Low-Power FPGA-Based Diabetic Retinopathy Detection Using Fundus Image Processing and Deep Learning

Paper ID: EIJTEM_2025_12_3_104-111

Author's Name: Dr.Venkata Reddy Adama, Dr. K Venugopal Rao, Dr.G.Koteswar Rao

Volume: 12

Issue: 3

Year: 2025

Page No: 104-111

Abstract:

Diabetic Retinopathy (DR) is a major retinal complication caused by diabetes and can lead to permanent vision loss if it is not detected at an early stage. Manual screening of fundus images requires expert ophthalmologists and may be time-consuming in large-scale healthcare environments. Image processing and deep learning techniques have shown strong potential for automated DR detection by identifying abnormal retinal features such as microaneurysms, hemorrhages, exudates, and blood vessel irregularities. However, conventional software-based deep learning systems require high computational resources, large memory, and significant power consumption, which limits their use in portable real-time medical screening devices. This paper proposes a low-power FPGA-based diabetic retinopathy detection system using fundus image processing and deep learning. The proposed architecture includes fundus image acquisition, preprocessing, retinal feature enhancement, lesion segmentation, lightweight CNN-based classification, and FPGA-based hardware acceleration. Fundus images are preprocessed using resizing, contrast enhancement, green-channel extraction, noise filtering, and normalization. Important retinal abnormalities are enhanced and passed to a compact deep learning model for classification. The CNN model is optimized using quantization, fixed-point representation, pipelining, parallel processing, and memory reuse for efficient FPGA implementation. Performance evaluation shows improved processing speed, reduced power consumption, and reliable classification performance compared with conventional CPU-based processing. The proposed system is suitable for low-cost, real-time, and portable DR screening applications.

Keywords: Diabetic Retinopathy, Fundus Image Processing, FPGA, Deep Learning, CNN, Low-Power Architecture, Retinal Disease Detection, Medical Image Analysis, Real-Time Processing, Edge AI.

References:

[1]. V. Gulshan et al., “Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs,” JAMA, vol. 316, no. 22, pp. 2402–2410, 2016.
[2]. H. Pratt, F. Coenen, D. M. Broadbent, S. P. Harding, and Y. Zheng, “Convolutional neural networks for diabetic retinopathy,” Procedia Computer Science, vol. 90, pp. 200–205, 2016.
[3]. M. D. Abràmoff et al., “Improved automated detection of diabetic retinopathy on a publicly available dataset through integration of deep learning,” Investigative Ophthalmology & Visual Science, vol. 57, no. 13, pp. 5200–5206, 2016.
[4]. T. Shanthi and R. S. Sabeenian, “Modified AlexNet architecture for classification of diabetic retinopathy images,” Computers & Electrical Engineering, vol. 76, pp. 56–64, 2019.
[5]. K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2016, pp. 770–778.
[6]. S. Mittal, “A survey of FPGA-based accelerators for convolutional neural networks,” Neural Computing and Applications, vol. 32, pp. 1109–1138, 2020.
[7]. M.D. Javeed, B. Hari Krishna, M.V.S.R Kishore, “Low Power Viterbi Decoder Design for TCM Decoders Using T-Algorithm”, ciiT International Journal of Programmable Device Circuits and Systems, vol. 4, no. 15,2012, pp. 764-768.
https://www.ciitresearch.org/dl/index.php/pdcs/article/view/PDCS112012001Google Scholar
[8]. K. Abdelouahab, M. Pelcat, J. Serot, and F. Berry, “Accelerating CNN inference on FPGAs: A survey,” arXiv preprint arXiv:1806.01683, 2018.
[9]. J. M. Rabaey, A. Chandrakasan, and B. Nikolic, Digital Integrated Circuits: A Design Perspective, 2nd ed. Upper Saddle River, NJ, USA: Prentice Hall, 2003.
[10]. Mohammad Javeed and B. Swapna, “High Speed Convolution Encoding and Viterbi Decoding Using Dynamic Shift Register,” International Journal of Electronic and Electrical Engineering, vol. 7, no. 5, pp. 483–490, 2014.
https://ijerece.com/abstract.php?id=8974&issue=Issue5
[11]. Mohammad Javeed, Gella Ravikanth " Design and Implementation of 64 Bit Multiplier By Using Carry Save Adder ", International Journal of Industrial Electronics and Electrical Engineering (IJIEEE) , Volume-2,Issue-11 (Nov, 2014)
https://ijieee.org.in/paper_detail.php?paper_id=1450&name=Design_And_Implementation_of_64_Bit_Multiplier_By_Using_Carry_Save_Adder
[12]. Mohammad Javeed ,Syed Shaheen ,Md Thahaseen , (2015) " Design Of SRAM Memory In Jpeg 2000 Image Compression With Multi Bit Flip Flops " , International Journal of Advance Computational Engineering and Networking (IJACEN) , pp. 7-10, Volume-3, Issue-2
https://ijacen.iraj.in/paper_detail.php?paper_id=1672&name=Design_Of_SRAM_Memory_In_Jpeg_2000_Image_Compression_With_Multi_Bit_Flip_Flops
[13]. M. D. Abràmoff, M. K. Garvin, and M. Sonka, “Retinal imaging and image analysis,” IEEE Reviews in Biomedical Engineering, vol. 3, pp. 169–208, 2010.
[14]. Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, pp. 436–444, 2015.
[15]. D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning representations by back-propagating errors,” Nature, vol. 323, pp. 533–536, 1986.

View PDF