International Journal of Image, Graphics and Signal Processing (IJIGSP)
IJIGSP Vol. 17, No. 6, 8 Dec. 2025
Cover page and Table of Contents: PDF (size: 1371KB)
Semantic Segmentation of Tuberculosis Bacilli from Microscopic Sputum Smear Images Using TransUNet
PDF (1371KB), PP.19-37
Views: 0 Downloads: 0
Author(s)
Ashutosh Satapathy
1,*
Praneeth Vallabhaneni
1
Manisha Indugula
1
Index Terms
Semantic Segmentation, Tuberculosis Bacilli, Sputum Smear Images, Attention Mechanism, Transunet
Abstract
According to the World Health Organization (WHO) touchstones of 2022, Tuberculosis is the second dominant disease after COVID-19. Around one-fourth of the comprehensive population is ascertained to have tuberculosis. Timely detection and prevention of tuberculosis is a must to overcome its harmful effects. The method most often used in ascertaining whether a patient has tuberculosis, is examining his or her sputum sample. In the process, the isolation of the bacilli is done manually, and hence it is prone to error. Segmentation illustrates and enlightens objects or particles within an image, thus extracting the Region of Interest (ROI). The contemplated study uses TransUNet architecture to segment tuberculosis bacilli from sputum images to increase diagnostic accuracy and performance. The attention mechanism used in the TransUNet model helps to identify the spatial hierarchies present in image. It is an extremely tough task for naive or traditional segmentation algorithms to deal with the inherent complexity of sputum images. Hence, this study introduces an approach to capture the intrinsic features and dependencies needed to segment mycobacterium or TB bacilli by leveraging the TransUNet model. The model achieved an average Dice Score of 92.795%, a mean Intersection over Union (IoU) of 88.845%, and a segmentation accuracy of 99.19% on the Mosaic and Ziehl-Neelsen datasets. These results surpassed several existing state-of-art methods like UNet, clustering and thresholding, depicting the superior capability of TransUNet in segmenting the TB bacilli. It deepens the potentiality of transformer-based CNN models, especially TransUNet, for improving the diagnosis of tuberculosis and supporting disease management.
Cite This Paper
Ashutosh Satapathy, Praneeth Vallabhaneni, Manisha Indugula, "Semantic Segmentation of Tuberculosis Bacilli from Microscopic Sputum Smear Images Using TransUNet", International Journal of Image, Graphics and Signal Processing(IJIGSP), Vol.17, No.6, pp. 19-37, 2025. DOI:10.5815/ijigsp.2025.06.02
Reference
[1]Mejbel, F. A. H., Aljanaby, I. A. J., & Aljanaby, A. A. J. (2023). Pulmonary tuberculosis risks and challenges. In E3S Web of Conferences (Vol. 381, p. 01101). EDP Sciences.J. Clerk Maxwell, A Treatise on Electricity and Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.68–73.
[2]Chakaya, J., Khan, M., Ntoumi, F., Aklillu, E., Fatima, R., Mwaba, P., ... & Zumla, A. (2021). Global Tuberculosis Report (2020).–Reflections on the Global TB burden, treatment and prevention efforts. International Journal of Infectious Diseases, 113, S7-S12.
[3]Ismail, Z., Musa, L. H., Muslimah Ithnin, N. A. M., Baharom, N., Ramli, S., & Pathmanathan, S. G. (2022). A Systematic Review of Prevalence and Risk Factors of Latent Tuberculosis Infection Among Medical and Nursing Students. Malaysian Journal of Medicine and Health Sciences, 18(6), 270-279.
[4]Cesur, S. (2023). Clinical forms and diagnosis of tuberculosis. Journal of Pulmonology and Intensive Care, 1(1), 13-19.
[5]Manalan, K., Barrett, J., & Kon, O. M. (2021). Tuberculosis clinical presentation and differential diagnosis. Essential Tuberculosis, 79-85.
[6]Zhang, Y., Zhou, T., Wang, S., Liang, P., Zhang, Y., & Chen, D. Z. (2023, October). Input augmentation with sam: Boosting medical image segmentation with segmentation foundation model. In International Conference on Medical Image Computing and Computer-Assisted Intervention (pp. 129-139). Cham: Springer Nature Switzerland.
[7]Costa Filho, C. F., Levy, P. C., Xavier, C. M., Costa, M. G., Fujimoto, L. B., & Salem, J. (2012, August). Mycobacterium tuberculosis recognition with conventional microscopy. In: 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp 6263–6268. IEEE.
