A Data-Driven Multiscale Nonlinear Segmentation Framework for Silicosis Detection Using Machine Learning
DOI:
https://doi.org/10.24237/djes.2026.19310Keywords:
Silicosis detection, Nonlinear scale space, Multiscale image segmentation, Machine learning, Computer-aided diagnosis, Lung imagingAbstract
Early-stage silicosis is challenging to identify since lung alterations might appear as micronodules, distorted parenchymal patterns, or fibrotic changes. In this paper, we introduce a nonlinear multiscale image segmentation framework, with machine-learning classification, for computer-aided assessment of silicosis. The proposed method combines nonlinear total-variation (TV)-flow scale-space construction, hierarchical region merging, data-driven rgion pruning, and region-level machine learning classification. In contrast to traditional image-based classifiers, the approach generates interpretable lung subregions featuring micronodular and fibrotic lesions prior to classification. A nonlinear scale-space stack is constructed by using diffusion-based partial differential equations while maintaining lesion boundaries at different anatomical scales. The resulting structures are connected via a hierarchical region tree, and a data - driven pruning procedure is employed to prune unstable or non-anatomical regions within each path without any manual intervention. Intensity, shape, texture, and multiscale features are extracted from the retained regions and then used for supervised classification. The study consisted of 64 subjects and 286 HRCT slices. The segmentation-based method performed 9–11 % better in classification accuracy compared to the single-scale approaches or non-oriented features and achieved ~ 92 % sensitivity in detecting early stage silicosis. The framework also showed better edge preservation, less over segmentation, and practical computational efficiency. These results motivate the continued investigation of the approach as an interpretable computer-aided silicosis assessment framework.
Downloads
References
[1] R. F. Hoy, M. F. Jeebhay, C. Cavalin, W. Chen, R. A. Cohen, E. Fireman, L. H. T. Go, A. León-Jiménez, A. Menéndez-Navarro, M. Ribeiro, and P.-A. Rosental, “Current global perspectives on silicosis—Convergence of old and newly emergent hazards,” Respirology, vol. 27, no. 6, pp. 387–398, 2022, doi: 10.1111/resp.14242.
[2] A. León-Jiménez, J. M. Mánuel, M. García-Rojo, M. G. Pintado-Herrera, J. A. López-López, A. Hidalgo-Molina, R. García, P. Muriel-Cueto, N. Maira-González, D. Del Castillo-Otero, and F. M. Morales, “Compositional and structural analysis of engineered stones and inorganic particles in silicotic nodules of exposed workers,” Part. Fibre Toxicol., vol. 18, Art. no. 41, 2021, doi: 10.1186/s12989-021-00434-x.
[3] A. León-Jiménez, A. Hidalgo-Molina, M. Á. Conde-Sánchez, A. Pérez-Alonso, J. M. Morales-Morales, E. M. García-Gámez, and J. A. Córdoba-Doña, “Artificial stone silicosis: Rapid progression following exposure cessation,” Chest, vol. 158, no. 3, pp. 1060–1068, 2020, doi: 10.1016/j.chest.2020.03.026.
[4] A. García-Núñez, G. Jiménez-Gómez, A. Hidalgo-Molina, J. A. Córdoba-Doña, A. León-Jiménez, and A. Campos-Caro, “Inflammatory indices obtained from routine blood tests show an inflammatory state associated with disease progression in engineered stone silicosis patients,” Sci. Rep., vol. 12, no. 1, Art. no. 8211, 2022, doi: 10.1038/s41598-022-11926-x.
[5] W. Sun, D. Wu, Y. Luo, L. Liu, H. Zhang, S. Wu, Y. Zhang, C. Wang, H. Zheng, J. Shen, and C. Luo, “A fully deep learning paradigm for pneumoconiosis staging on chest radiographs,” IEEE J. Biomed. Health Inform., vol. 26, no. 10, pp. 5154–5164, 2022, doi: 10.1109/JBHI.2022.3190923.
[6] L. Zhang, R. Rong, Q. Li, D. M. Yang, B. Yao, D. Luo, X. Zhang, X. Zhu, J. Luo, Y. Liu, X. Yang, X. Ji, Z. Liu, Y. Xie, Y. Sha, Z. Li, and G. Xiao, “A deep learning-based model for screening and staging pneumoconiosis,” Sci. Rep., vol. 11, no. 1, Art. no. 2201, 2021, doi: 10.1038/s41598-020-77924-z.
