摘要
目的评价基于糖尿病视网膜病变(DR)图像数据库的临床应用技术,探讨其在临床DR早期诊断中的作用。方法采集1100例DR患者的眼底图像,经过初选和归纳分析后选择300只眼,建立微型DR图像数据库及应用软件技术。对100例糖尿病患者的眼底照相和荧光素眼底血管造影(FFA)检查图像进行人工直接诊断,并进行基于DR图像数据库技术的虚拟FFA诊断,对两者各层诊断符合情况进行统计分析,确定其诊断一致性。结果在DR患者不同分期的层次中,FFA检查诊断相符程度均大于80%,人工直接诊断和虚拟FFA诊断的总符合率为90%。结论虚拟FFA诊断与DR眼底照相诊断有较高的一致性,该图像数据库的建立和应用技术可指导和规范临床DR的早期诊疗,也为有创性FFA检查在DR诊治中合理应用提供依据和参考。
Objective To evaluate the image-based diabetic retinopathy（DR） database,and access its application in early diagnosis.Methods A total of 1 100 DR patients with fundus images were collected analyzed and summarized through the primaries and established 300 DR mini image database and software technology.The 100 images of diabetes fundus and fluorescein angiography（FFA） were direct diagnosed with manual.The DR from image database technology based on virtual FFA diagnosis was made,diagnostic accuracy of each layer using the 2 compliances were compared.Results The diagnosis consistent of FFA was more than 80 % in different stages of DR patients,the total compliance of manual and virtual FFA were 90 %.Conclusion It is demonstrated that the diagnosis of virtual FFA and DR fundus photography has high consistency,establishment and application of the image database could guide and standardize the early diagnosis and treatment in DR,and FFA in DR could provide a rational basis and reference for invasive diagnosis.
出处
《生物医学工程与临床》
CAS
2011年第2期163-168,共6页
Biomedical Engineering and Clinical Medicine
基金
西安市科技发展引导项目-科技创新与应用示范工程(SF1022)
