SE-ResNet18: Attention-Enhanced Deep Learning for Multi-Class Classification of Tuberculosis and Pneumonia from Chest X-ray Images

Öz

Tuberculosis and pneumonia are major causes of respiratory mortality worldwide, requiring accurate and timely diagnosis. This study proposes SE-ResNet18, an attention-enhanced deep learning model for multi-class classification of chest X-ray images into Normal, Pneumonia, Tuberculosis, and Unknown categories. The model integrates Squeeze-and-Excitation (SE) blocks into the ResNet18 architecture to improve channel-wise feature representation. A dataset of 15,316 chest radiographs was used, split into training (13,028), validation (761), and testing (1,527) sets. Transfer learning was applied using ImageNet-pretrained weights, followed by fine-tuning for 10 epochs with the Adam optimizer (learning rate: 1×10⁻⁵). To enhance generalization, limited data augmentation (horizontal flipping and ±5° rotation) was applied only to the training set. Dropout (p = 0.4) was used in the classification head to reduce overfitting. The proposed model achieved 98.03% accuracy and a macro F1-score of 0.97 on the test set, indicating balanced performance across classes. Class-wise results were: Unknown (1.00 precision, recall, F1-score), Tuberculosis (0.95 precision, 0.99 recall, 0.97 F1-score), Pneumonia (0.98 precision, 0.96 recall, 0.97 F1-score), and Normal (0.96 precision, 0.95 recall, 0.95 F1-score). No misclassification occurred between Pneumonia and Tuberculosis. Confidence analysis showed well-calibrated predictions, with higher confidence for correct predictions (0.947) than errors (0.823), enabling identification of uncertain cases for expert review.

Anahtar Kelimeler

• Deep learning
• Attention mechanisms
• ResNet18
• Transfer learning
• Medical image analysis

SE-ResNet18: Göğüs Röntgeni Görüntülerinden Tüberküloz ve Pnömoninin Çok Sınıflı Sınıflandırılması için Dikkat Mekanizması ile Güçlendirilmiş Derin Öğrenme Yaklaşımı

Öz

Tüberküloz ve pnömoni, dünya genelinde solunum kaynaklı ölümlerin başlıca nedenleri arasında yer almakta olup doğru ve hızlı tanı gerektirmektedir. Bu çalışmada, göğüs röntgeni görüntülerini Normal, Pnömoni, Tüberküloz ve Bilinmeyen olarak sınıflandırmak için SE-ResNet18 adı verilen dikkat mekanizması ile güçlendirilmiş bir derin öğrenme modeli önerilmiştir. Model, kanal bazlı özellik temsillerini iyileştirmek amacıyla ResNet18 mimarisine Squeeze-and-Excitation blokları entegre edilerek geliştirilmiştir. Toplam 15.316 görüntü kullanılmış ve veri seti eğitim (13.028), doğrulama (761) ve test (1.527) olarak ayrılmıştır. Transfer öğrenme yaklaşımı kapsamında ImageNet ön-eğitimli ağırlıklar kullanılmış ve model, Adam optimizasyon algoritması (öğrenme oranı: 1×10⁻⁵) ile 10 epoch boyunca eğitilmiştir. Genelleme performansını artırmak için yalnızca eğitim verisinde sınırlı veri artırma teknikleri (yatay çevirme ve ±5° döndürme) uygulanmıştır. Aşırı öğrenmeyi önlemek amacıyla sınıflandırma katmanında dropout (p = 0.4) kullanılmıştır. Model, test kümesinde %98.03 doğruluk ve 0.97 makro F1-skoru elde etmiştir. Pnömoni ile Tüberküloz arasında yanlış sınıflandırma gözlenmemiştir. Güven analizi sonuçları, doğru tahminlerde daha yüksek güven değeri (0.947) olduğunu göstermiştir.

