Improved Pyramid Histogram of Oriented Gradients and Statistical Features for Tuberculosis Detection via a Hybrid Classifier Using X-Ray Images
Received: 15 April 2025 | Revised: 5 June 2025, 16 June 2025, and 24 June 2025 | Accepted: 27 June 2025 | Online: 8 December 2025
Corresponding author: Rathod Dharmesh Ishwerlal
Abstract
Tuberculosis (TB) is a bacterial infection primarily affecting the lungs, which is diagnosed using methods such as skin tests, Purified Protein Derivative (PPD) test, blood tests, and Chest X-rays (CXRs). Traditionally, diagnostic test results were examined manually by doctors; however, in recent years, the advancements of Machine Learning (ML) have benefited the medical diagnosis process by offering an additional automated classification based on prognostic tests performed. In this study, a novel Hybrid LinkNet-Recurrent Neural Network (LinkNet-RNN) model is proposed for TB detection trained on CXR images. In the models' preprocessing stage, input X-ray images are enhanced using a median filter to reduce noise, while during feature extraction, an Improved Pyramid Histogram of Oriented Gradients (I-PHOG) method is employed to capture gradient information across multiple scales. In the final stage, the LinkNet-RNN model classifies each case as TB or not TB. The model's performance was evaluated through comprehensive experimental analysis, using the TB Chest X-ray Database as the train/test set, and compared against existing prediction models to validate its effectiveness, including standalone LinkNet, standalone RNN, SqueezeNet, LeNet, and Convolutional Neural Networks (CNN).
Keywords:
tuberculosis detection, LinkNet-Recurrent Neural Network, X-ray image, Improved Pyramid Histogram of Oriented Gradients (I-PHOG), median filterDownloads
References
E. Kotei and R. Thirunavukarasu, "Tuberculosis Detection From Chest X-Ray Image Modalities Based on Transformer and Convolutional Neural Network," IEEE Access, vol. 12, pp. 97417–97427, 2024. DOI: https://doi.org/10.1109/ACCESS.2024.3428446
T. Rahman et al., "Reliable Tuberculosis Detection Using Chest X-Ray With Deep Learning, Segmentation and Visualization," IEEE Access, vol. 8, pp. 191586–191601, 2020. DOI: https://doi.org/10.1109/ACCESS.2020.3031384
A. F. M. Saif, T. Imtiaz, C. Shahnaz, W.-P. Zhu, and M. O. Ahmad, "Exploiting Cascaded Ensemble of Features for the Detection of Tuberculosis Using Chest Radiographs," IEEE Access, vol. 9, pp. 112388–112399, 2021. DOI: https://doi.org/10.1109/ACCESS.2021.3102077
T. Xu and Z. Yuan, "Convolution Neural Network With Coordinate Attention for the Automatic Detection of Pulmonary Tuberculosis Images on Chest X-Rays," IEEE Access, vol. 10, pp. 86710–86717, 2022. DOI: https://doi.org/10.1109/ACCESS.2022.3199419
C.-F. Chen et al., "A deep learning-based algorithm for pulmonary tuberculosis detection in chest radiography," Scientific Reports, vol. 14, no. 1, Jun. 2024, Art. no. 14917. DOI: https://doi.org/10.1038/s41598-024-65703-z
S. Urooj, S. Suchitra, L. Krishnasamy, N. Sharma, and N. Pathak, "Stochastic Learning-Based Artificial Neural Network Model for an Automatic Tuberculosis Detection System Using Chest X-Ray Images," IEEE Access, vol. 10, pp. 103632–103643, 2022. DOI: https://doi.org/10.1109/ACCESS.2022.3208882
T. Rahman et al., "TB-CXRNet: Tuberculosis and Drug-Resistant Tuberculosis Detection Technique Using Chest X-ray Images," Cognitive Computation, vol. 16, no. 3, pp. 1393–1412, May 2024. DOI: https://doi.org/10.1007/s12559-024-10259-3
V. T. Q. Huy and C.-M. Lin, "An Improved Densenet Deep Neural Network Model for Tuberculosis Detection Using Chest X-Ray Images," IEEE Access, vol. 11, pp. 42839–42849, 2023. DOI: https://doi.org/10.1109/ACCESS.2023.3270774
Tuberculosis (TB) Chest X-ray Database. (2020), T. Rahman et al.. [Online]. Available: https://www.kaggle.com/datasets/tawsifurrahman/tuberculosis-tb-chest-xray-dataset.
A. Nithya, A. Appathurai, N. Venkatadri, D. R. Ramji, and C. Anna Palagan, "Kidney disease detection and segmentation using artificial neural network and multi-kernel k-means clustering for ultrasound images," Measurement, vol. 149, Jan. 2020, Art. no. 106952. DOI: https://doi.org/10.1016/j.measurement.2019.106952
A. Saidani and A. K. Echi, "Pyramid histogram of oriented gradient for machine-printed/handwritten and Arabic/Latin word discrimination," in 2014 6th International Conference of Soft Computing and Pattern Recognition (SoCPaR), Tunis, Tunisia, Aug. 2014, pp. 267–272. DOI: https://doi.org/10.1109/SOCPAR.2014.7008017
F. Bimantoro, A. A. Aziz, A. Y. Husodo, A. Musnansyah, A. E. Minarno, and A. Kurniawardhani, "Image Retrieval using Modified Multi Texton and Rotation Invariant Local Binary Pattern," in 2020 International Conference on Advancement in Data Science, E-learning and Information Systems (ICADEIS), Lombok, Indonesia, Oct. 2020, pp. 1–5. DOI: https://doi.org/10.1109/ICADEIS49811.2020.9277462
V. Ayzenberg and S. F. Lourenco, "Skeletal descriptions of shape provide unique perceptual information for object recognition," Scientific Reports, vol. 9, no. 1, Art. no. 9359, Jun. 2019. DOI: https://doi.org/10.1038/s41598-019-45268-y
B. Krishnakumar and K. Kousalya, "Optimal Trained Deep Learning Model for Breast Cancer Segmentation and Classification," Information Technology and Control, vol. 52, no. 4, pp. 915–934, Dec. 2023. DOI: https://doi.org/10.5755/j01.itc.52.4.34232
L. Zhang and X. Liang, "Image Segmentation of Plant Leaves in Natural Environments Based on LinkNet," Journal of Computing and Electronic Information Management, vol. 11, no. 3, pp. 67–72, Dec. 2023. DOI: https://doi.org/10.54097/jceim.v11i3.15
M. Islam, Z. Islam, A. Asraf, M. S. Al-Rakhami, W. Ding, and A. H. Sodhro, "Diagnosis of COVID-19 from X-rays using combined CNN-RNN architecture with transfer learning," BenchCouncil Transactions on Benchmarks, Standards and Evaluations, vol. 2, no. 4, Oct. 2022, Art. no. 100088. DOI: https://doi.org/10.1016/j.tbench.2023.100088
Downloads
How to Cite
License
Copyright (c) 2025 Rathod Dharmesh Ishwerlal, Reshu Agarwal, K. S. Sujatha
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
