Penerapan Squeeze-and-Excitation Attention pada DenseNet169 untuk Klasifikasi Multi-Kelas Citra X-Ray Dada: Squeeze-and-Excitation Attention on DenseNet169 for Multi-Class Classification of Chest X-Ray Images

Lyra Zulyanda Daulay; Benny Sukma Negara; Yelfi Vitriani; Iwan Iskandar; Fitra Kurnia

doi:10.57152/malcom.v6i2.2665

Authors

Lyra Zulyanda Daulay Universitas Islam Negeri Sultan Syarif Kasim Riau
Benny Sukma Negara Universitas Islam Negeri Sultan Syarif Kasim Riau
Yelfi Vitriani Universitas Islam Negeri Sultan Syarif Kasim Riau
Iwan Iskandar Universitas Islam Negeri Sultan Syarif Kasim Riau
Fitra Kurnia Universitas Islam Negeri Sultan Syarif Kasim Riau

DOI:

https://doi.org/10.57152/malcom.v6i2.2665

Keywords:

Attention Mechanism, Chest X-Ray Image, Densenet169, Multi-Class Classification, Squeeze-and-Excitation

Abstract

Citra X-ray dada (CXR) ialah satu dari metode pencitraan medis yang banyak diterapkan guna mendeteksi banyak penyakit paru diantaranya COVID-19 serta pneumonia. Meskipun model deep learning yang basisnya Convolutional Neural Network (CNN) sudah banyak diterapkan untuk klasifikasi citra medis, model CNN konvensional masih memiliki keterbatasan dalam menyoroti fitur penting pada citra yang memiliki kompleksitas tinggi serta kemiripan pola antar kelas penyakit. Kondisi tersebut dapat menyebabkan model kesulitan dalam mengekstraksi fitur yang paling relevan untuk proses klasifikasi. Oleh sebab itu, penelitian mengusulkan penerapan attention mechanism berupa modul Squeeze-and-Excitation (SE) pada arsitektur DenseNet169 guna meningkatkan kemampuan model dalam menekankan fitur penting dari CXR. Kontribusi utama penelitian ini adalah menganalisis pengaruh mekanisme attention terhadap peningkatan kualitas ekstraksi fitur pada klasifikasi multi-kelas citra CXR. Dataset diterapkan pada penelitian ini terdiri dari 3.000 citra yang diklasifikasikan kedalam tiga kelas yakni Normal, COVID-19, serta Pneumonia dengan pembagian data sebesar 80% untuk pelatihan dan 20% untuk validasi. Pelatihan model menggunakan optimizer Adam. Evaluasi model menggunakan confusion matrix dengan accuracy, precision, recall, dan F1 score. Temuan eksperimen mengindikasi bahwasanya model DenseNet169 baseline memperoleh akurasi senilai 95,8%, sedangkan model DenseNet169 dengan modul SE mencapai akurasi 96,5%. Temuan ini menegaskan bahwa SE meningkatkan representasi fitur sehingga performa klasifikasi lebih optimal.

Downloads

Download data is not yet available.

References

N. Rezaei, Ed., Pneumonia, vol. 13. in Infectious Diseases, vol. 13. IntechOpen, 2022. doi: 10.5772/intechopen.73895.

M. E. H. Chowdhury et al., “Can AI Help in Screening Viral and COVID-19 Pneumonia?,” IEEE Access, vol. 8, pp. 132665–132676, 2020, doi: 10.1109/ACCESS.2020.3010287.

T. Rahman et al., “Exploring the effect of image enhancement techniques on COVID-19 detection using chest X-ray images,” Comput. Biol. Med., vol. 132, May 2021, doi: 10.1016/j.compbiomed.2021.104319.

P. Afshar et al., “COVID-CT-MD, COVID-19 computed tomography scan dataset applicable in machine learning and deep learning,” Sci. Data, vol. 8, no. 1, Dec. 2021, doi: 10.1038/s41597-021-00900-3.

M.Kranthi, S.Sailaja, and E.V.N.Jyothi, “Deep Learning Approaches for Medical Image Processing in the Big Data Era,” International Journal of Scientific Methods in Computational Science and Engineering, vol. 01, no. 01, pp. 24–31, Jun. 2024, doi: 10.58599/ijsmcse.2024.1108.

F. Breve, “COVID-19 Detection on Chest X-Ray Images: A comparison of CNN architectures and ensembles,” May 2022, [Online]. Available: http://arxiv.org/abs/2111.09972

B. Oltu, S. Güney, S. E. Yuksel, and B. Dengiz, “Automated classification of chest X-rays: a deep learning approach with attention mechanisms,” BMC Med. Imaging, vol. 25, no. 1, Dec. 2025, doi: 10.1186/s12880-025-01604-5.

