Scholarlink

<p>Diabetic retinopathy (DR) detection focuses on identifying and diagnosing retinal damage caused by prolonged high blood sugar levels in individuals with diabetes. This condition affects the delicate blood vessels in the retina, leading to potential complications such as vision impairment and even blindness if left untreated. Early detection and timely intervention are essential to prevent severe vision loss and improve patient outcomes.Existing models face challenges in handling the noise and image quality variations, and dealing with high-dimensional data. Nevertheless, the existing techniques fail to extract the highly nonlinear featuresoften resulting in misdetections.Hence, the Volans Finch Optimization-based Quantum Convolutional Neural Network (VF-QCNN) is proposed for effective DR detection. The VF-QCNN encodes retinal image data into quantum states, leveraging quantum superposition and entanglement for efficient parallel processing, which enhances the extraction of intricate features crucial for identifying early Diabetic Retinopathy indicators.More specifically, the Volans Finch optimization inherits the adaptive learning and memory-based search capabilities to reach theoptimal solution, ensuring optimal tuning of network parameters and improving the segmentation of retinal structures. Consequently, the VF-QCNN model captures and preserves subtle spatial and textural patterns during diagnosis, leading to improved accuracy and reliability. The proposed model achieves impressive performance metrics, with an accuracy of 96.47%, an F1 score of 96.33%, a precision of 95.27%, and a recall of 97.43%.</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

VF-QCNN: Diabetic Retinopathy Detection Using Volans Finch Optimization-Based Quantum Convolutional Neural Network Model

  • 期刊论文

  • 发表日期：

    2025年7月11日

查看全文
  • L. S. Kalkonde,
  • K. N. Kasat,
  • Chetan K. Videkar,
  • K. S. Kalkonde,
  • D. S.Chandak

摘要

Diabetic retinopathy (DR) detection focuses on identifying and diagnosing retinal damage caused by prolonged high blood sugar levels in individuals with diabetes. This condition affects the delicate blood vessels in the retina, leading to potential complications such as vision impairment and even blindness if left untreated. Early detection and timely intervention are essential to prevent severe vision loss and improve patient outcomes.Existing models face challenges in handling the noise and image quality variations, and dealing with high-dimensional data. Nevertheless, the existing techniques fail to extract the highly nonlinear featuresoften resulting in misdetections.Hence, the Volans Finch Optimization-based Quantum Convolutional Neural Network (VF-QCNN) is proposed for effective DR detection. The VF-QCNN encodes retinal image data into quantum states, leveraging quantum superposition and entanglement for efficient parallel processing, which enhances the extraction of intricate features crucial for identifying early Diabetic Retinopathy indicators.More specifically, the Volans Finch optimization inherits the adaptive learning and memory-based search capabilities to reach theoptimal solution, ensuring optimal tuning of network parameters and improving the segmentation of retinal structures. Consequently, the VF-QCNN model captures and preserves subtle spatial and textural patterns during diagnosis, leading to improved accuracy and reliability. The proposed model achieves impressive performance metrics, with an accuracy of 96.47%, an F1 score of 96.33%, a precision of 95.27%, and a recall of 97.43%.