Dilation Layer Menggunakan EfficientNetV2 untuk Mendeteksi Penyakit Retinopati Diabetik
DOI:
https://doi.org/10.25126/jtiik.2025126Kata Kunci:
Diabetic Retinopathy, EfficientNetV2, dilated convolution, klasifikasi citra, CNNAbstrak
Diabetic Retinopathy (DR) merupakan salah satu komplikasi serius akibat diabetes melitus yang dapat menyebabkan kebutaan jika tidak dideteksi dan ditangani secara dini. Deteksi otomatis berbasis citra fundus retina telah menjadi fokus utama dalam penelitian deep learning. Namun, fitur-fitur kecil seperti mikroaneurisma dan lesi halus seringkali sulit dikenali, terutama pada citra dengan kualitas rendah. Penelitian ini mengusulkan modifikasi arsitektur EfficientNetV2 dengan menambahkan dilation layer guna meningkatkan kemampuan model dalam menangkap fitur kontekstual dengan cakupan spasial yang lebih luas tanpa menambah kompleksitas secara signifikan. Dilated convolution memungkinkan perluasan receptive field tanpa mengurangi resolusi spasial, sehingga lebih efektif dalam mendeteksi fitur-fitur kritis pada DR. Pengujian dilakukan menggunakan dataset APTOS 2019 Blindness Detection yang menyediakan citra fundus retina beranotasi untuk lima tingkat keparahan DR. Model dilatih menggunakan optimizer Adam dengan learning rate sebesar 0,001 untuk mencapai konvergensi yang optimal. Evaluasi dilakukan terhadap beberapa varian EfficientNetV2, yaitu B0, B1, B2, B3, dan S. Hasil eksperimen menunjukkan bahwa model EfficientNetV2-B0 dan B1 memberikan akurasi terbaik, masing-masing sebesar 97,14%. Di antara keduanya, EfficientNetV2-B0 menghasilkan nilai presisi dan f1-score yang lebih tinggi. Temuan ini menunjukkan bahwa integrasi dilation layer pada arsitektur EfficientNetV2 berpotensi meningkatkan performa sistem klasifikasi DR otomatis secara signifikan dibandingkan model baseline tanpa modifikasi.
Abstract
Diabetic Retinopathy (DR) is a serious complication of diabetes mellitus that can lead to blindness if not detected and treated early. Automated detection based on retinal fundus images has become a primary focus in deep learning research. However, small features such as microaneurysms and subtle lesions are often difficult to identify, particularly in low-quality images. This study proposes a modification to the EfficientNetV2 architecture by incorporating dilation layers to enhance the model’s ability to capture contextual features over a broader spatial range without significantly increasing complexity. Dilated convolutions allow the expansion of the receptive field without reducing spatial resolution, making them more effective in detecting critical features associated with DR. The evaluation was conducted using the APTOS 2019 Blindness Detection dataset, which provides annotated retinal fundus images across five levels of DR severity. The model was trained using the Adam optimizer with a learning rate of 0.001 to achieve optimal convergence. The study assessed several variants of EfficientNetV2 - B0, B1, B2, B3, and S. Experimental results indicate that the EfficientNetV2-B0 and B1 models achieved the highest accuracy, each reaching 97.14%. Among these, EfficientNetV2-B0 yielded higher precision and F1-score values. These findings suggest that integrating dilation layers into the EfficientNetV2 architecture can significantly enhance the performance of automated DR classification systems compared to the unmodified baseline models.
Downloads
Referensi
CHE, Z., BI, F., SUN, Y., XING, W., HUANG, H., dan ZHANG, X. 2023. Dilated residual FPN-based segmentation for mouse retinal images. Heliyon, 9(8), e18605. https://doi.org/10.1016/j.heliyon.2023.e18605
CHOIRUNISA, N. A., KARLITA, T., & ASMARA, R. 2021. Deteksi Ras Kucing Menggunakan Compound Model Scaling Convolutional Neural Network. Technomedia Journal, 6(2), 236–251. https://doi.org/10.33050/tmj.v6i2.1704
CITRA, E. E., FUDHOLI, D. H., dan DEWA, C. K. 2023. Implementasi Arsitektur EfficientNetV2 Untuk Klasifikasi Gambar Makanan Tradisional Indonesia. JURNAL MEDIA INFORMATIKA BUDIDARMA, 7(2), 766. https://doi.org/10.30865/mib.v7i2.5881
DAS, S., & SAHA, S. K. 2022. Diabetic retinopathy detection and classification using CNN tuned by genetic algorithm. Multimedia Tools and Applications, 81(6), 8007–8020. https://doi.org/10.1007/s11042-021-11824-w
DEWI, C., SANTOSO, A., INDRIATI, I., DEWI, N. A., & ARBAWA, Y. K. 2021. Evaluasi Performasi Ruang Warna pada Klasifikasi Diabetic Retinophaty Menggunakan Convolution Neural Network. Jurnal Teknologi Informasi Dan Ilmu Komputer, 8(3), 619–624. https://doi.org/10.25126/jtiik.2021834459
Ditjen P2P. 2024, January 11. Saatnya Mengatur Si Manis. Kemenkes Ditjen P2P. https://p2p.kemkes.go.id/saatnya-mengatur-si-manis/#:~:text=Menurut%20IDF%2C%20Indonesia%20menduduki%20peringkat,merupakan%20ibu%20dari%20segala%20penyakit.