Anahtar Kelimeler

• Derin öğrenme
• Dikkat mekanizmaları
• ResNet18
• Transfer öğrenme
• Tıbbi görüntü analizi

Kaynakça

1. Bagcchi S. WHO's global tuberculosis report 2022. The Lancet Microbe. 2023;4(1): e20.
2. Iqbal A, Usman M, Ahmed Z. An efficient deep learning-based framework for tuberculosis detection using chest X-ray images. Tuberculosis. 2022;136, 102234.
3. Jaeger S, Candemir S, Antani S, Wáng Y X J, Lu P X, Thoma G. Two public chest X-ray datasets for computer-aided screening of pulmonary diseases. Quantitative imaging in medicine and surgery. 2014:4(6), 475.
4. Rajaraman S, Antani S K. Modality-specific deep learning model ensembles toward improving TB detection in chest radiographs. IEEE Access. 2020:8, 27318-27326.
5. Hooda R, Mittal A, Sofat S. Automated TB classification using ensemble of deep architectures. Multimedia Tools and Applications. 2019:78(22), 31515-31532.
6. Rajpurkar P, Irvin J, Zhu K, Yang B, Mehta H, Duan T, Ng AY. Chexnet: Radiologist-level pneumonia detection on chest x-rays with deep learning. arXiv preprint. 2017:arXiv:1711.05225.
7. Capellán-Martín D, Gómez-Valverde J J, Bermejo-Peláez D, Ledesma-Carbayo M J. A lightweight, rapid and efficient deep convolutional network for chest x-ray tuberculosis detection. In 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI); 2023. p. 1-5.
8. Tan M, Le Q. Efficientnet: Rethinking model scaling for convolutional neural networks. In International conference on machine learning; 2019. p. 6105-6114.

