研究目的
To propose a tool for diagnosing the stage and zones of Retinopathy of Prematurity (ROP) in infants easily, which can be used by nurses and paramedics without the need for a specialist.
研究成果
The developed tool successfully classifies the stage, zone, and severity of Retinopathy of Prematurity (ROP) in infants, facilitating early detection and treatment. It is user-friendly for nurses and paramedics, reducing the dependency on specialists. Future work includes integrating the tool into the 3D Retcam system for broader application and telemedicine purposes.
研究不足
The tool's effectiveness is dependent on the quality of the retinal images acquired by Retcam. Noise in images can affect the accuracy of edge detection and classification.
1：Experimental Design and Method Selection:
The tool uses image processing techniques in Matlab 2015a to analyze retinal images acquired by Retcam. It involves extracting the optic disc, marking the ridge, and calculating the distance of the optic nerve to classify the severity of ROP.
2：Sample Selection and Data Sources:
Retinal images of premature infants at risk of ROP were non-invasively acquired using Retcam.
3：List of Experimental Equipment and Materials:
Retcam for retinal imaging, Matlab 2015a for image processing.
4：Experimental Procedures and Operational Workflow:
The process includes image acquisition, preprocessing, binarization, thresholding, finding the centroid of the optic disc, calculating Euclidean distance to the ridge, classification of stages and zones, grading the severity, and displaying the results.
5：Data Analysis Methods:
The algorithm was validated on a set of pre-classified images to ensure accuracy in detecting the stage, zone, and severity of ROP.
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