S. Potharaju, S. N. Tambe, K. Dasari, N. Srikanth, R. Venkatarao, and S. Tambe, “Enhanced X-ray image classification for pneumonia detection using deep learning based CBAM and SE mechanisms,” Intell. Based. Med., vol. 12, Jan. 2025, doi: 10.1016/j.ibmed.2025.100299.

A. Torbicki, “Pulmonary hypertension: Diagnosis and management,” Jun. 01, 2021, Multidisciplinary Digital Publishing Institute (MDPI). doi: 10.3390/diagnostics11061066.

Y. J. Wei et al., “Effects of Diesel Hydrocarbon Components on Cetane Number and Engine Combustion and Emission Characteristics,” Applied Sciences (Switzerland), vol. 12, no. 7, Apr. 2022, doi: 10.3390/app12073549.

T. Finck et al., “Faster and Better: How Anomaly Detection Can Accelerate and Improve Reporting of Head Computed Tomography,” Diagnostics, vol. 12, no. 2, Feb. 2022, doi: 10.3390/diagnostics12020452.

S. Kumar et al., “LiteCovidNet: A lightweight deep neural network model for detection of COVID-19 using X-ray images,” Int. J. Imaging Syst. Technol., vol. 32, no. 5, pp. 1464–1480, Sep. 2022, doi: 10.1002/ima.22770.

S. Shastri, I. Kansal, S. Kumar, K. Singh, R. Popli, and V. Mansotra, “CheXImageNet: a novel architecture for accurate classification of Covid-19 with chest x-ray digital images using deep convolutional neural networks,” Health Technol. (Berl)., vol. 12, no. 1, pp. 193–204, Jan. 2022, doi: 10.1007/s12553-021-00630-x.

G. S. P. Ramadhan, M. Maimunah, and S. Nugroho, “Optimasi Data Preprocessing dan Hyperparameter Tuning pada Klasifikasi Penyakit Daun Apel menggunakan DenseNet169,” Building of Informatics, Technology and Science (BITS), vol. 6, no. 3, pp. 1352–1362, Dec. 2024, doi: 10.47065/bits.v6i3.6134.

D. Husen, F. Ilmu Komputer, U. Kuningan Kuningan, and J. Barat, “Husen-Evaluasi Teknik Augmentasi Data Untuk Klasifikasi Tumor Otak Menggunakan CNN Pada Citra MRI Evaluasi Teknik Augmentasi Data Untuk Klasifikasi Tumor Otak Menggunakan CNN Pada Citra MRI (Performance Evaluation OF CNN Models With Various Data Augmentasi Techniques ON MRI Images For Brain Tumor Classification).”

H. Bichri, A. Chergui, and M. Hain, “Investigating the Impact of Train / Test Split Ratio on the Performance of Pre-Trained Models with Custom Datasets,” 2024. [Online]. Available: www.ijacsa.thesai.org

T. Zhou, X. Ye, H. Lu, X. Zheng, S. Qiu, and Y. Liu, “Dense Convolutional Network and Its Application in Medical Image Analysis,” 2022, Hindawi Limited. doi: 10.1155/2022/2384830.

P. P. Dalvi, D. R. Edla, and B. R. Purushothama, “Diagnosis of Coronavirus Disease From Chest X-Ray Images Using DenseNet-169 Architecture,” May 01, 2023, Springer. doi: 10.1007/s42979-022-01627-7.

A. Hernández and J. M. Amigó, “Attention mechanisms and their applications to complex systems,” Entropy, vol. 23, no. 3, pp. 1–18, Mar. 2021, doi: 10.3390/e23030283.

K. A. Huang, H. K. Choudhary, A. Santiago, and N. S. Prakash, “Squeeze-and-Excitation Enhanced Convolutional Neural Networks for Multi-class Pneumonia Classification on Chest Radiographs,” Cureus, Dec. 2025, doi: 10.7759/cureus.99649.

C. Halkiopoulos, E. Gkintoni, A. Aroutzidis, and H. Antonopoulou, “Advances in Neuroimaging and Deep Learning for Emotion Detection: A Systematic Review of Cognitive Neuroscience and Algorithmic Innovations,” Feb. 01, 2025, Multidisciplinary Digital Publishing Institute (MDPI). doi: 10.3390/diagnostics15040456.

K. A. Huang and N. Prakash, “Evaluating the Impact of Attention Mechanisms on a Fine-Tuned Neural Network for Magnetic Resonance Imaging Tumor Classification: A Comparative Analysis,” Cureus, Mar. 2025, doi: 10.7759/cureus.80872.

S. Sathyanarayanan, “Confusion Matrix-Based Performance Evaluation Metrics,” African Journal of Biomedical Research, pp. 4023–4031, Nov. 2024, doi: 10.53555/ajbr.v27i4s.4345.

B. Kocak et al., “Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendations,” European Journal of Radiology Artificial Intelligence, vol. 3, p. 100030, Sep. 2025, doi: 10.1016/j.ejrai.2025.100030.