研究目的
Investigating the identification process of diabetes mellitus using a computer-based intelligent system to classify diabetic maculopathy stages from fundus images.
研究成果
The ANN classifier performs better than the PNN classifier with an accuracy of more than 96% of correct classification, and sensitivity of more than 96% and specificity of exactly 100%. The system's accuracy can further be improved with additional features and larger training datasets.
研究不足
The accuracy of the system could be improved with proper input features such as microaneurysms and hemorrhages, and by increasing the size of the training data.
1：Experimental Design and Method Selection:
The study uses morphological image processing techniques to extract features from fundus images and applies ANN and PNN classifiers for classification.
2：Sample Selection and Data Sources:
Fundus images from various ill persons are used, considering differences due to iris color, skin pigmentation, and other factors.
3：List of Experimental Equipment and Materials:
Not explicitly mentioned.
4：Experimental Procedures and Operational Workflow:
Includes pre-processing (RGB to grayscale conversion, intensity adjustment), image segmentation, detection of optic disc, and classification using ANN and PNN.
5：Data Analysis Methods:
The area covered by exudates in specific regions is considered for identifying maculopathy stages, with ANN and PNN classifiers used for classification.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