[7] F. Yang, Z.-R. Tang, J. Chen, M. Tang, S. Wang, W. Qi, C. Yao, Y. Yu, Y. Guo, and Z. Yu, “Pneumoconiosis computer aided diagnosis system based on X-rays and deep learning,” BMC Med. Imaging, vol. 21, no. 1, Art. no. 189, 2021, doi: 10.1186/s12880-021-00723-z.
[8] F. Mushtaq, S. Bhattacharjee, S. Mandia, K. Singh, S. S. Chouhan, R. Kumar, and P. Harjule, “Artificial intelligence for computer aided detection of pneumoconiosis: A succinct review since 1974,” Eng. Appl. Artif. Intell., vol. 133, Art. no. 108516, 2024, doi: 10.1016/j.engappai.2024.108516.
[9] M. Mirbabaie, S. Stieglitz, and N. R. J. Frick, “Artificial intelligence in disease diagnostics: A critical review and classification on the current state of research guiding future direction,” Health Technol., vol. 11, no. 4, pp. 693–731, 2021, doi: 10.1007/s12553-021-00555-5.
[10] L. Devnath, S. Luo, P. Summons, and D. Wang, “Automated detection of pneumoconiosis with multilevel deep features learned from chest X-ray radiographs,” Comput. Biol. Med., vol. 129, Art. no. 104125, 2021, doi: 10.1016/j.compbiomed.2020.104125.
[11] D. Sanchez-Morillo, A. León-Jiménez, M. Guerrero-Chanivet, G. Jiménez-Gómez, A. Hidalgo-Molina, and A. Campos-Caro, “Integrating routine blood biomarkers and artificial intelligence for supporting diagnosis of silicosis in engineered stone workers,” Bioeng. Transl. Med., vol. 9, no. 6, Art. no. e10694, 2024, doi: 10.1002/btm2.10694.
[12] X. Xue, D. Ji, C. Xu, Y. Zhao, Y. Li, and C. Hu, “Adaptive orthogonal directional total variation with kernel regression for CT image denoising,” J. X-Ray Sci. Technol., vol. 32, no. 5, pp. 1253–1271, 2024, doi: 10.3233/XST-230416.
[13] Z. Mustafa and H. Nsour, “Using computer vision techniques to automatically detect abnormalities in chest X-rays,” Diagnostics, vol. 13, no. 18, Art. no. 2979, 2023, doi: 10.3390/diagnostics13182979.
[14] L. Devnath, S. Luo, P. Summons, D. Wang, K. Shaukat, I. A. Hameed, and F. S. Alrayes, “Deep ensemble learning for the automatic detection of pneumoconiosis in coal worker’s chest X-ray radiography,” J. Clin. Med., vol. 11, no. 18, Art. no. 5342, 2022, doi: 10.3390/jcm11185342.
[15] A. Davri, E. Birbas, T. Kanavos, G. Ntritsos, N. Giannakeas, A. T. Tzallas, and A. Batistatou, “Deep learning for lung cancer diagnosis, prognosis and prediction using histological and cytological images: A systematic review,” Cancers, vol. 15, no. 15, Art. no. 3981, 2023, doi: 10.3390/cancers15153981.
[16] M. A. Balcı, L. M. Batrancea, Ö. Akgüller, and A. Nichita, “A series-based deep learning approach to lung nodule image classification,” Cancers, vol. 15, no. 3, Art. no. 843, 2023, doi: 10.3390/cancers15030843.
[17] Y. H. Bhosale and K. S. Patnaik, “Application of deep learning techniques in diagnosis of COVID-19 (Coronavirus): A systematic review,” Neural Process. Lett., vol. 55, pp. 3551–3603, 2023, doi: 10.1007/s11063-022-11023-0.
[18] J. Yi, K. Tapia, J. W. Robinson, Z. Gandomkar, M. E. Suleiman, P. C. Brennan, N. Sommerfeld, and S. T. Taba, “Radiologists’ performance in diagnosing silicosis on high-resolution computed tomography (HRCT) scans: An online platform,” in Medical Imaging 2024: Image Perception, Observer Performance, and Technology Assessment, vol. 12929. Bellingham, WA, USA: SPIE, 2024, pp. 32–38, doi: 10.1117/12.3006114.
