Explainable Deep Learning for Lung Disease Detection from Chest X-rays via Layer-wise Relevance Propagation
Downloads
This study proposes an approach to classify lung diseases based on X-ray images using the VGG16 architecture equipped with the Layer-wise Relevance Propagation (LRP) interpretability method. The dataset consists of three classes: COVID-19, pneumonia, and normal, which are processed through augmentation and normalization. The model is trained with a data ratio of 70:30, a learning rate of 0.001, a batch size of 32, and an Adam optimizer. The training results show high accuracy of 96.78% with balanced precision, recall, and F1-score values. The LRP method was used to highlight important areas in the image that contributed to the model's prediction, thereby increasing decision transparency. The main contribution of this research is the integration of VGG16 with LRP in multi-class X-ray image classification, which provides accurate results along with visual interpretations that support confidence in medical applications.
D. Septhya et al., “Implementation of Decision Tree and Support Vector Machine Algorithms for Lung Cancer Classification,” Malcom Indonesian Journal of Machine Learning and Computer Science, vol. 3, no. 1, pp. 15–19, 2023, doi: 10.57152/malcom.v3i1.591.
N. Jain, A. Choudhury, J. Sharma, V. Kumar, D. De, and R. Tiwari, “A review of novel coronavirus infection (Coronavirus Disease- 19),” Global Journal of Transfusion Medicine, vol. 5, no. 1, pp. 22–26, 2020, doi: 10.4103/GJTM.GJTM_24_20.
M. drg. Oscar Primadi and M. dr. Anas Ma'ruf, "PROFILE OF HEALTH IN INDONESIA 2020," 2020.
F. N. Azizah and D. Juniati, “Analysis of Lung Disease Types Based on Chest X-Ray Using the Fuzzy C-Means Method,” Mathunesa Jurnal Ilmiah Matematika, vol. 9, no. 2, pp. 322–331, 2021.-Ray Using the Fuzzy C-Means Method," Mathunesa Scientific Journal of Mathematics, vol. 9, no. 2, pp. 322–331, 2021, doi: 10.26740/mathunesa.v9n2.p322-331.
S. Saturi and S. Banda, “Advanced Lung Disease Detection and Classification Using Ge-U-Net-ODL with Gabor Filters and Entropy- Based Feature Selection,” Journal of Sensors, IoT & Health Sciences, vol. 2, no. 2, pp. 69–86, Jun. 2024, doi: 10.69996/jsihs.2024011.
M. Kranthi, S. Sailaja, and E. Jyothi, “Deep Learning Approaches for Medical Image Processing in the Big Data Era,” vol. 01, no. 01,
pp. 24–31, 2024, doi: 10.58599/ijsmcse.2024.1108. M. E. Karar, E. E. Hemdan, and M. A. Shouman, “Cascaded Deep Learning Classifiers for Computer-Aided Diagnosis of COVID-19 and Pneumonia Diseases in X-Ray Scans,” Complex & Intelligent Systems, vol. 7, no. 1, pp. 235–247, 2020, doi: 10.1007/s40747-020- 00199-4.
C. Karima, N. Bourbia, K. Messaoudi, and E.-B. Bourennane, “Analysis and Classification of Medical Images Using Deep Learning Algorithms,” Recent Advances in Computer Science and Communications, vol. 18, Oct. 2024, doi: 10.2174/0126662558327739240925073925.
M. M. Rahman, K. Matsuo, S. Matsuzaki, and S. Purushotham, “DeepPseudo: Pseudo Value Based Deep Learning Models for Competing Risk Analysis,” Proceedings of the Aaai Conference on Artificial Intelligence, vol. 35, no. 1, pp. 479–487, 2021, doi: 10.1609/aaai.v35i1.16125.
P. Gort, C. Claessens, F. van der Sommen, and P. H. N. de With, “Evaluating the Interpretability of Prototype Networks for Medical Image Analysis,” p. 54, 2025, doi: 10.1117/12.3046678.
D. A. Safitri, Y. P. M. Mamesah, and J. F. J. Timban, “Chest X-ray Findings in Patients with Pulmonary Tuberculosis and Chronic Kidney Disease at Prof. Dr. R. D. Kandou General Hospital, January–June 2022,” Medical Scope Journal, vol. 4, no. 1, pp. 93–98, 2023, doi: 10.35790/msj.v4i1.44722.
