Impact of Optimizer Algorithm on NasNetMobile Model for Eight-class Retinal Disease Classification from OCT Images

Madhumithaa Selvarajan; Masoodhu Banu N. M

doi:10.35882/jeeemi.v8i2.1464

Madhumithaa Selvarajan Department of Biomedical Engineering, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Chennai, India https://orcid.org/0009-0006-9215-7989
Masoodhu Banu N. M Department of Biomedical Engineering, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Chennai, India https://orcid.org/0000-0002-7954-059X

DOI: https://doi.org/10.35882/jeeemi.v8i2.1464

Keywords: Retinal diseases, NasNetMobile model, Deep learning, SGD optimizer, Nadam optimizer, AdamW optimizer, GradCAM XAI

Abstract

Artificial intelligence (AI) is an emerging technology that plays a vital role in various fields, including the medical field. Ophthalmology is the earliest field to adopt AI for diagnosing several retinal diseases. Many imaging techniques are available, but Optical Coherence Tomography (OCT) is particularly useful for early-stage diagnosis. OCT is a non-invasive imaging method that offers high-resolution visualization of the retinal structure, aiding the ophthalmologist in differentiating between normal and abnormal retina. Automated OCT-based retinal disease classification using deep learning (DL) is important for early disease detection. Most DL models achieved high performance, but the influence of the optimizer on model behaviour, convergence, and explainability remains a challenge. To bridge the gap, this study evaluates the performance and convergence of five optimizers, such as RMSprop, AdamW, Adam, Nadam, and SGD, on the NasNetMobile model. The model was trained on the OCT-8 dataset, which comprises seven diseased retinal classes and one normal class of Optical Coherence Tomography (OCT) images. The seven diseases are Age-related Macular Degeneration (AMD), choroidal neovascularization (CNV), Central Serous retinopathy (CSR), diabetic macular edema (DME), diabetic retinopathy (DR), DRUSEN, and Macular Hole (MH). The study also analyzes convergence behaviour and explainability through early stopping regularization technique and GradCAM XAI, respectively. The model achieved 71%, 93%, 96%, 97%, and 97% of accuracy, respectively. Compared with other optimizers, the SGD optimizer achieved high accuracy in 22 epochs, which indicates better generalization. GradCAM XAI highlights the disease-relevant region across different retinal diseases. This framework emphasizes the significance of selecting an appropriate optimizer for robust retinal disease classification using a DL model trained on OCT images

Downloads

Download data is not yet available.

References

B. Das, “Sustainable global vision care,” Kerala Journal of Ophthalmology, vol. 34, no. 2, p. 92, 2022, doi: 10.4103/kjo.kjo_64_22.

S. G. Honavar, “Self-sustainable and inclusive eye care – Where equity meets excellence,” Indian J. Ophthalmol., vol. 70, no. 5, pp. 1441–1442, May 2022, doi: 10.4103/ijo.IJO_1008_22.

J. Mai and U. Schmidt-Erfurth, “Rolle der künstlichen Intelligenz bei verschiedenen retinalen Erkrankungen,” Klin. Monbl. Augenheilkd., vol. 241, no. 09, pp. 1023–1031, Sep. 2024, doi: 10.1055/a-2378-6138.

A. Choudhary, S. Ahlawat, S. Urooj, N. Pathak, A. Lay-Ekuakille, and N. Sharma, “A Deep Learning-Based Framework for Retinal Disease Classification,” Healthcare, vol. 11, no. 2, p. 212, Jan. 2023, doi: 10.3390/healthcare11020212.

D. S. Kermany et al., “Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning,” Cell, vol. 172, no. 5, pp. 1122-1131.e9, Feb. 2018, doi: 10.1016/j.cell.2018.02.010.

N. Rajagopalan, V. N., A. N. Josephraj, and S. E., “Diagnosis of retinal disorders from Optical Coherence Tomography images using CNN,” PLoS One, vol. 16, no. 7, p. e0254180, Jul. 2021, doi: 10.1371/journal.pone.0254180.

M. S, M. B. N.M, J. M. M, and A. P. Sarma, “Comparison of DL model for retinal diseases classification using OCT images,” in 2024 Tenth International Conference on Bio Signals, Images, and Instrumentation (ICBSII), IEEE, Mar. 2024, pp. 1–5. doi: 10.1109/ICBSII61384.2024.10562391.

A. Naik, B. S. Pavana, and K. Sooda, “Retinal Disease Classification from Retinal-OCT Images Using Deep Learning Methods,” 2022, pp. 95–104. doi: 10.1007/978-981-19-2347-0_8.

A. Kaur, V. Kukreja, M. Kumar, A. Choudhary, and R. Sharma, “RetDense: A Fine-tuned DenseNet121 Model for Retinal Eye Disease Detection,” in 2024 IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), IEEE, Mar. 2024, pp. 1–5. doi: 10.1109/IATMSI60426.2024.10502833.

