Scholarlink

Retinal degeneration is a major cause of visual impairment, significantly reducing quality of life and imposing economic burdens, particularly in low-income communities. Genetic predisposition, environmental factors, and traumatic injuries contribute to progressive retinal diseases, which, if undiagnosed, can lead to irreversible vision loss. The prevalence of retinal degeneration is rising, especially among younger populations in India, with a higher incidence in Southern states. Financial limitations further delay diagnosis and treatment, worsening the condition in rural areas. Deep learning has shown great potential in medical imaging for automated detection and classification of retinal diseases. This study develops DL architectures to diagnose retinal degeneration, incorporating advanced image preprocessing and segmentation techniques. The models are evaluated using benchmark datasets and Indian retinal images, optimized through hyperparameter tuning. By leveraging AI-driven approaches, this research aims to enable early diagnosis, enhance accessibility to retinal care, and mitigate the socioeconomic impact of vision impairment.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Investigations on Retinal Degeneration Detection and Pathology Classification Using Various Multilayer Deep Learning Architectures

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • N. B. S. Venkatesh,
  • Chinmayee Dora,
  • Viswanadham Ravuri

摘要

Retinal degeneration is a major cause of visual impairment, significantly reducing quality of life and imposing economic burdens, particularly in low-income communities. Genetic predisposition, environmental factors, and traumatic injuries contribute to progressive retinal diseases, which, if undiagnosed, can lead to irreversible vision loss. The prevalence of retinal degeneration is rising, especially among younger populations in India, with a higher incidence in Southern states. Financial limitations further delay diagnosis and treatment, worsening the condition in rural areas. Deep learning has shown great potential in medical imaging for automated detection and classification of retinal diseases. This study develops DL architectures to diagnose retinal degeneration, incorporating advanced image preprocessing and segmentation techniques. The models are evaluated using benchmark datasets and Indian retinal images, optimized through hyperparameter tuning. By leveraging AI-driven approaches, this research aims to enable early diagnosis, enhance accessibility to retinal care, and mitigate the socioeconomic impact of vision impairment.