ANALISIS PERBANDINGAN CNN DENGAN COST-SENSITIVE LEARNING UNTUK DETEKSI TUBERKULOSIS PARU PADA CITRA X-RAY THORAX 4 1
DOI:
https://doi.org/10.26623/elektrika.v18i2.16277Keywords:
chest X-ray, convolutional neural network, cost-sensitive learning, EfficientNetB0, tuberculosisAbstract
Tuberculosis (TB) remains one of the heaviest infectious disease burdens in Indonesia, with an estimated 1.09 million cases and 125,000 deaths each year, while access to radiologists in primary health facilities is limited. This study compares two EfficientNetB0 models for classifying chest X-ray images as normal or TB: a baseline model trained without class weighting and a cost-sensitive model that applies class weights of 0.6 for the normal class and 3.0 for the TB class. The Tuberculosis (TB) Chest X-ray Database, containing 3,500 normal and 700 TB images, was divided into 3,000 training, 950 validation, and 250 test images (200 normal, 50 TB). On the test set, the baseline model achieved 97.20% accuracy, 98.00% sensitivity, 97.00% specificity, 89.09% precision, and a 93.33% F1-score, with 6 false positives and 1 false negative. The cost-sensitive model classified all 250 images correctly, reaching 100% on every metric with a ROC-AUC of 1.0000. The cost-sensitive model therefore performed better on this test set, mainly by eliminating false positives and the single missed TB case. The small test set, the single training run, and the absence of external validation limit these findings, so they need confirmation through repeated training with different random seeds, statistical testing, and external validation.
ABSTRAK
Tuberkulosis (TB) tergolong beban penyakit infeksi terberat di Indonesia dengan estimasi 1,09 juta kasus dan 125.000 kematian setiap tahun, sementara ketersediaan radiolog di fasilitas kesehatan primer masih terbatas. Penelitian ini membandingkan dua model EfficientNetB0 untuk mengklasifikasikan citra X-ray toraks sebagai normal atau TB, yaitu model baseline tanpa pembobotan kelas dan model cost-sensitive dengan bobot kelas 0,6 untuk kelas normal dan 3,0 untuk kelas TB. Tuberculosis (TB) Chest X-ray Database yang memuat 3.500 citra normal dan 700 citra TB dibagi menjadi 3.000 citra latih, 950 citra validasi, dan 250 citra uji (200 normal, 50 TB). Pada data uji, model baseline memperoleh akurasi 97,20%, sensitivitas 98,00%, spesifisitas 97,00%, presisi 89,09%, dan F1-score 93,33% dengan 6 false positive dan 1 false negative. Model cost-sensitive mengklasifikasikan seluruh 250 citra dengan benar sehingga seluruh metriknya 100% dengan ROC-AUC 1,0000. Dengan demikian, model cost-sensitive lebih unggul pada data uji ini, terutama karena tidak menghasilkan false positive dan tidak melewatkan kasus TB. Jumlah data uji yang kecil, pelatihan yang hanya dilakukan sekali, dan belum adanya validasi eksternal membatasi temuan ini, sehingga perlu dikonfirmasi melalui pelatihan ulang dengan beberapa random seed, uji statistik, dan validasi eksternal.
Kata Kunci: citra X-ray toraks, convolutional neural network, cost-sensitive learning, EfficientNetB0, tuberculosis
Downloads
References
[1] WHO, Global Tuberculosis Report 2024 (2024th ed). 2024. https://iris.who.int/server/api/core/bitstreams/7292c91e-ffb0-4cef-ac39-0200f06961ea/content
[2] H. Lee, J. Kim, J. Kim, and Y.-J. Park, “Review of the Global Burden of Tuberculosis in 2023: Insights from the WHO Global Tuberculosis Report 2024,” Jugan geon-gang gwa jilbyeong, vol. 18, no. 11 Suppl, pp. 55–69, Mar. 2025, doi: 10.56786/PHWR.2025.18.11suppl.5.
[3] Suryanti and I. A. Ahmed, “Recent Review Tuberculosis in Indonesia: Burden and the Challenge of Under-Reporting.,” Iranian journal of public health, vol. 54, no. 2, pp. 445–446, Feb. 2025. https://pmc.ncbi.nlm.nih.gov/articles/PMC11992917/
[4] K. L. Tobing et al., “Under-reporting of Childhood Tuberculosis Cases in Indonesia and Associated Factors,” Journal of Preventive Medicine and Public Health, vol. 59, no. 3, pp. 289–297, 2026, doi: 10.3961/jpmph.25.894.
[5] World Health Organization. WHO consolidated guidelines on tuberculosis. Module 2: screening – systematic screening for tuberculosis disease. Geneva: World Health Organization; 2021. https://www.who.int/publications/i/item/9789240022676.
[6] Z. Z. Qin et al., “Tuberculosis detection from chest x-rays for triaging in a high tuberculosis-burden setting: an evaluation of five artificial intelligence algorithms,” The Lancet Digital Health, vol. 3, no. 9, pp. e543–e554, 2021, doi: 10.1016/S2589-7500(21)00116-3.