[8]Han, K., Wang, Y., Chen, H., Chen, X., Guo, J., Liu, Z., ... & Tao, D. (2022). A survey on vision transformer. IEEE transactions on pattern analysis and machine intelligence, 45(1), 87-110.
[9]Alzubaidi, L., Zhang, J., Humaidi, A. J., Al-Dujaili, A., Duan, Y., Al-Shamma, O., ... & Farhan, L. (2021). Review of deep learning: concepts, CNN architectures, challenges, applications, future directions. Journal of big Data, 8, 1-74.
[10]Chen, J., Lu, Y., Yu, Q., Luo, X., Adeli, E., Wang, Y., ... & Zhou, Y. (2021). Transunet: Transformers make strong encoders for medical image segmentation. arXiv preprint arXiv:2102.04306.
[11]Ali, O., Ali, H., Shah, S. A. A., & Shahzad, A. (2022). Implementation of a modified U-Net for medical image segmentation on edge devices. IEEE Transactions on Circuits and Systems II: Express Briefs, 69(11), 4593-4597.
[12]Wulandari, S. A., Mayasari, D. A., Kurniatie, M. D., & Tjahyono, R. (2021, September). Simplification of Mycobacterium tuberculosis segmenting algorithm in sputum images based on auto-thresholding. In: 2021 International Seminar on Application for Technology of Information and Communication (iSemantic), pp 12–16. IEEE.
[13]Mithra, K. S., & Emmanuel, W. S. (2021). GFNN: Gaussian-Fuzzy-Neural network for diagnosis of tuberculosis using sputum smear microscopic images. Journal of King Saud University-Computer and Information Sciences, 33(9), 1084-1095.
[14]Núñez-Fernández, D., Ballan, L., Jiménez-Avalos, G., Coronel, J., & Zimic, M. (2020). Automatic semantic segmentation for prediction of tuberculosis using lens-free microscopy images. arXiv preprint arXiv:2007.02482.
[15]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.
[16]Mithra, K. S., & Emmanuel, W. S. (2017, April). Segmentation and classification of mycobacterium from Ziehl Neelsen stained sputum images for tuberculosis diagnosis. In 2017 International Conference on Communication and Signal Processing (ICCSP) (pp. 1672-1676). IEEE.
[17]Shwetha, V., Prasad, K., Mukhopadhyay, C., Banerjee, B., & Chakrabarti, A. (2021, December). Automatic detection of bacilli bacteria from Ziehl-Neelsen sputum smear images. In 2021 2nd International Conference on Communication, Computing and Industry 4.0 (C2I4) (pp. 1-5). IEEE.
[18]Mithra, K. S., & Emmanuel, W. S. (2018, December). Segmentation of Mycobacterium tuberculosis bacterium from ZN stained microscopic sputum images. In 2018 International Conference on Smart Systems and Inventive Technology (ICSSIT) (pp. 150-154). IEEE.
[19]Ayas, S., & Ekinci, M. (2014). Automatic segmentation of Mycobacterium tuberculosis in Ziehl-Neelsen sputum slide images using support vector machines.
[20]Xavier, C. M. (2012). Segmentação, classificação e quantificação de bacilos de tuberculose em imagens de baciloscopia de campo claro através do emprego de uma nova técnica de classificação de pixels utilizando máquinas de vetores de suporte.
[21]Vanitha, V., Angayarkanni, S. P., & Sathishkumar, M. (2021, December). Segmentation of tuberculosis bacteria: Evaluation and comparison. In: 2021 International Conference on Computational Science and Computational Intelligence (CSCI), pp 1690–1696. IEEE.
[22]Serrão, M. K. M., Costa, M. G. F., Fujimoto, L. B. M., & Ogusku, M. M. (2024). Automatic Bright-field Smear Microscopy for Diagnosis of Pulmonary Tuberculosis. Computers in Medicine and Biology, accepted for publication Feb. 15, 2024.
[23]Misra, S., & Wu, Y. (2019). Machine learning assisted segmentation of scanning electron microscopy images of organic-rich shales with feature extraction and feature ranking. Machine learning for subsurface characterization, 289(3), 4.
[24]Singhal, P., Verma, A., & Garg, A. (2017, January). A study in finding effectiveness of Gaussian blur filter over bilateral filter in natural scenes for graph based image segmentation. In 2017 4th international conference on advanced computing and communication systems (ICACCS) (pp. 1-6). IEEE.
[25]MR, D. S. (2020). Comparison of Gaussian and Median Filters to Remove Noise in Dental Images. Institute of Scholars (InSc).
[26]C., & Khoshgoftaar, T. M. (2019). A survey on image data augmentation for deep learning. Journal of big data, 6(1), 1-48.