G.M.Karthik; D. Lakshmi Padmaja; Nageswara Rao Medikondu; C. Sateesh Kumar Reddy; Pichili Vijaya Bhaskar Reddy; Anoop V.

Efficient ResNet-based Deep Learning Model for Asthma and COPD Detection Using Short-Time Fourier Transform Spectrograms

PDF (1047KB), PP.1-15

Views: 0 Downloads: 0

Author(s)

G.M.Karthik ^1,* D. Lakshmi Padmaja ² Nageswara Rao Medikondu ³ C. Sateesh Kumar Reddy ⁴ Pichili Vijaya Bhaskar Reddy ⁵ Anoop V. ⁶

1. Department of Information Security, School of Computer science and engineering (SCOPE), Vellore Institute of technology, Vellore, TN, 632014, India

2. Department of Information Technology , Anurag University School of Engineering , Hyderabad , 500088, India

3. Department Of Mechanical Engineering, Koneru Lakshmaiah Foundation, Vaddeswaram, Guntur, 522302, India

4. Department of Advanced Computer Science and Engineering, Vignan's Foundation for Science, Technology and Research, Guntur, 522213, India

5. Department of Life Science and Bioinformatics, Assam University Diphu Campus, Diphu, 782462, India

6. Department of Artificial Intelligence and Data Science, Jyothi Engineering College, Thrissur, 679531, India

* Corresponding author.

DOI: https://doi.org/10.5815/ijigsp.2026.05.01

Received: 23 Jan. 2026 / Revised: 17 Feb. 2026 / Accepted: 23 Mar. 2026 / Published: 8 Oct. 2026

Index Terms

Chronic Obstructive Pulmonary Disease, Wiener Filtering, Silence Trimming, Spectrograms, EfficientNet.

Abstract

This study proposes a deep learning-based framework for the automated detection of asthma and chronic obstructive pulmonary disease (COPD) using respiratory sound analysis. Breath sound recordings are preprocessed through resampling, silence trimming, and Wiener filtering to enhance signal quality. Short-Time Fourier Transform (STFT) is employed to convert audio signals into spectrogram representations, which are further augmented using time masking, frequency masking, and time warping to improve model generalization. The proposed model utilizes EfficientNet with multi-scale feature fusion to capture both local and global patterns in respiratory sounds. Experimental results demonstrate that the proposed approach achieves superior performance, with an accuracy of 99.21%, sensitivity of 99.56%, and specificity of 100%, outperforming existing CNN, ResNet, and SVM-based methods. The findings indicate that the proposed method is a reliable and efficient tool for non-invasive respiratory disease diagnosis.

Cite This Paper

G. M. Karthik, D. Lakshmi Padmaja, Nageswara Rao Medikondu, C. Sateesh Kumar Reddy, Pichili Vijaya Bhaskar Reddy, Anoop V., "Efficient ResNet-based Deep Learning Model for Asthma and COPD Detection Using Short-Time Fourier Transform Spectrograms", International Journal of Image, Graphics and Signal Processing(IJIGSP), Vol.18, No.5, pp. 1-15, 2026. DOI:10.5815/ijigsp.2026.05.01