9. Hu J, Shen L, Sun G. Squeeze-and-excitation networks. In Proceedings of the IEEE conference on computer vision and pattern recognition; 2018. p. 7132-7141.
10. Wong A, Lee J R H, Rahmat-Khah H, Sabri A, Alaref A, Liu H. TB-Net: a tailored, self-attention deep convolutional neural network design for detection of tuberculosis cases from chest X-ray images. Frontiers in Artificial Intelligence. 2022:5, 827299.
11. Mirugwe A, Tamale L, Nyirenda J. Improving tuberculosis detection in chest X-ray images through transfer learning and deep learning: comparative study of convolutional neural network architectures. JMIRx Med. 2025:6, e66029.
12. Showkatian E, Salehi M, Ghaffari H, Reiazi R, Sadighi N. Deep learning-based automatic detection of tuberculosis disease in chest X-ray images. Polish journal of radiology. 2022:87(1), 118-124.
13. Rahman K M, Zulaikha S, Dhafer B, Ahmed R. Advancing tuberculosis screening: A tailored CNN approach for accurate chest X-ray analysis and practical clinical integration. Intelligence-Based Medicine. 2025:11, 100196.
14. Maheswari B U, Sam D, Mittal N, Sharma A, Kaur S, Askar S S, Abouhawwash M. Explainable deep-neural-network supported scheme for tuberculosis detection from chest radiographs. BMC Medical Imaging. 2024:24(1), 32.
15. Salkade S A, Rathi S V. An adaptive convolution neural network model for tuberculosis detection and diagnosis using semantic segmentation. Polish Journal of Radiology. 2025:90, e124.
16. Sarawagi K, Pagrotra A, Dhiman H, Singh N. Self-trained convolutional neural network (CNN) for tuberculosis diagnosis in medical imaging. Cureus. 2024:16(6).
17. Vanitha K, Mahesh T R, Kumar V V, Guluwadi S. Enhanced tuberculosis detection using Vision Transformers and explainable AI with a Grad-CAM approach on chest X-rays. BMC Medical Imaging. 2025:25(1), 96.
18. Iqbal A, Usman M, Ahmed Z. Tuberculosis chest X-ray detection using CNN-based hybrid segmentation and classification approach. Biomedical Signal Processing and Control. 2023:84, 104667.
19. Kotei E, Thirunavukarasu R. Ensemble technique coupled with deep transfer learning framework for automatic detection of tuberculosis from chest X-ray radiographs. Healthcare. 2022:10(11):2335.
20. Hansun S, Argha A, Liaw S T, Celler B G, Marks G B. Machine and deep learning for tuberculosis detection on chest X-rays: Systematic literature review. Journal of medical Internet research. 2023:25, e43154.
21. Ayaz M, Shaukat F, Raja G. Ensemble learning based automatic detection of tuberculosis in chest X-ray images using hybrid feature descriptors. Physical and Engineering Sciences in Medicine. 2021:44(1):183-194.
22. Sharma V, Nillmani Gupta S K, Shukla K K. Deep learning models for tuberculosis detection and infected region visualization in chest X-ray images. Intelligent Medicine. 2024:4(02):104-113.
23. Rajpurkar P, O’Connell C, Schechter A, Asnani N, Li J, Kiani A, Lungren M P. CheXaid: deep learning assistance for physician diagnosis of tuberculosis using chest x-rays in patients with HIV. NPJ digital medicine. 2020:3(1), 115.
24. Ifty T T, Shafin S A, Shahriar S M, Towhid T. Explainable lung disease classification from chest x-ray images utilizing deep learning and xai. In 2024 IEEE 3rd International Conference on Computing and Machine Intelligence (ICMI); 2024. p. 1-5.
25. de Camargo T F O, Ribeiro G A S, da Silva M C B, da Silva L O, Torres P P T E S, Rodrigues D D S D S, de Paiva J P Q. Clinical validation of an artificial intelligence algorithm for classifying tuberculosis and pulmonary findings in chest radiographs. Frontiers in Artificial Intelligence. 2025:8, 1512910.
26. Chaykam V R, Koduru A R, Arun C. Enhanced Tuberculosis detection using deep learning and GAN data augmentation techniques for Chest X-ray analysis. In 2025 3rd International Conference on Intelligent Data Communication Technologies and Internet of Things (IDCIoT); 2025. p. 1171-1178.
27. Santosh K C, Allu S, Rajaraman S, Antani S. Advances in deep learning for tuberculosis screening using chest X-rays: the last 5 years review. Journal of Medical Systems. 2022:46(11), 82.
28. Hansun S, Argha A, Bakhshayeshi I, Wicaksana A, Alinejad-Rokny H, Fox G J, Marks G B. Diagnostic performance of artificial intelligence–based methods for tuberculosis detection: Systematic review. Journal of medical Internet research. 2025: 27, e69068.
29. Riasat K, Jamil A, Al-Otaibi S, Zeb S, Riasat S, Kanwal S. Tuberculosis detection using few shot learning. Scientific Reports. 2025:15(1), 13167.
30. Ding Z, Yaakob R, Azman A, Rum S N M, Zakaria N, Nazri A S A. Artificial intelligence-powered tuberculosis detection with complementary domain attention model. Neurocomputing. 2025:637, 130089.
31. Parveen Rahamathulla M, Sam Emmanuel W R, Bindhu A, Mustaq Ahmed M. YOLOv8's advancements in tuberculosis identification from chest images. Frontiers in big Data. 2024:7, 1401981.
32. Amin S U, Taj S, Hussain A, Seo S. An automated chest X-ray analysis for COVID-19, tuberculosis, and pneumonia employing ensemble learning approach. Biomedical Signal Processing and Control. 2024:87, 105408.
33. Cai W, Zhu X, Bai K, Ye A, Zhang R. An explainable dual-mode convolutional neural network for multivariate time series classification. Knowledge-Based Systems. 2024:299, 112015.
34. Majumder R I. Pneumonia & tuberculosis with normal & non-X-ray Dataset [Internet]. Kaggle; 2024 [cited 2026 Jun 30]. Available from: https://www.kaggle.com/datasets/rifatulmajumder23/combined-unknown-pneumonia-and-tuberculosis%20.
35. Coskun M. Enhancing Reliability in Deep Learning Diagnosis of Brain Tumors Using Grad-CAM. Turkish Journal of Nature and Science. 2026:15(2), 39-48.
36. Arı A. ECG-Based Arrhythmia Classification with an Explainable Two-Stage Hybrid Learning Approach. Türk Doğa ve Fen Dergisi. 2026:15(1), 213-220.