[7] S. Hansun, A. Argha, S. T. Liaw, B. G. Celler, and G. B. Marks, “Machine and Deep Learning for Tuberculosis Detection on Chest X-Rays: Systematic Literature Review,” Journal of Medical Internet Research, vol. 25, 2023, doi: 10.2196/43154.
[8] M. Oloko-Oba and S. Viriri, “A Systematic Review of Deep Learning Techniques for Tuberculosis Detection From Chest Radiograph,” Frontiers in Medicine, vol. 9, 2022, doi: 10.3389/fmed.2022.830515.
[9] E. Showkatian, M. Salehi, H. Ghaffari, R. Reiazi, and N. Sadighi, “Deep learning-based automatic detection of tuberculosis disease in chest X-ray images,” Polish Journal of Radiology, vol. 87, no. 1, pp. 118–124, 2022, doi: 10.5114/pjr.2022.113435.
[10] A. Mirugwe, L. Tamale, and J. Nyirenda, “Improving Tuberculosis Detection in Chest X-Ray Images Through Transfer Learning and Deep Learning: Comparative Study of Convolutional Neural Network Architectures,” JMIRx Med, vol. 6, pp. e66029–e66029, 2025, doi: 10.2196/66029.
[11] D. Witarsyah, H. Almohab, R. S. Sianturi, and D. Syamsuar, “Enhancing Tuberculosis Detection from Chest X-Ray Images Using Deep Learning: Evaluating Multi-Architecture Performance and Efficiency,” Engineering, Technology and Applied Science Research, vol. 16, no. 2, pp. 34365–34372, 2026, doi: 10.48084/etasr.16321.
[12] V. T. Q. Huy and C. M. Lin, “An Improved Densenet Deep Neural Network Model for Tuberculosis Detection Using Chest X-Ray Images,” IEEE Access, vol. 11, pp. 42839–42849, 2023, doi: 10.1109/ACCESS.2023.3270774.
[13] V. Pal, H. Pabari, S. Indoria, S. Patel, D. Krishnan, and V. Ravi, “Multifaceted Disease Diagnosis: Leveraging Transfer Learning with Deep Convolutional Neural Networks on Chest X-Rays for COVID-19, Pneumonia, and Tuberculosis,” The Open Bioinformatics Journal, vol. 17, no. 1, p. 55, 2024, doi: 10.2174/0118750362303182240516043224.
[14] K. K. Goswami, R. Kumar, R. Kumar, A. J. Reddy, and S. K. Goswami, “Deep Learning Classification of Tuberculosis Chest X-rays,” Cureus, 2023, doi: 10.7759/cureus.41583.
[15] M. Owda, A. Abumihsan, A. Y. Owda, and M. Abumohsen, “A Lightweight Hybrid Deep Learning Model for Tuberculosis Detection from Chest X-Rays.,” Diagnostics (Basel, Switzerland), vol. 15, no. 24, Dec. 2025, doi: 10.3390/diagnostics15243216..
[16] K. K. Mujeeb Rahman, S. Zulaikha, B. Dhafer, and R. Ahmed, “Advancing tuberculosis screening: A tailored CNN approach for accurate chest X-ray analysis and practical clinical integration,” Intelligence-Based Medicine, vol. 11, p. 100196, 2025, doi: https://doi.org/10.1016/j.ibmed.2024.100196.
[17] A. Supriyanto, R. R. Isnanto, and O. D. Nur-hayati, “Klasifikasi penyakit daun kopi ro-busta menggunakan metode SVM dengan ekstraksi ciri GLCM”, Jurnal Nasional Teknik Elektro dan Teknologi Informasi, vol. 12, no. 4, hal. 241-248, Nov. 2023. doi: https://doi.org/10.22146/jnteti.v12i4.8044
[18] Y. D. Al Iman, S. E. Putri, and A. Supriyanto, “Klasifikasi Pneumonia Pada Citra X-Ray Dada Menggunakan Ekstraksi Ciri Convolutional Neural Network dan Algoritma K-Nearest Neighbor”, Elektrika, vol. 18, no. 1, pp. 31–39, Apr. 2026, doi: 10.26623/elektrika.v18i1.14211.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Elektrika
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Authors who publish this journal agree to the following terms:
The author owns the copyright and grants the journal the first publication rights with the work simultaneously licensed under the Creative Commons Attribution 4.0 International License which allows others to share the work with recognition of the authorship of the work and initial publication in the journal.
Authors may enter into separate additional contractual agreements for non-exclusive distribution of the published journal version of the work (e.g., posting it to an institutional repository or publishing it in a book), in recognition of its initial publication in this journal.
Authors are allowed and encouraged to post their work online (e.g., in institutional repositories or on their websites) before and during the submission process, as it can lead to productive exchanges, as well as earlier and larger citations of published works (See The Effects of Open Access).
This work is licensed under the Creative Commons Attribution 4.0 International License.
