A computer-aided diagnosis system for plus disease in retinopathy of prematurity with structure adaptive segmentation...

Area of Science:

• Ophthalmology
• Medical Imaging
• Computational Biology

Background:

• Retinopathy of Prematurity (ROP) is a significant cause of childhood blindness in premature infants.
• Accurate diagnosis of ROP, particularly plus disease, is crucial for timely intervention and prevention of vision loss.
• Current diagnostic methods for ROP can be subjective and require specialized expertise.

Purpose of the Study:

• To develop and evaluate an automated computer-based system for objective assessment of plus disease in ROP.
• To improve the accuracy and efficiency of ROP diagnosis by analyzing retinal blood vessel features.
• To provide a tool that assists non-physician graders in identifying treatment-requiring ROP.

Main Methods:

• Development of a computer-based analysis system employing structure adaptive filtering, connectivity analysis, and image fusion for blood vessel segmentation.
• Extraction of novel retinal features, including leaf node count and vessel density, alongside traditional features like tortuosity and width.
• Automated selection of relevant vessels for feature extraction to mimic clinical diagnosis.

Main Results:

• The system achieved high classification accuracy for plus disease, with 95% sensitivity and 93% specificity.
• The proposed segmentation algorithm and novel vessel-based features demonstrated superiority in identifying plus disease.
• Automated feature extraction process enhances usability for non-specialist graders.

Conclusions:

• The developed computer-based system offers an objective and accurate method for assessing plus disease in ROP.
• The system's automated nature and high performance make it a valuable tool for ROP screening and diagnosis, especially where specialists are scarce.
• This technology has the potential to significantly reduce childhood blindness caused by ROP through improved diagnostic capabilities.