S. Imran, A. L U, D. M S, T. Sharma, and T. Khanum, “Deep Learning-Based Classification of Retinal Diseases from OCT Images with LLM-Powered Patient Query Support,” Int. J. Health Sci. Res., vol. 15, no. 7, pp. 314–323, Jul. 2025, doi: 10.52403/ijhsr.20250738.

E. Hassan et al., “Enhanced Deep Learning Model for Classification of Retinal Optical Coherence Tomography Images,” Sensors, vol. 23, no. 12, p. 5393, Jun. 2023, doi: 10.3390/s23125393.

M. E. Subasi, S. Patnaik, and A. Subasi, “Optical coherence tomography image classification for retinal disease detection using artificial intelligence,” in Applications of Artificial Intelligence in Healthcare and Biomedicine, Elsevier, 2024, pp. 289–323. doi: 10.1016/B978-0-443-22308-2.00009-3.

J. Subhedar and A. Mahajan, “A Review on Recent Work On OCT Image Classification for Disease Detection,” in 2022 OPJU International Technology Conference on Emerging Technologies for Sustainable Development (OTCON), IEEE, Feb. 2023, pp. 1–6. doi: 10.1109/OTCON56053.2023.10114003.

Q. Chen et al., “Automated drusen segmentation and quantification in SD-OCT images,” Med. Image Anal., vol. 17, no. 8, pp. 1058–1072, Dec. 2013, doi: 10.1016/j.media.2013.06.003.

S. Saha et al., “Automated detection and classification of early AMD biomarkers using deep learning,” Sci. Rep., vol. 9, no. 1, Dec. 2019, doi: 10.1038/s41598-019-47390-3.

S. M. Waldstein, W. D. Vogl, H. Bogunovic, A. Sadeghipour, S. Riedl, and U. Schmidt-Erfurth, “Characterization of Drusen and Hyperreflective Foci as Biomarkers for Disease Progression in Age-Related Macular Degeneration Using Artificial Intelligence in Optical Coherence Tomography,” JAMA Ophthalmol., vol. 138, no. 7, pp. 740–747, Jul. 2020, doi: 10.1001/jamaophthalmol.2020.1376.

S. Akhtar et al., “A deep learning based model for diabetic retinopathy grading,” Sci. Rep., vol. 15, no. 1, p. 3763, Jan. 2025, doi: 10.1038/s41598-025-87171-9.

T. A. Ciulla, A. G. Amador, and B. Zinman, “Diabetic Retinopathy and Diabetic Macular EdemaPathophysiology, screening, and novel therapies,” Diabetes Care, vol. 26, no. 9, pp. 2653–2664, Sep. 2003, doi: 10.2337/DIACARE.26.9.2653.

A. Markan, A. Agarwal, A. Arora, K. Bazgain, V. Rana, and V. Gupta, “Novel imaging biomarkers in diabetic retinopathy and diabetic macular edema,” Ther. Adv. Ophthalmol., vol. 12, Jan. 2020, doi: 10.1177/2515841420950513.

S. A. Hassan, S. Akbar, A. Rehman, T. Saba, H. Kolivand, and S. A. Bahaj, “Recent Developments in Detection of Central Serous Retinopathy Through Imaging and Artificial Intelligence Techniques&x2013;A Review,” 2021, Institute of Electrical and Electronics Engineers Inc. doi: 10.1109/ACCESS.2021.3108395.

M. Subramanian, K. Shanmugavadivel, O. S. Naren, K. Premkumar, and K. Rankish, “Classification of Retinal OCT Images Using Deep Learning,” in 2022 International Conference on Computer Communication and Informatics, ICCCI 2022, Institute of Electrical and Electronics Engineers Inc., 2022. doi: 10.1109/ICCCI54379.2022.9740985.

H. Benhar, A. Idri, and J. L. Fernández-Alemán, “Data preprocessing for heart disease classification: A systematic literature review,” Comput. Methods Programs Biomed., vol. 195, p. 105635, Oct. 2020, doi: 10.1016/j.cmpb.2020.105635.

A. Soliman and J. Terstriep, “Keras Spatial,” in Proceedings of the 3rd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery, New York, NY, USA: ACM, Nov. 2019, pp. 69–76. doi: 10.1145/3356471.3365240.

L. Alzubaidi et al., “Review of deep learning: concepts, CNN architectures, challenges, applications, future directions,” J. Big Data, vol. 8, no. 1, p. 53, Mar. 2021, doi: 10.1186/s40537-021-00444-8.

L. Bottou, “Large-Scale Machine Learning with Stochastic Gradient Descent,” in Proceedings of COMPSTAT’2010, Heidelberg: Physica-Verlag HD, 2010, pp. 177–186. doi: 10.1007/978-3-7908-2604-3_16.

H. Robbins and S. Monro, “A Stochastic Approximation Method,” The Annals of Mathematical Statistics, vol. 22, no. 3, pp. 400–407, Sep. 1951, doi: 10.1214/aoms/1177729586.

