatic method for prostate segmentation on 3d mri scans using local phylogenetic indexes and xgboost. In Anais do XXI Simpósio Brasileiro de Computação Aplicada à Saúde, pages 165–176. SBC.

Diniz, J. O., Quintanilha, D. B., de Carvalho Filho, A. O., Gomes Jr, D. L., Silva, A. C., Braz Jr, G., de Paiva, A. C., and Luz, D. d. S. (2023). Detecção de covid-19 em imagens de raio-x de tórax através de seleção automática de pré-processamento e de rede neural convolucional. In Anais do XXIII Simpósio Brasileiro de Computação Aplicada à Saúde, pages 162–173. SBC.

Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020). An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929.

Evangelista, L. G. C. and Guedes, E. B. (2019). Ensembles of convolutional neural networks on computer-aided pulmonary tuberculosis detection. IEEE Latin America Transactions, 17(12):1954–1963.

Islam, M. T., Aowal, M. A., Minhaz, A. T., and Ashraf, K. (2017). Abnormality detection and localization in chest x-rays using deep convolutional neural networks. arXiv preprint arXiv:1705.09850.

Jaeger, S., Candemir, S., Antani, S., Wang, Y.-X. J., Lu, P.-X., and Thoma, G. R. (2014). Two public chest x-ray datasets for computer-aided screening of pulmonary diseases. Quantitative imaging in medicine and surgery.

Lin, T.-Y., Goyal, P., Girshick, R., He, K., and Dollár, P. (2017). Focal loss for dense object detection. In Proceedings of the IEEE international conference on computer vision, pages 2980–2988.

Mahbub, M. K., Biswas, M., Gaur, L., Alenezi, F., Alenezi, F., and Santosh, K. C. (2022). Deep features to detect pulmonary abnormalities in chest x-rays due to infectious diseasex: Covid-19, pneumonia, and tuberculosis. Information Sciences.

Munzlinger, C., Yepes, I., and Rieder, R. (2023). Uso de uma rede neural convolucional para análise de exames de radiografia de pulmão com detecção de covid-19, pneumonia e tuberculose. In Anais Estendidos do XXIII Simpósio Brasileiro de Computação Aplicada à Saúde, pages 25–30, Porto Alegre, RS, Brasil. SBC.

Murphy, K., Habib, S. S., Zaidi, S. M. A., Khowaja, S., Khan, A., Melendez, J., Scholten, E. T., Amad, F., Schalekamp, S., Verhagen, M., et al. (2020). Computer aided detection of tuberculosis on chest radiographs: An evaluation of the cad4tb v6 system. Scientific reports, 10(1):5492.

Pande, T., Pai, M., Khan, F. A., and Denkinger, C. M. (2015). Use of chest radiography in the 22 highest tuberculosis burden countries. European Respiratory Journal.

Pasa, F., Golkov, V., Pfeiffer, F., Cremers, D., and Pfeiffer, D. (2019). Efficient deep network architectures for fast chest x-ray tuberculosis screening and visualization. Scientific reports, 9(1):6268.

Siracusano, G., La Corte, A., Gaeta, M., Cicero, G., Chiappini, M., and Finocchio, G. (2020). Pipeline for advanced contrast enhancement (pace) of chest x-ray in evaluating covid-19 patients by combining bidimensional empirical mode decomposition and contrast limited adaptive histogram equalization (clahe). Sustainability, 12(20):8573.

World Health Organization (2016). Chest radiography in tuberculosis detection: summary of current WHO recommendations and guidance on programmatic approaches. World Health Organization.

World Health Organization (2023). Global tuberculosis report.

Wu, H., Xiao, B., Codella, N., Liu, M., Dai, X., Yuan, L., Liu, Y., and Zhang, L. (2021). Cvt: Introducing convolutions to vision transformers.

Yang, F., Lu, P. X., Deng, M., Wáng, Y. X. J., Rajaraman, S., Xue, Z., Folio, L. R., Antani, S. K., and Jaeger, S. (2022). Annotations of lung abnormalities in the shenzhen chest x-ray dataset for computer-aided screening of pulmonary diseases. Data, 7(7).