Analisis Penerapan Flexmatch Pada Semi-Supervised Deep Learning Untuk Deteksi Penyakit Paru-Paru Berbasis Citra Chest X-Ray
DOI:
https://doi.org/10.52060/juptik.v4i1.4439Abstract
Penyakit paru-paru seperti COVID-19 dan Pneumonia merupakan penyebab utama morbiditas dan mortalitas di dunia, sehingga diperlukan metode deteksi dini yang akurat dan efisien. Penelitian ini bertujuan menganalisis penerapan FlexMatch pada skema Semi-Supervised Deep Learning untuk klasifikasi penyakit paru-paru berbasis citra Chest X-ray (CXR), dengan memanfaatkan data berlabel dan tidak berlabel melalui mekanisme Curriculum Pseudo Labeling dan class-adaptive thresholding, serta DenseNet-169 sebagai ekstraktor fitur utama. Dataset yang digunakan terdiri dari 3.000 citra CXR yang mencakup tiga kelas, yaitu COVID-19, pneumonia, dan normal. Tahapan penelitian meliputi preprocessing data, augmentasi citra, pembagian data, pelatihan model, serta evaluasi menggunakan accuracy, precision, recall, F1-score, dan Grad-CAM. Hasil penelitian menunjukkan bahwa model mencapai akurasi validasi sebesar 96,65% dengan F1-score masing-masing sebesar 99,02% untuk COVID-19, 95,74% untuk Normal, dan 94,48% untuk Pneumonia. Visualisasi Grad-CAM membuktikan bahwa model mampu memfokuskan perhatian pada area paru yang relevan secara klinis, sehingga FlexMatch terbukti efektif meningkatkan performa klasifikasi pada kondisi data berlabel terbatas dan berpotensi mendukung sistem diagnosis penyakit paru-paru berbasis kecerdasan buatan.
References
[1] W. H. Organization, “Global Tuberculosis Report 2022,” WHO, Geneva, 2022. [Online]. Available: https://www.who.int/teams/global-programme-on-tuberculosis-and-lung-health/tb-reports/global-tuberculosis-report-2022.
[2] P. Rajpurkar, E. Chen, E. O’Keefe, and N. Shah, “Deep learning for chest radiograph diagnosis: A retrospective comparison of CheXNeXt to practicing radiologists,” PLoS Med., vol. 15, no. 11, p. e1002686, 2022, doi: https://doi.org/10.1371/journal.pmed.1002686.
[3] J. A. Maharani and M. Masriadi, “Pengaruh Artificial Intelegen Dalam Mendeteksi Kasus Penyakit Tidak Menular,” J. Artif. Intell. Digit. Bus., vol. 4, no. 2, pp. 7343–7349, 2025, https://doi.org/10.31004/riggs.v4i2.1864
[4] C. J. Ryerson et al., “Update of the international multidisciplinary classification of the interstitial pneumonias: an ERS/ATS statement,” Eur. Respir. J., 2025, https://doi.org/10.1183/13993003.00158-2025
[5] M. Merk, “Still a role for the chest radiograph – humble but helpful!,” African J. Thorac. Crit. Care Med., vol. 28, p. 143, 2022, https://doi.org/10.7196/AJTCCM.2022.v28i4.300
[6] H. K. Ahmad et al., “Machine Learning Augmented Interpretation of Chest X-rays: A Systematic Review,” Diagnostics, vol. 13, no. 4, p. 743, 2023, https://doi.org/10.3390/diagnostics13040743
[7] P. Syifa, Safwandi, and Z. Fitri, “Sistem pakar diagnosis penyakit paru menggunakan metode convolutional neural network dan rule based system,” Rabit J. Teknol. Dan Sist. Inf. Univrab, vol. 10, no. 2, pp. 1380–1392, 2025, https://doi.org/10.1016/j.bea.2023.100076
[8] S. Modak, “Applications of deep learning in disease diagnosis of chest radiographs: A survey on materials and methods,” Biomed. Eng. Adv., vol. 5, p. 100076, 2023, https://doi.org/10.1016/j.bea.2023.100076
[9] D. Banik and D. Bhattacharjee, “Mitigating Data Imbalance Issues in Medical Image Analysis,” 2021, pp. 66–89. https://doi.org/10.4018/978-1-7998-7371-6.ch004
[10] Y. Xu, Y. Li, D. Wang, Y. Zhang, and D. Huang, “Addressing the current challenges in the clinical application of AI-based Radiomics for cancer imaging,” Front. Med., vol. 12, 2025, https://doi.org/10.3389/fmed.2025.1674397
[11] A. Chowdhury, J. Rosenthal, J. Waring, and R. Umeton, “Applying Self-Supervised Learning to Medicine: Review of the State of the Art and Medical Implementations,” Informatics, vol. 8, no. 3, p. 59, 2021, https://doi.org/10.3390/informatics8030059
[12] E. M. J. S. Fdo, A. A. A. Rekshin, G. G. Gains, X. A. M. Mcchenzi, and A. J. P. Peniel, “Self-supervised learning for small-scale medical imaging dataset,” World J. Adv. Eng. Technol. Sci., vol. 13, no. 2, pp. 142–145, 2024, https://doi.org/10.30574/wjaets.2024.13.2.0526
[13] L.-Z. Guo, L.-H. Jia, J.-J. Shao, and Y. Li, “Robust semi-supervised learning in open environments,” Front. Comput. Sci., vol. 19, no. 8, 2025, https://doi.org/10.1007/s11704-024-40646-w
[14] B. Zhang et al., “FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling,” N/A, p. 34, 2021, [Online]. Available: https://proceedings.neurips.cc/paper/2021/hash/995693c15f439e3d189b06e89d145dd5-Abstract.html.
