Simulations for FASTPAC and the standard 4-2 dB full-threshold strategy of the Humphrey Field Analyzer

Area of Science:

• Ophthalmology
• Visual Science
• Medical Technology

Background:

• Automated perimetry is crucial for diagnosing and monitoring visual field defects.
• Standard 4-2 dB full threshold strategy is widely used but time-consuming.
• New strategies aim to improve efficiency without compromising diagnostic accuracy.

Purpose of the Study:

• To evaluate the accuracy, reproducibility, and efficiency of FASTPAC, a novel 3-dB step automated perimetry strategy.
• To compare FASTPAC against the standard 4-2 dB full threshold strategy using computer simulations.

Main Methods:

• Development of an "artificial patient" module using Monte-Carlo simulations.
• 10,200 simulations conducted with threshold values from 0 to 50 dB.
• Comparison of FASTPAC (3-dB steps) and the standard 4-2 dB full threshold strategy.

Main Results:

• FASTPAC reduced the number of presentations per threshold determination by 18%, decreasing testing time.
• Both strategies showed a strong influence of starting deviation on threshold error.
• FASTPAC exhibited a more pronounced relationship between starting deviation and threshold error (slope 0.18) compared to the 4-2 dB strategy (slope 0.13).
• Threshold fluctuations increased with absolute starting deviations, particularly for FASTPAC.

Conclusions:

• The starting value significantly impacts threshold determination in both strategies, potentially leading to underestimation of visual field defects.
• FASTPAC offers time savings but at the cost of reduced accuracy and reliability compared to the 4-2 dB strategy.
• The increased fluctuation and sensitivity to starting values are intrinsic properties of FASTPAC's staircase procedure.