Scholarlink

Purpose <p>This study introduces a new method for classifying Diabetic Retinopathy (DR) with enhanced accuracy, focusing on addressing the limitations of existing approaches. DR classification is still a difficult task due to the presence of various complex and overlapping lesions in the retinal images. The primary goal is to develop an extremely effective model that can identify the different phases of diabetic retinopathy (DR) very accurately from the fundus images, therefore reducing the misclassification rates and helping in the very early diagnosis. The focus of our inquiry is on the determination of whether the new Hierarchical Auto-Associative Inception Polynomial Transformer Convolutional Neural Network with Snake Optimizer (HAutoAIPTCNNet + SO) will result in a significantly higher level of accuracy when compared with the existing methods.</p> Methods <p>Fundus images from the EyePACS and DIARETDB1 datasets are processed using Gradient Domain Guided Filtering (GDGF) for enhanced contrast, noise reduction, and normalization. Segmentation of DR-affected regions is achieved with the EfficientNet and Cascaded Visual Attention Network (ENet-CVAN) framework. The classification process then employs the Hierarchical Auto-Associative Inception Polynomial Transformer Convolutional Neural Network (HAutoAIPTCNNet) further refined through the Snake Optimizer (SO). The HAutoAIPTCNNet + SO model, developed in Python, features both hierarchical and polynomial transformations for the precise imaging of retinal characteristics.</p> Results <p>The suggested HAutoAIPTCNNet + SO approach reached a classification accuracy of 99.8% on the EyePACS and DIARETDB1 datasets, which was a notable improvement over current methods. The model also showed better specificity and lower error rates in differentiating DR stages. By effectively segmenting and classifying intricate retinal features, it minimized false positives and negatives, providing robust support for early DR diagnosis. The findings suggest that this model is a significant improvement for diabetic retinopathy screening high-accuracy.</p> Conclusion <p>The HAutoAIPTCNNet + SO model represents a substantial progress in diabetic retinopathy classification, integrating high-precision and low-error-rate segmentation and classification. By means of powerful feature extraction and optimization, it permits the accurate detection of DR in fundus images, overcoming the shortcomings that have been involved in the previous techniques. The excellent performance of the model indicates its suitability for the detection of early stages of DR with the consequent reduction of the risk of severe vision impairment by making timely intervention possible. These results strengthen the case for deep learning in medical imaging analysis for win–win clinical outcomes in DR management.</p>

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

Snake-optimized hierarchical polynomial transformer network for fundus-based diabetic retinopathy detection

  • 期刊论文

  • 发表日期：

    2026年4月29日

查看全文
  • D. Poornima,
  • M. Therasa,
  • Vinodpuri Rampuri Gosavi,
  • Shashank Shekhar Tiwari

摘要

Purpose

This study introduces a new method for classifying Diabetic Retinopathy (DR) with enhanced accuracy, focusing on addressing the limitations of existing approaches. DR classification is still a difficult task due to the presence of various complex and overlapping lesions in the retinal images. The primary goal is to develop an extremely effective model that can identify the different phases of diabetic retinopathy (DR) very accurately from the fundus images, therefore reducing the misclassification rates and helping in the very early diagnosis. The focus of our inquiry is on the determination of whether the new Hierarchical Auto-Associative Inception Polynomial Transformer Convolutional Neural Network with Snake Optimizer (HAutoAIPTCNNet + SO) will result in a significantly higher level of accuracy when compared with the existing methods.

Methods

Fundus images from the EyePACS and DIARETDB1 datasets are processed using Gradient Domain Guided Filtering (GDGF) for enhanced contrast, noise reduction, and normalization. Segmentation of DR-affected regions is achieved with the EfficientNet and Cascaded Visual Attention Network (ENet-CVAN) framework. The classification process then employs the Hierarchical Auto-Associative Inception Polynomial Transformer Convolutional Neural Network (HAutoAIPTCNNet) further refined through the Snake Optimizer (SO). The HAutoAIPTCNNet + SO model, developed in Python, features both hierarchical and polynomial transformations for the precise imaging of retinal characteristics.

Results

The suggested HAutoAIPTCNNet + SO approach reached a classification accuracy of 99.8% on the EyePACS and DIARETDB1 datasets, which was a notable improvement over current methods. The model also showed better specificity and lower error rates in differentiating DR stages. By effectively segmenting and classifying intricate retinal features, it minimized false positives and negatives, providing robust support for early DR diagnosis. The findings suggest that this model is a significant improvement for diabetic retinopathy screening high-accuracy.

Conclusion

The HAutoAIPTCNNet + SO model represents a substantial progress in diabetic retinopathy classification, integrating high-precision and low-error-rate segmentation and classification. By means of powerful feature extraction and optimization, it permits the accurate detection of DR in fundus images, overcoming the shortcomings that have been involved in the previous techniques. The excellent performance of the model indicates its suitability for the detection of early stages of DR with the consequent reduction of the risk of severe vision impairment by making timely intervention possible. These results strengthen the case for deep learning in medical imaging analysis for win–win clinical outcomes in DR management.