Penerapan Semi-Supervised Deep Learning dengan AdaMatch untuk Klasifikasi Penyakit Paru-paru pada Citra X-ray Dada: Application of Semi-Supervised Deep Learning with AdaMatch for Classification of Lung Disease on Chest X-ray Image

Hilya Zalwana; Benny Sukma Negara; Muhammad Irsyad; Suwanto Sanjaya; Muhammad Fikry

doi:10.57152/malcom.v6i3.2811

Authors

Hilya Zalwana Universitas Islam Negeri Sultan Syarif Kasim Riau
Benny Sukma Negara Universitas Islam Negeri Sultan Syarif Kasim Riau
Muhammad Irsyad Universitas Islam Negeri Sultan Syarif Kasim Riau
Suwanto Sanjaya Universitas Islam Negeri Sultan Syarif Kasim Riau
Muhammad Fikry Universitas Islam Negeri Sultan Syarif Kasim Riau

DOI:

https://doi.org/10.57152/malcom.v6i3.2811

Keywords:

AdaMatch, Chest X-Ray, DenseNet-169, Klasifikasi Multi-Kelas, Semi-Supervised Deep Learning

Abstract

Penyakit paru-paru, seperti Coronavirus Disease 2019 (COVID-19) dan pneumonia, masih menjadi tantangan kesehatan yang memerlukan diagnosis cepat dan akurat. Citra chest X-ray (CXR) banyak digunakan untuk mendukung diagnosis, namun interpretasinya masih bergantung pada radiolog dan ketersediaan data berlabel. Keterbatasan data berlabel menjadi kendala dalam pengembangan model deep learning berbasis supervised learning. Penelitian ini menerapkan pendekatan semi-supervised deep learning menggunakan AdaMatch dengan DenseNet-169 untuk klasifikasi multikelas citra CXR menjadi COVID-19, Pneumonia, dan Normal. Dataset publik Mendeley Data yang digunakan terdiri atas 5.228 citra CXR, dengan pembagian 70% data pelatihan, 10% validasi, dan 20% pengujian. Tiga skenario proporsi data berlabel, yaitu 5%, 10%, dan 20%, digunakan untuk mengevaluasi performa model. AdaMatch memanfaatkan data berlabel dan tidak berlabel melalui mekanisme adaptive thresholding, distribution alignment, dan consistency regularization. Hasil terbaik diperoleh pada skenario 20% data berlabel dengan akurasi 98,19%, sensitivitas 98,23%, dan F1-score 98,24%. Performa tersebut mendekati model supervised learning pembanding yang memperoleh akurasi 98,95%, sensitivitas 98,98%, dan F1-score 98,98%. Temuan ini menunjukkan bahwa AdaMatch merupakan pendekatan semi-supervised yang efektif untuk meningkatkan klasifikasi citra CXR pada kondisi keterbatasan data berlabel.

Downloads

Download data is not yet available.

References

WHO, “Pneumonia in children,” 11 November 2022. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/pneumonia

WHO, “WHO COVID-19 dashboard,” July 2023. [Online]. Available: https://data.who.int/dashboards/covid19/deaths

M. Sakaida et al., “The Effectiveness of Semi-Supervised Learning Techniques in Identifying Calcifications in X-ray Mammography and the Impact of Different Classification Probabilities,” Appl. Sci., vol. 14, no. 14, 2024, doi: 10.3390/app14145968.

A. Al Foysal and S. Sultana, “AI-Driven Pneumonia Diagnosis Using Deep Learning?: A Comparative Analysis of CNN Models on Chest X-Ray Images,” vol. 12, 2025, doi: 10.4236/oalib.1112899.

P. Sahoo, I. Roy, R. Ahlawat, S. Irtiza, and L. Khan, “Potential diagnosis of COVID-19 from chest X-ray and CT findings using semi-supervised learning,” Phys. Eng. Sci. Med., vol. 45, no. 1, pp. 31–42, 2022, doi: 10.1007/s13246-021-01075-2.

J. Yang, M. Chen, Q. Jia, and S. Liu, “Unlabeled Insight , Labeled Boost?: Contrastive Learning and Class-Adaptive Pseudo-Labeling for Semi-Supervised Medical Image Classification,” pp. 1–20, 2025, doi: doi.org/10.3390/e27101015.

J. Zhou, B. Jing, Z. Wang, H. Xin, and H. Tong, “SODA: Detecting COVID-19 in Chest X-Rays With Semi-Supervised Open Set Domain Adaptation,” IEEE/ACM Trans. Comput. Biol. Bioinforma., vol. 19, no. 5, pp. 2605–2612, 2022, doi: 10.1109/TCBB.2021.3066331.

G. Huang, Q. Fu, M. Gu, N. Lu, K. Liu, and T. Chen, “Deep Transfer Learning for the Multilabel Classification of Chest X-ray Images,” 2022, doi: doi.org/10.3390/ diagnostics12061457.

