advertisement
Editors Selection IGR 8-4
Imaging: Structure and function
Comment by Ronald Harwerth on:
15093 Structure and function in glaucoma: The relationship between a functional visual field map and an anatomic retinal map, Strouthidis NG; Vinciotti V; Tucker AJ et al., Investigative Ophthalmology and Visual Science, 2006; 47: 5356-5362
Measurement variability is a major concern in the use of standard clinical perimetry for diagnosis and assessment of progression in glaucoma. A general strategy to reduce measurement noise in digital image processing is called spatial filtering, and it is reasonable to adapt a similar strategy to visual field testing. Strouthidis et al. (1093), developed spatial filters that are specific to perimetry data. With each spatial filter, the sensitivity of a given test location will be a weighted sum of the sensitivities of all test locations included in the filter and, thus, the first issue involved the determination of which test locations should be included and the weighting of their individual sensitivities. The data for designating empirical filters were based on two parameters. The first was the angular distance, at the optic nerve head, between the axons of retinal ganglion cells that enter the optic nerve from each of a pair of perimetric test locations. The second parameter was the angular retinal distance between test locations on the retina. It is not clear why two distance parameters were used or that the anatomic map is sufficiently precise to map each retinal test location onto the optic nerve, but functional correlations for either parameter were significant, with higher functional correlations for smaller distances. In the final model, both of the angular distances and the product of the two were incorporated in the best-fit derived from multiple regression analysis. The utility of the model was demonstrated by its application to the large Moorfields Eye Hospital dataset of visual fields, for which the predictions accounted for 75% of the variance for functional correlations between test locations. The specific spatial filters for each test location were comprised of all locations that had high functional correlations with the nominal location which, for the published example, incorporated contiguous test locations. Overall, the study presents an interesting empiric approach to an important problem in clinical perimetry, but whether these spatial filters will improve the signal-to-noise for clinical perimetry must be determined by clinical trials.
Comments
The comment section on the IGR website is restricted to WGA#One members only. Please log-in through your WGA#One account to continue.Log-in through WGA#One
