Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM69067
Volume:
10 |
11 |
Nov 2026
Pages:
1-8
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.

Computer Science and Engineering
KLS Vishwanathrao Deshpande Institute of Technology Haliyal

Computer Science and Engineering
KLS Vishwanathrao Deshpande Institute of Technology Haliyal

Computer Science and Engineering
KLS Vishwanathrao Deshpande Institute of Technology Haliyal

Computer Science and Engineering
KLS Vishwanathrao Deshpande Institute of Technology Haliyal

Computer Science and Engineering
KLS Vishwanathrao Deshpande Institute of Technology Haliyal
VisionCare is a connected prototype for preliminary retinal screening with deep learning. Retinal disease can end in irreversible blindness, yet cost and distance keep many people from reaching a specialist. The prototype provides first-level screening and lets patients read their own reports. It combines a React/Vite interface for healthcare workers, a FastAPI backend, OpenCV and Pillow for preprocessing, a PyTorch Vision Transformer (ViT-Base Patch16-224), PostgreSQL for records, ReportLab for PDF generation and a Flutter app for patients. A healthcare worker uploads a retinal image; the system scales it to 224 × 224 pixels, applies CLAHE contrast enhancement and a circular mask, and normalises the pixel values, after which the Vision Transformer assigns one of five classes: Normal, Diabetic Retinopathy, Glaucoma, Cataract or Age-related Macular Degeneration (AMD). The paper describes the architecture, algorithms, interfaces and an evaluation plan. Formal testing is incomplete and the proposed Raspberry Pi imaging unit has not been built into the system, so no accuracy, latency or clinical-validity figures are reported. The output is a first screening aid only and cannot stand in for a doctor’s diagnosis.
KEYWORDS
VisionCare connects a healthcare-worker interface, FastAPI services, image preprocessing, a five-class ViT, PostgreSQL records, PDF reports and Flutter patient access. This paper has set out its architecture, diagrams, algorithms, module responsibilities and evaluation plan. The Raspberry Pi device is not yet integrated, no test outcomes or model metrics exist, and clinical validity has not been shown. VisionCare is a screening-support prototype and not a diagnostic system. The next milestone is evaluation that can be traced across every software and hardware boundary. Future work will build and test the acquisition unit, verify the class mapping and preprocessing against the training code, run the software tests listed here and evaluate the model independently. Results must be reported separately: a web upload test does not validate camera capture, a model metric does not show that image acquisition is safe, and a working report does not prove clinical benefit. Any later clinical use requires qualified professional oversight and proper clinical validation. Until then, results remain preliminary and should direct patients to professional eye care.
1. V. Gulshan, L. Peng, M. Coram 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. D. S. W. Ting, L. R. Pasquale, L. Peng et al., “Artificial Intelligence and Deep Learning in Ophthalmology,” Br. J. Ophthalmol., vol. 103, no. 2, pp. 167–175, 2019. 3. Z. Zhou, X. Chen, E. Li, L. Zeng, K. Luo, and J. Zhang, “Edge Intelligence: Paving the Last Mile of Artificial Intelligence with Edge Computing,” Proc. IEEE, vol. 107, no. 8, 2019. 4. J. De Fauw, J. R. Ledsam, B. Romera-Paredes et al., “Clinically Applicable Deep Learning for Diagnosis and Referral in Retinal Disease,” Nat. Med., vol. 24, pp. 1342–1350, 2018. 5. P. Wankhede, T. Bhusari, V. Tammewar, R. Agrawal, C. Dhule, and N. C. Morris, “IoT and AI-ML Based Portable Non-Contact Retinal Imaging for Glaucoma and Eye Disease Monitoring at Home,” in Proc. 2025 12th Int. Conf. Emerging Trends in Engineering and Technology – Signal and Information Processing, Nagpur, India, 2025. 6. A. Dosovitskiy et al., “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,” in Proc. ICLR, 2021. 7. R. T. Fielding, “Architectural Styles and the Design of Network-Based Software Architectures,” Ph.D. dissertation, University of California, Irvine, 2000. 8. M. Jones, J. Bradley, and N. Sakimura, “JSON Web Token (JWT),” IETF RFC 7519, May 2015. 9. A. Paszke et al., “PyTorch: An Imperative Style, High-Performance Deep Learning Library,” in Advances in Neural Information Processing Systems, vol. 32, 2019. 10. React documentation. [Online]. Available: https://react.dev
Publication Date
11/17/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
1
Total views since publication
PDF Downloads
2
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...