D. Kim, J. Lee, J. Moon, and T. Moon, “Interpretable Deep Learning-based Hippocampal Sclerosis Classification,” Epilepsia Open, vol. 7, no. 4, pp. 747–757, 2022, doi: 10.1002/epi4.12655.
S. Gupta et al., “Four Transformer-Based Deep Learning Classifiers Embedded with an Attention U-Net-Based Lung Segmenter and Layer-Wise Relevance Propagation-Based Heatmaps for COVID-19 X-ray Scans,” Diagnostics, vol. 14, no. 14, p. 1534, Jul. 2024, doi: 10.3390/diagnostics14141534.
F. Ahmed et al., “Identification of kidney stones in KUB X-ray images using VGG16 empowered with explainable artificial intelligence,” Sci Rep, vol. 14, no. 1, p. 6173, Mar. 2024, doi: 10.1038/s41598-024-56478-4.
L. Sağır, E. Kaba, M. H. Yiğit, F. Taşçı, and H. Uzun, “Predicting Semen Analysis Parameters From Testicular Ultrasonography Images Using Deep Learning Algorithms: An Innovative Approach to Male Infertility Diagnosis,” J Clin Med, vol. 14, no. 2, p. 516, 2025, doi: 10.3390/jcm14020516.
D. Shen, G. Wu, and H.-I. Suk, “Deep Learning in Medical Image Analysis.,” Annu Rev Biomed Eng, vol. 19, pp. 221–248, Jun. 2017, doi: 10.1146/annurev-bioeng-071516-044442.
H. E. Kim, A. Cosa-Linan, N. Santhanam, M. Jannesari, M. E. Maros, and T. Ganslandt, “Transfer learning for medical image classification: a literature review,” BMC Med Imaging, vol. 22, no. 1, p. 69, Dec. 2022, doi: 10.1186/s12880-022-00793-7.
R. Hao, K. Namdar, L. Liu, M. A. Haider, and F. Khalvati, “A Comprehensive Study of Data Augmentation Strategies for Prostate Cancer Detection in Diffusion-Weighted MRI Using Convolutional Neural Networks,” Journal of Digital Imaging, vol. 34, no. 4,
pp. 862–876, Aug. 2021, doi: 10.1007/s10278-021-00478-7. B. Liu, C. Tan, S. Li, J. He, and W. Hong-yan, “A Data Augmentation Method Based on Generative Adversarial Networks for Grape Leaf Disease Identification,” IEEE Access, vol. 8, pp. 102188–102198, 2020, doi: 10.1109/access.2020.2998839.
H. Iiduka, “Appropriate Learning Rates of Adaptive Learning Rate Optimization Algorithms for Training Deep Neural Networks,” 2020, doi: 10.48550/arxiv.2002.09647.
G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Densely Connected Convolutional Networks,” in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jul. 2017, pp. 2261–2269. doi: 10.1109/CVPR.2017.243.
S.-C. Kim and Y. Cho, “Implementation of a Predictive System to Improve the Accuracy of Urine Self-Diagnosis Using Smartphones: Application of a Confusion Matrix-Based Learning Model Through RGB Semiquantitative Analysis,” Sensors, vol. 22, no. 14,
p. 5445, 2022, doi: 10.3390/s22145445. A. K. M. Nor, S. R. Pedapati, M. Muhammad, and V. Leiva, “Overview of Explainable Artificial Intelligence for Prognostic and Health Management of Industrial Assets Based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses,” Sensors, vol. 21, no. 23, p. 8020, 2021, doi: 10.3390/s21238020.
S. Lebovitz, H. Lifshitz-Assaf, and N. Levina, “To Engage or Not to Engage With AI for Critical Judgments: How Professionals Deal With Opacity When Using AI for Medical Diagnosis,” Organization Science, vol. 33, no. 1, pp. 126–148, 2022, doi: 10.1287/orsc.2021.1549.
P. R. A. S. Bassi, S. S. J. Dertkigil, and A. Cavalli, “Improving deep neural network generalization and robustness to background bias via layer-wise relevance propagation optimization,” Nat Commun, vol. 15, no. 1, p. 291, 2024, doi: 10.1038/s41467-023-44371-z.
