No abstract

We applied new methods which take available knowledge of visual field physiology and pathophysiology into account, and employ modern computer-intensive mathematical methods for real time estimates of threshold values and threshold error estimates. In this way it was possible to design a family of testing algorithms which significantly reduced perimetric test time without any loss of quality in results.

We have devised a package for the statistical analysis of computerized visual fields. It is based on a new mathematical model of the normal visual field and intended to facilitate interpretation of single fields and to illustrate changes over time in consecutive threshold fields. Single field analyses include maps showing pointwise total and pattern deviations from the age-corrected normal reference field. These maps are displayed both numerically, in dB, and as noninterpolated greyscaled probability maps illustrating the statistical significance of measured deviations. These probability maps help emphasize shallow, but significant, depressions in the paracentral field while frequently occurring false positive deviations occurring in the midperiphery are de-emphasized. Visual field indices, summarizing the deviations of height (Mean Deviation) and shape (Pattern Standard Deviation and Corrected Pattern Standard Deviation) of the measured field are weighted according to the normal variance among healthy individuals and printed out together with level of statistical significance. For follow-up the programme contains several different options. These range from an Overview format where threshold printouts and probability maps from several tests are printed in reduced size, but without any reduction of data, on a single sheet of paper, to a box plot format where the development of the field is shown with an intermediate degree of data reduction and a format employing a high degree of data reduction: graphs over visual field indices over time. If five or more tests are available a linear regression analysis of Mean Deviation is automatically performed. The programme will become available in the Humphrey Field Analyzer.

ABSTRACT.Purpose: We developed a new family of test algorithms. This is an evaluation in normal subjects. Methods: One eye in each of twenty normal subjects, with a mean age of 37 years (range 26 to 59), was tested twice with each of the SITA, Full Threshold and Fastpac strategies of the Humphrey perimeter at 3 separate visits. Actual test times and number of stimulus exposures were compared. Test-retest variability and levels of threshold estimates were also calculated and compared between strategies. Results: In all subjects test times were shortest with SITA, 6.14 minutes in average, which was 50% as compared to Full Threshold (p∞0.001) with an average of 12.27 minutes, and a reduction of 16% as compared to Fastpac (mean 7.28 minutes, p∞0.0001). SITA required 287 stimulus exposures on the average, significantly fewer (p∞0.0001) than corresponding numbers with Full Threshold (mean of 404), and significantly more (p∞0.0001) than with Fastpac (average 240). SITA results showed significantly lower test-retest variability than results obtained with Fastpac (pΩ0.0002), and just as low as those of the Full Threshold strategy (pΩ0.0979). Threshold values obtained with SITA were slightly higher than those produced by the other two strategies. Conclusions:The results confirm those of previously reported simulated tests, that improved test algorithms using advanced visual field models and mathematical analyses performed in real time may effectively shorten computerized perimetry tests, while achieving the same or better test quality than today's standard methods.

No abstract