研究目的
To estimate the risk of suffering from glaucoma in patients through automatic analysis of OCT images by calculating nerve fiber and ganglionary vessels areas.
研究成果
The proposal allows us to obtain a correct segmentation of the RNFL layer and calculate its thickness which is a parameter for the diagnosis of the disease in order to compare it with a set of reference patients. The obtained results are valid in most of the images where the correct segmentation and thickness of the fibers can be obtained.
研究不足
One of the most important is related to the calculation of ROI areas (in comparison with the values provided by the equipment used to obtain the OCT images).
1：Experimental Design and Method Selection:
The system uses computer vision techniques and the Newton Interpolation Method to separate and analyze OCT images.
2：Sample Selection and Data Sources:
30 OCT images from patients with different diagnoses (healthy, at risk, or with glaucoma) were used.
3：List of Experimental Equipment and Materials:
OCT machine for image acquisition, computer for processing.
4：Experimental Procedures and Operational Workflow:
Steps include image acquisition, color conversion to HSV, filtering (median and Gaussian blur), segmentation using erode and dilation operations and bandpass filters, ROI splitting with Newton interpolation, and area calculation.
5：Data Analysis Methods:
Kernel Density Estimation (KDE) for area analysis and expert feedback on criteria like pixel loss and accuracy.
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