EL-ATEIF, S., & IDRI, A. 2022. Single-modality and joint fusion deep learning for diabetic retinopathy diagnosis. Scientific African, 17. https://doi.org/10.1016/j.sciaf.2022.e01280
FU, Y., JU, Y., & ZHANG, D. 2024. MSEF-Net: A multi-scale EfficientNet Fusion for Diabetic Retinopathy grading. Biomedical Signal Processing and Control, 98, 106714. https://doi.org/10.1016/j.bspc.2024.106714
HU, H., YU, C., ZHOU, Q., GUAN, Q., & FENG, H. 2024. HDConv: Heterogeneous kernel-based dilated convolutions. Neural Networks, 179, 106568. https://doi.org/10.1016/j.neunet.2024.106568
JORDI DE LA TORRE, A. V. D. P. 2020. A deep learning interpretable classifier for diabetic retinopathy disease grading. Neurocomputing, 396, 365–476.
KARTHIK, R., VAICHOLE, T. S., KULKARNI, S. K., YADAV, O., & KHAN, F. 2022. Eff2Net: An efficient channel attention-based convolutional neural network for skin disease classification. Biomedical Signal Processing and Control, 73, 103406. https://doi.org/10.1016/j.bspc.2021.103406
LIU, R., CAI, W., LI, G., NING, X., & JIANG, Y. 2022. Hybrid Dilated Convolution Guided Feature Filtering and Enhancement Strategy for Hyperspectral Image Classification. IEEE Geoscience and Remote Sensing Letters, 19, 1–5. https://doi.org/10.1109/LGRS.2021.3100407
MADARAPU, S., ARI, S., & MAHAPATRA, K. 2024. A multi-resolution convolutional attention network for efficient diabetic retinopathy classification. Computers and Electrical Engineering, 117, 109243. https://doi.org/10.1016/j.compeleceng.2024.109243
MAJUMDAR, S., PRAMANIK, P., & SARKAR, R. 2023. Gamma function based ensemble of CNN models for breast cancer detection in histopathology images. Expert Systems with Applications, 213. https://doi.org/10.1016/j.eswa.2022.119022
MEI, H., WANG, Q., YU, L., & ZENG, Q. 2024. A deep learning-based algorithm for intelligent prediction of adverse geologic bodies in tunnels. Measurement Science and Technology, 35(9), 096119. https://doi.org/10.1088/1361-6501/ad4c8a
MUNTASA, A., WAHYUNINGRUM, R. T., HUSNI, SUGIARTI, A., YUSUF, M., MOTWAKEL, A., DEWI, D. A., ASMARA, Y. P., TUZZAHRA, Z., & MAHMUDI, W. F. 2025. EfficientNetV2-S Enhancement to Classify Acute Lymphoblastic Leukemia: Integrating Pre-Trained Models and Grid Search for Optimal Performance. International Journal of Intelligent Engineering and Systems, 18(1), 409–421. https://doi.org/10.22266/ijies2025.0229.30
PHRIDVIRAJ, M. S. B., BHUKYA, R., MADUGULA, S., MANJULA, A., VODITHALA, S., & WASEEM, M. S. 2023. A bi-directional Long Short-Term Memory-based Diabetic Retinopathy detection model using retinal fundus images. Healthcare Analytics, 3. https://doi.org/10.1016/j.health.2023.100174
QOMARIAH, D. U. N., TJANDRASA, H., & FATICHAH, C. 2019. Classification of Diabetic Retinopathy and Normal Retinal Images using CNN and SVM. 2019 12th International Conference on Information & Communication Technology and System (ICTS), 152–157. https://doi.org/10.1109/ICTS.2019.8850940
QUMMAR, S., KHAN, F. G., SHAH, S., KHAN, A., SHAMSHIRBAND, S., REHMAN, Z. U., KHAN, I. A., & JADOON, W. 2019. A Deep Learning Ensemble Approach for Diabetic Retinopathy Detection. IEEE Access, 7, 150530–150539. https://doi.org/10.1109/ACCESS.2019.2947484
RAIAAN, M. A. K., FATEMA, K., KHAN, I. U., AZAM, S., RASHID, M. R. U., MUKTA, M. S. H., JONKMAN, M., & DE BOER, F. 2023. A Lightweight Robust Deep Learning Model Gained High Accuracy in Classifying a Wide Range of Diabetic Retinopathy Images. IEEE Access, 11, 42361–42388. https://doi.org/10.1109/ACCESS.2023.3272228
RIZAL, S., IBRAHIM, N., PRATIWI, N. K. C., SAIDAH, S., & FU’ADAH, R. Y. N. 2020. Deep Learning untuk Klasifikasi Diabetic Retinopathy menggunakan Model EfficientNet. ELKOMIKA: Jurnal Teknik Energi Elektrik, Teknik Telekomunikasi, & Teknik Elektronika, 8(3), 693. https://doi.org/10.26760/elkomika.v8i3.693