F. A. Breve, “COVID-19 detection on Chest X-ray images?: A comparison of CNN architectures and ensembles,” Expert Syst. Appl., vol. 204, no. May, p. 117549, 2022, doi: 10.1016/j.eswa.2022.117549.

P. Pradeep, D. Damodar, and R. Edla, “Diagnosis of Coronavirus Disease From Chest X ? Ray Images Using DenseNet ? 169 Architecture,” SN Comput. Sci., vol. 4, no. 3, pp. 1–6, 2023, doi: 10.1007/s42979-022-01627-7.

I. Sultan, H. Gharaibeh, A. Gharaibeh, B. Lahham, M. Al-tarawneh, and R. Al-qawabah, “LungVisionNet?: A Hybrid Deep Learning Model for Chest X-Ray Classification — A Case Study at King Hussein Cancer Center ( KHCC ),” pp. 1–25, 2025.

C. Disease and D. Using, “A Review of Recent Advances in Deep Learning Models for Chest Disease Detection Using Radiography,” 2023, doi: doi.org/10.3390/ diagnostics13010159.

S. Calderon-ramirez et al., “Dealing with Scarce Labelled Data?: Semi-supervised Deep Learning with Mix Match for Covid-19 Detection Using Chest X-ray Images,” pp. 5294–5301, 2021, doi: 10.1109/ICPR48806.2021.9412946.

A. R. Sajun and I. Zualkernan, “applied sciences Investigating the Performance of FixMatch for COVID-19 Detection in Chest X-rays,” 2022, doi: doi.org/10.3390/ app12094694.

A. H. Edoardo Vantaggiato, Emanuela Paladini, Fares Bougourzi, Cosimo Distante, “COVID-19 Recognition Using Ensemble-CNNs in Two New Chest X-ray Databases,” pp. 1–20, 2021, doi: doi.org/10.3390/s21051742 Academic.

S. Zoha, J. G. Lee, and Y. W. Ko, “CAT: Class-aware adaptive-thresholding for robust semi-supervised domain generalization,” PLoS One, vol. 20, no. 9 September, pp. 1–20, 2025, doi: 10.1371/journal.pone.0329799.

D. Berthelot, R. Roelofs, K. Sohn, N. Carlini, and A. Kurakin, “Adamatch: a Unified Approach To Semi-Supervised Learning and Domain Adaptation,” ICLR 2022 - 10th Int. Conf. Learn. Represent., no. 2, pp. 1–50, 2022.

U. Chutia, A. Shanker, T. Jyoti, P. Singh, and V. Kumar, “Classification of Lung Diseases Using an Attention ? Based Modified DenseNet Model,” pp. 1625–1641, 2024, doi: 10.1007/s10278-024-01005-0.

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)., pp. 193–204, 2022, doi: 10.1007/s12553-021-00630-x.

S. Kumar and R. Rani, “LiteCovidNet?: A lightweight deep neural network model for detection of COVID-19 using X-ray images,” no. March 2021, pp. 1464–1480, 2022, doi: 10.1002/ima.22770.

D. Zhang, F. Ren, Y. Li, L. Na, and Y. Ma, “Pneumonia Detection from Chest X-ray Images Based on Convolutional Neural Network,” 2021, doi: doi.org/10.3390/electronics10131512 1.

C. Mosquera, L. Ferrer, D. H. Milone, D. Luna, and E. Ferrante, “Class imbalance on medical image classification: towards better evaluation practices for discrimination and calibration performance.,” Eur. Radiol., vol. 34, no. 12, pp. 7895–7903, Dec. 2024, doi: 10.1007/s00330-024-10834-0.

M. Elgendi et al., “The Effectiveness of Image Augmentation in Deep Learning Networks for Detecting COVID-19?: A Geometric Transformation Perspective,” vol. 8, no. March, pp. 1–12, 2021, doi: 10.3389/fmed.2021.629134.

D. L. Rubin, C. P. Langlotz, and A. S. Chaudhari, “Exploring Image Augmentations for Siamese Representation Learning with Chest X-Rays,” pp. 444–467, 2023.

A. Ke, W. Ellsworth, and A. Y. Ng, CheXtransfer?: Performance and Parameter Efficiency of ImageNet Models for Chest X-Ray Interpretation, vol. 1, no. 1. Association for Computing Machinery, 2021. doi: 10.1145/3450439.3451867.

B. Kocak et al., “European Journal of Radiology Artificial Intelligence Evaluation metrics in medical imaging AI?: fundamentals , pitfalls , misapplications , and recommendations,” vol. 3, no. July, 2025, doi: 10.1016/j.ejrai.2025.100030.

R. Rs, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization,” Int. J. Comput. Vis., vol. 128, Feb. 2020, doi: 10.1007/s11263-019-01228-7.