[15] G. Gui, Z. Zhao, L. Qi, L. Zhou, L. Wang, and Y.-H. Shi, “Enhancing Sample Utilization through Sample Adaptive Augmentation in Semi-Supervised Learning,” arXiv.Org, 2023, https://doi.org/10.1109/ICCV51070.2023.01455
[16] V. Leon, A. Pasko, and B. Terenzio, “Progress and Prospects in Deep Learning for Chest X-Ray Interpretation,” 2025, https://doi.org/10.36227/techrxiv.173592311.10695741/v1
[17] M. O. Sahetai, D. Chauhan, and R. Jani, “Harnessing Artificial Intelligence for Precise Pulmonary Disease Diagnosis,” 2023, pp. 151–157, https://doi.org/10.1109/ICSSAS57918.2023.10331667
[18] S. Kumar, “Covid19-Pneumonia-Normal Chest X-Ray Images,” vol. V1. Mendeley Data, 2022, doi: 10.17632/dvntn9yhd2.1.
[19] 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, https://doi.org/10.3390/diagnostics12061457
[20] K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770–778, https://doi.org/10.1109/CVPR.2016.90
[21] X. Zhao, S. Qi, B. Zhang, H. Lian, S. Yao, and W. Sun, “Deep Learning for Pulmonary Nodule Detection: A Review,” Front. Oncol., vol. 12, p. 837279, 2022. https://doi.org/10.3389/fonc.2022.983833
[22] 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, https://doi.org/10.3389/fmed.2021.629134
[23] H. Jin, H. Che, Y. Lin, and H. Chen, “Promptmrg: Diagnosis-driven prompts for medical report generation,” 2024, [Online]. Available: https://ojs.aaai.org/index.php/AAAI/article/download/28038/28087.
[24] Y. Wang et al., “FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling,” Adv. Neural Inf. Process. Syst., vol. 34, pp. 25324–25336, 2021.
[25] 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, https://doi.org/10.1007/s42979-022-01627-7
[26] M. N. Dawami, B. S. Negara, M. Irsyad, and F. Yanto, “Klasifikasi Multikelas Citra Chest X-Ray Menggunakan Semi-Supervised SoftMatch pada Label Terbatas,” vol. 6, no. 12, pp. 2345–2360, 2026, doi: 10.47065/tin.v6i12.9848.
[27] M. Shetty and S. Shetty, “Computer-Aided Diagnostic of COVID-19 Using Chest X-Ray Analysis BT - Proceedings of Second Doctoral Symposium on Computational Intelligence,” 2022, pp. 675–681. https://doi.org/10.1007/978-981-16-3346-1_54
[28] G. Huang and K. Q. Weinberger, “Densely Connected Convolutional Networks,” doi: https://doi.org/10.48550/arXiv.1608.06993.
[29] F. Shahira and B. S. Negara, “Lung X-Ray Image Classification Using DenseNet-169 and Bayesian Optimization,” vol. 05, no. 01, pp. 18–27, 2025. https://doi.org/10.30983/knowbase.v5i1.9618
[30] J. Van Zyl, “Analysis of classification metric behaviour under class imbalance,” Egypt. Informatics J., vol. 31, no. July, p. 100711, 2025, doi: 10.1016/j.eij.2025.100711.
[31] E. Selby, M. Woodruff, and J. Carter, “Explainable AI in Medical Imaging: A Systematic Review of Grad-CAM Applications,” J. Med. Imaging Heal. Informatics, vol. 13, no. 2, pp. 145–162, 2023, doi: 10.1166/jmihi.2023.3821.
[32] B. Sukma Negara, T. Hariyanto, and R. Pratiwi, “Grad-CAM Visualization on DenseNet Architecture for Chest X-Ray Classification,” Int. J. Intell. Syst. Appl. Eng., vol. 11, no. 3, pp. 512–521, 2023, doi: 10.18201/ijisae.2023.412.
[33] E. Tjoa and C. Guan, “A Survey on Explainable Artificial Intelligence (XAI): Toward Medical XAI,” IEEE Trans. Neural Networks Learn. Syst., vol. 32, no. 11, pp. 4793–4813, 2021, https://doi.org/10.1109/TNNLS.2020.3027314
[34] K. Sohn et al., “FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence,” in Advances in Neural Information Processing Systems (NeurIPS), 2020, vol. 33, doi: 10.48550/arXiv.2001.07685.
[35] A. Setiawan, I. K. E. Purnama, and M. H. Purnomo, “Chest X-Ray Classification for COVID-19 Detection Using DenseNet-121,” J. Electron. Electromed. Eng. Med. Informatics, vol. 4, no. 2, pp. 78–86, 2022, https://doi.org/10.35882/jeeemi.v4i2.5
| Keywords | : |
Keywords:
FlexMatch, Semi-Supervised Learning, Deep Learning, DenseNet-169, Chest X-ray, Klasifikasi Penyakit Paru-Paru Grad-CAM
|
| Galleys | : | |
| Published | : |
2026-06-01
|
| Issue | : |
Copyright (c) 2026 Jurnal Pengembangan Teknologi Informasi dan Komunikasi (JUPTIK)
This work is licensed under a Creative Commons Attribution 4.0 International License.