[19] G. K. Sharma, P. Harjule, B. Agarwal, and R. Kumar, “Silicosis detection using extended transfer learning model,” in Recent Trends in Image Processing and Pattern Recognition. Cham, Switzerland: Springer, 2024, pp. 111–126, doi: 10.1007/978-3-031-53085-2_10.
[20] Y. Zhang, B. Zheng, F. Zeng, X. Cheng, T. Wu, Y. Peng, Y. Zhang, Y. Xie, W. Yi, W. Chen, J. Wu, and L. Li, “Potential of digital chest radiography-based deep learning in screening and diagnosing pneumoconiosis: An observational study,” Medicine, vol. 103, no. 25, Art. no. e38478, 2024, doi: 10.1097/MD.0000000000038478.
[21] P. G. Anderson, H. Tarder-Stoll, M. Alpaslan, N. Keathley, D. L. Levin, S. Venkatesh, E. Bartel, S. Sicular, S. Howell, R. V. Lindsey, and R. M. Jones, “Deep learning improves physician accuracy in the comprehensive detection of abnormalities on chest X-rays,” Sci. Rep., vol. 14, Art. no. 25151, 2024, doi: 10.1038/s41598-024-76608-2.
[22] C. Liu, Y. Fang, Y. Xie, H. Zheng, X. Li, D. Wu, and T. Zhang, “Deep learning pneumoconiosis staging and diagnosis system based on multi-stage joint approach,” BMC Med. Imaging, vol. 24, no. 1, Art. no. 165, 2024, doi: 10.1186/s12880-024-01337-x.
[23] B. Priego-Torres, D. Sanchez-Morillo, E. Khalili, M. Á. Conde-Sánchez, A. García-Gámez, and A. León-Jiménez, “Automated engineered-stone silicosis screening and staging using deep learning with X-rays,” Comput. Biol. Med., vol. 191, Art. no. 110153, 2025, doi: 10.1016/j.compbiomed.2025.110153.
[24] X. Li, M. Xu, Z. Yan, F. Xia, S. Li, Y. Zhang, Z. Xing, and L. Guan, “Deep convolutional network-based chest radiographs screening model for pneumoconiosis,” Front. Med., vol. 11, Art. no. 1290729, 2024, doi: 10.3389/fmed.2024.1290729.
[25] S. Hao, X. Li, W. Peng, Z. Fan, Z. Ji, and I. Ganchev, “YOLO-CXR: A novel detection network for locating multiple small lesions in chest X-ray images,” IEEE Access, vol. 12, pp. 156003–156019, 2024, doi: 10.1109/ACCESS.2024.3482102.
[26] Z. Wang, J. Zhang, T. Wang, Z. Liu, W. Zhang, Y. Sun, X. Wu, H. Shao, and Z. Du, “The value of single biomarkers in the diagnosis of silicosis: A meta-analysis,” iScience, vol. 27, no. 6, Art. no. 109948, 2024, doi: 10.1016/j.isci.2024.109948.
[27] W. Sun, D. Wu, Y. Luo, L. Liu, H. Zhang, S. Wu, Y. Zhang, C. Wang, H. Zheng, J. Shen, and C. Luo, “ExpertNet: Defeat noisy labels by deep expert consultation paradigm for pneumoconiosis staging on chest radiographs,” Expert Syst. Appl., vol. 232, Art. no. 120710, 2023, doi: 10.1016/j.eswa.2023.120710.
[28] L. Xiong, X. Liu, X. Qin, and W. Li, “Accurate pneumoconiosis staging via deep texture encoding and discriminative representation learning,” Front. Med., vol. 11, Art. no. 1440585, 2024, doi: 10.3389/fmed.2024.1440585.
[29] B. Priego-Torres, I. Sopo-Lambea, E. Khalili, A. Martín-Carrillo, A. Campos-Caro, A. León-Jiménez, and D. Sanchez-Morillo, “Multimodal fusion of chest X-rays and blood biomarkers for automated silicosis staging,” J. Clin. Med., vol. 14, no. 22, Art. no. 8074, 2025, doi: 10.3390/jcm14228074.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 N. Shivaanivarsha, P. Kavipriya
This work is licensed under a Creative Commons Attribution 4.0 International License.
