Multispectral pattern recognition measures change in drusen area in age-related macular degeneration with high congruency to expert graders

Researchers at the Centre for Eye Health have analysed the change in drusen in eyes with intermediate AMD across two separate visits using a novel technique known as multispectral pattern recognition for en face retinal images from various imaging modalities. Multispectral pattern recognition change analysis was in 91% agreement in direction of drusen change and exhibited a sensitivity of 100% with a specificity of 73%. This was greater congruency than a commercially available change analysis tool, the Cirrus Advanced RPE analysis regardless of the direction of drusen change.

The study was originally conceived as drusen are a hallmark lesion of AMD and changes in their area and/or volume are strongly associated with disease progression. Assessment of longitudinal change in drusen size in clinical practice however is limited to a single commercial tool or manual inspection by clinicians. Thus, the aim of this study was to determine the ability of multispectral pattern recognition to accurately quantify change in drusen over time relative to expert graders and compare this to a commercially available, semi-automated drusen quantification method, the ‘Advanced RPE Analysis’ of the Cirrus SD-OCT.

Considering the association of drusen change and AMD progression, multispectral pattern recognition could be a highly useful tool in assessing drusen in clinical practice and other contexts such as clinical trials.

https://protect-au.mimecast.com/s/5CIGCYWL2MF4wXQU0r-e7?domain=nature.com