R. Elshamy, O. Abu-Elnasr, M. Elhoseny, and S. Elmougy, “Improving the efficiency of RMSProp optimizer by utilizing Nestrove in deep learning,” Sci. Rep., vol. 13, no. 1, p. 8814, May 2023, doi: 10.1038/s41598-023-35663-x.

A. N. T. Kissiedu, G. K. Aggrey, M. G. Asante-Mensah, and A. Asante, “Development of Pneumonia Identification System: A Comparative Analysis of Some Selected CNN Architectures Using Adam, Nadam, and RAdam Optimizers,” in 2024 IEEE SmartBlock4Africa, IEEE, Sep. 2024, pp. 1–12. doi: 10.1109/SmartBlock4Africa61928.2024.10779552.

H. Azis, M. S. Andini, Herman, L. Syafie, A. R. Manga’, and L. B. Ilmawan, “A Comparative Study of SGD, Adam, AdamW, and RMSprop Optimizers for VGG19-Based Rupiah Banknote Classification,” in 2025 9th International Conference On Electrical, Electronics And Information Engineering (ICEEIE), IEEE, Sep. 2025, pp. 1–6. doi: 10.1109/ICEEIE66203.2025.11251942.

I. Khalil, A. Mehmood, H. Kim, and J. Kim, “OCTNet: A Modified Multi-Scale Attention Feature Fusion Network with InceptionV3 for Retinal OCT Image Classification,” Mathematics, vol. 12, no. 19, Oct. 2024, doi: 10.3390/math12193003.

A. M. Alenezi et al., “Multiscale attention-over-attention network for retinal disease recognition in OCT radiology images,” Front. Med. (Lausanne)., vol. 11, 2024, doi: 10.3389/fmed.2024.1499393.

N. Novely, S. Mahmud Shuvo, and Md. F. Faruk, “Improving Pre-Trained CNNs with CBAM and Skip Connections for Multi-Class Retinal Diseases Classification using OCT Images,” Association for Computing Machinery (ACM), Oct. 2024, pp. 946–953. doi: 10.1145/3723178.3723304.

Z. Ma, Q. Xie, P. Xie, F. Fan, X. Gao, and J. Zhu, “HCTNet: A Hybrid ConvNet-Transformer Network for Retinal Optical Coherence Tomography Image Classification,” Biosensors (Basel)., vol. 12, no. 7, Jul. 2022, doi: 10.3390/bios12070542.

P. Dutta, K. A. Sathi, M. A. Hossain, and M. A. A. Dewan, “Conv-ViT: A Convolution and Vision Transformer-Based Hybrid Feature Extraction Method for Retinal Disease Detection,” J. Imaging, vol. 9, no. 7, Jul. 2023, doi: 10.3390/jimaging9070140.

S. F. Rabbi, M. Al Mamun, M. F. Faruk, S. M. Mahedy Hasan, and A. Y. Srizon, “A Multi-branch and Attention Based CNN Architecture for the Classification of Retinal Diseases from OCT Images,” in 2023 International Conference on Information and Communication Technology for Sustainable Development, ICICT4SD 2023 - Proceedings, Institute of Electrical and Electronics Engineers Inc., 2023, pp. 36–40. doi: 10.1109/ICICT4SD59951.2023.10303384.

Q. Wang, K. Chen, W. Dou, and Y. Ma, “Cross-Attention Based Multi-Resolution Feature Fusion Model for Self-Supervised Cervical OCT Image Classification,” IEEE/ACM Trans. Comput. Biol. Bioinform., vol. 20, no. 4, pp. 2541–2554, Jul. 2023, doi: 10.1109/TCBB.2023.3246979.

A. Laouarem, C. Kara-Mohamed, E. B. Bourennane, and A. Hamdi-Cherif, “HTC-retina: A hybrid retinal diseases classification model using transformer-Convolutional Neural Network from optical coherence tomography images,” Comput. Biol. Med., vol. 178, Aug. 2024, doi: 10.1016/j.compbiomed.2024.108726.

Z. Shahzadi and M. Zubair, “Multiclass Classification of Retinal Disorders Using Optical Coherence Tomography Images,” in 2024 Horizons of Information Technology and Engineering (HITE), IEEE, Oct. 2024, pp. 1–6. doi: 10.1109/HITE63532.2024.10777173.

F.-E. Jannat, S. Gholami, M. N. Alam, and H. Tabkhi, “OCT-SelfNet: a self-supervised framework with multi-source datasets for generalized retinal disease detection,” Front. Big Data, vol. 8, Jul. 2025, doi: 10.3389/fdata.2025.1609124.

H. Pan et al., “A lightweight model for the retinal disease classification using optical coherence tomography,” Biomed. Signal Process. Control, vol. 101, p. 107146, Mar. 2025, doi: 10.1016/j.bspc.2024.107146.