ROY, S., TYAGI, M., BANSAL, V., & JAIN, V. 2022. SVD-CLAHE boosting and balanced loss function for Covid-19 detection from an imbalanced Chest X-Ray dataset. Computers in Biology and Medicine, 150, 106092. https://doi.org/10.1016/j.compbiomed.2022.106092
SHANG, Z., YU, C., HUANG, H., & LI, R. 2024. DCNet: A lightweight retinal vessel segmentation network. Digital Signal Processing, 153, 104651. https://doi.org/10.1016/j.dsp.2024.104651
SILVA-TINOCO, R., CABRERA-GERARDO, D., CASTILLO-MARTÍNEZ, L., & CUATECONTZI-XOCHITIOTZI, T. 2024. Diabetic retinopathy prevalence in Mexico: Results from a primary public health access initiative for screening in patients with type 2 diabetes. Diabetes Epidemiology and Management, 13. https://doi.org/10.1016/j.deman.2023.100182
TAN, M., & LE, Q. V. 2019. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. http://arxiv.org/abs/1905.11946
TAN, M., & LE, Q. V. 2021. EfficientNetV2: Smaller Models and Faster Training. http://arxiv.org/abs/2104.00298
TUMMALA, S., THADIKEMALLA, V. S. G., KADRY, S., SHARAF, M., & RAUF, H. T. 2023. EfficientNetV2 Based Ensemble Model for Quality Estimation of Diabetic Retinopathy Images from DeepDRiD. Diagnostics, 13(4). https://doi.org/10.3390/diagnostics13040622
XU, X., LIU, D., WANG, M., & LEI, M. 2023. Application of End-To-End EfficientNetV2 in Diabetic Retinopathy Grading. 153, 182–190. https://doi.org/10.1007/978-3-031-20738-9_22
YAZID, R. K., 2021. Pendeteksian Diabetik Retinopatik Menggunakan Convolutional Neural Network (CNN). Universitas Sriwijaya.
YE, Y., ZHOU, H., YU, H., HU, H., ZHANG, G., HU, J., & HE, T. 2022. An Improved EfficientNetV2 Model Based on Visual Attention Mechanism: Application to Identification of Cassava Disease. Computational Intelligence and Neuroscience, 2022. https://doi.org/10.1155/2022/1569911
YUDISTIRA, N., WIDODO, A. W., & RAHAYUDI, B. 2020. Deteksi Covid-19 pada Citra Sinar-X Dada Menggunakan Deep Learning yang Efisien. Jurnal Teknologi Informasi Dan Ilmu Komputer, 7(6), 1289. https://doi.org/10.25126/jtiik.2020763651
ZHANG, Y.-D., DONG, Z., CHEN, X., JIA, W., DU, S., MUHAMMAD, K., & WANG, S.-H. 2019. Image based fruit category classification by 13-layer deep convolutional neural network and data augmentation. Multimedia Tools and Applications, 78(3), 3613–3632. https://doi.org/10.1007/s11042-017-5243-3
Unduhan
Diterbitkan
Terbitan
Bagian
Lisensi
Hak Cipta (c) 2025 Jurnal Teknologi Informasi dan Ilmu Komputer
Artikel ini berlisensiCreative Commons Attribution-ShareAlike 4.0 International License.
Artikel ini berlisensi Creative Common Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)
Penulis yang menerbitkan di jurnal ini menyetujui ketentuan berikut:
- Penulis menyimpan hak cipta dan memberikan jurnal hak penerbitan pertama naskah secara simultan dengan lisensi di bawah Creative Common Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) yang mengizinkan orang lain untuk berbagi pekerjaan dengan sebuah pernyataan kepenulisan pekerjaan dan penerbitan awal di jurnal ini.
- Penulis bisa memasukkan ke dalam penyusunan kontraktual tambahan terpisah untuk distribusi non ekslusif versi kaya terbitan jurnal (contoh: mempostingnya ke repositori institusional atau menerbitkannya dalam sebuah buku), dengan pengakuan penerbitan awalnya di jurnal ini.
- Penulis diizinkan dan didorong untuk mem-posting karya mereka online (contoh: di repositori institusional atau di website mereka) sebelum dan selama proses penyerahan, karena dapat mengarahkan ke pertukaran produktif, seperti halnya sitiran yang lebih awal dan lebih hebat dari karya yang diterbitkan. (Lihat Efek Akses Terbuka).