Reference

[1]A. Nair, H. Vadher, P. Patel, T. Vyas, C. Bhatt, and A. Bruno, “Lung sound disease detection using attention over pre-trained EfficientNet architecture,” Multimedia Tools and Applications, pp. 1–35, 2024. doi:10.1007/s11042-024-20078-1
[2]M. Habib, V. Vicente-Palacios, and P. García-Sánchez, “Bio-inspired optimization of feature selection and SVM tuning for voice disorders detection” Knowledge-Based Systems, vol. 310, pp.112950, 2025. doi:10.1016/j.knosys.2024.112950
[3]N. Vodnala, P. S. Yarlagadda, M. Ch, and K. Sailaja, “Novel deep learning approaches to differentiate asthma and COPD based on cough sounds” in Proceedings of the 2024 Parul International Conference on Engineering and Technology (PICET), IEEE, pp. 1–4, 2024. doi: 10.1109/PICET60765.2024.10716083.
[4]P. Sahu, S. Kumar, and A. K. Behera, “DeepFusion: Early diagnosis of COPD, asthma, and pneumonia using lung sound analysis with a multimodal BiGRU network,” Computer Methods in Biomechanics and Biomedical Engineering, pp. 1–14, 2025. doi:10.1080/10255842.2025.2511228.
[5]P. Sahu, S. Jha, and S. Kumar, “Optimized 1D CNNs for enhanced early detection and accurate prediction of COPD and other pulmonary diseases,” in Proc. IEEE Region 10 Symp. (TENSYMP), 2024, pp. 1–6. doi: 10.1109/TENSYMP61132.2024.10752156.
[6]P. Sahu, P. Prasad, V. P. Verma, and S. Kumar, “Deep learning framework for early diagnosis of COPD and respiratory diseases using lung sound analysis,” in Proc. Int. Conf. Big Data Analytics, Cham: Springer Nature Switzerland, 2024, pp. 295–304. doi: 10.1007/978-3-031-81821-9_17.
[7]H. Chen, X. Yuan, Z. Pei, M. Li, and J. Li, “Triple-classification of respiratory sounds using optimized S-transform and deep residual networks,” IEEE Access, vol. 7, pp. 32845–32852, 2019. doi: 10.1109/ACCESS.2019.2903859.
[8]M. A. Iqbal, K. Devarajan, and S. M. Ahmed, “An optimal asthma disease detection technique for voice signal using hybrid machine learning technique,” Concurrency Comput.: Pract. Exp., vol. 34, no. 11, p. e6856, 2022. doi: 10.1002/cpe.6856.
[9]N. Zakaria, M. Fezari, A. Redjati, and K. Sundaraj, “Three ResNet deep learning architectures applied in pulmonary pathologies classification,” in Proc. Int. Conf. AI for Cyber Security Systems and Privacy (AI-CSP), 2021, pp. 1–8.
[10]R. Karla and R. Yalavarthi, “A hybrid RNN-based deep learning model for lung cancer and COPD detection,” Eng., Technol. Appl. Sci. Res., vol. 14, no. 5, pp. 16847–16853, 2024.doi: 10.48084/etasr.8181.
[11]V. Koshta, B. K. Singh, A. K. Behera, and T. G. Ranganath, “Fourier decomposition-based automated classification of healthy, COPD, and asthma using single-channel lung sounds,” IEEE Trans. Med. Robot. Bionics, vol. 6, no. 3, pp. 1270–1284, 2024. doi: 10.1109/TMRB.2024.3408325.
[12]A. Roy, B. Gyanchandani, A. Oza, and A. Singh, “TriSpectraKAN: A novel approach for COPD detection via lung sound analysis,” Scientific Reports, vol. 15, no. 1, p. 6296, 2025. doi: 10.1038/s41598-024-82781-1
[13]X. Zhang, J. Fu, W. Wang, and L. Yu, “Diagnosis of chronic obstructive pulmonary disease based on deep learning and auscultation lung sound,” Biomedical Signal Processing and Control, vol. 112, p. 108438, 2026, doi: 10.1016/j.bspc.2025.108438.
[14]S. D. Aaron et al., “Early diagnosis and treatment of COPD and asthma—a randomized, controlled trial,” N. Engl. J. Med., vol. 390, no. 22, pp. 2061–2073, 2024. doi: 10.1056/NEJMoa2401389
[15]Y. Yan, L. van Bemmel, F. M. E. Franssen, S. O. Simons, and V. Urovi, “Developing a multi-feature fusion model for exacerbation classification in asthma and COPD,” Comput. Methods Programs Biomed., vol. 268, p. 108796, 2025.doi: 10.1016/j.cmpb.2025.108796.
[16]G. Shanthakumari, S. Gayathri, M. Parvathababu, J. Raamkumar, and E. Prajin, “Lung disease diagnosis through Mel spectrogram and acoustic feature analysis,” in Proc. Int. Conf. Computing and Communication Technologies (ICCCT), 2025, pp. 1–8. doi: 10.1109/ICCCT63501.2025.11019006.
[17]S. Xu, R. C. Deo, O. Faust, P. D. Barua, J. Soar, and R. Acharya, “Automated lightweight model for asthma detection using respiratory and cough sound signals,” Diagnostics, vol. 15, no. 9, p. 1155, 2025. doi:10.3390/diagnostics15091155
[18]P. Sahu, S. Kumar, and A. K. Behera, “SoundNet: Leveraging deep learning for the severity classification of chronic obstructive pulmonary disease based on lung sound analysis,” in Proc. IEEE Int. Conf. Electronics, Computing and Communication Technologies (CONECCT), 2024, pp. 1–6. doi: 10.1109/CONECCT62155.2024.10677193.
[19]K.-R. Lee et al., “A wearable stethoscope for accurate real-time lung sound monitoring and automatic wheezing detection based on an AI algorithm,” Engineering, 2025. doi: 10.1016/j.eng.2024.12.031
[20]M. M. Ramyasri, M. Yoga, C. Harish, M. M. Rizwan, and K. Monisha, “Utilizing neural networks to analyze respiratory sounds for the detection of lung diseases,” in Proc. Int. Conf. Advances in Computing and Communications (ICACC), 2024, pp. 1–6. doi: 10.1109/ICACC63692.2024.10845461.
[21]B. Zhang, Y. Gu, D. Fu, L. Zhou, X. Wang, L. Guo, and J. Wang, “MSMF-DBNet: Dual-branch model for pulmonary disease recognition with multi-scale feature enhancement and multichannel fusion,” Applied Soft Computing, p. 114847, 2026, doi: 10.1016/j.asoc.2026.114847.
[22]W. N. A. Ismail, N. Ghazali, and R. Ngadiran, “Development of a robust deep learning network to distinguish COPD and asthma based on the augmentation of imbalanced respiratory sounds,” in AIP Conf. Proc., vol. 3338, no. 1, p. 020009, 2025. doi: 10.1063/5.0295943.
[23]M. L. Bennett, “Automated identification of respiratory anomalies from cough acoustics using spectrogram-driven deep learning,” Frontiers Emerg. AI Mach. Learn., vol. 2, no. 1, pp. 8–13, 2025.
[24]T. Albiges, “Artificial intelligence and signal analysis for COPD classification: Detecting and understanding of respiratory disease severities,” Ph.D. dissertation, Bournemouth Univ., 2025.

International Journal of Image, Graphics and Signal Processing (IJIGSP)