Evaluation of a New Algorithm-Based Approach in Subjective Refraction

Area of Science:

• Ophthalmology
• Optometry
• Medical Technology

Background:

• Subjective refraction is crucial for determining precise optical correction.
• Traditional manual refraction relies on expert skill and can have inter-examiner variability.
• New algorithm-based systems aim to standardize and improve the accuracy of subjective refraction.

Purpose of the Study:

• To evaluate a novel algorithm-based subjective refraction system, SiviewExam Expert.
• To compare its refractive outcomes against those of a professional expert using a standardized manual method.
• To assess the system's performance on a large and diverse patient population.

Main Methods:

• Prospective comparison of two subjective refraction methods: manual expert refraction and the SiviewExam Expert system.
• Inclusion of 107 patients with a wide range of refractive errors (mean SE -2.10 ± 3.37 D).
• Analysis of outcomes using power vector analysis (M, J0, J45) and comparison of limits of agreement with established inter-examiner variability.

Main Results:

• No statistically significant bias was observed between the SiviewExam Expert system and expert manual refraction for M, J0, and J45 components.
• The limits of agreement for M, J45, and J0 were significantly lower than the expected inter-examiner variability (0.7 D).
• The SiviewExam Expert system's generated report was deemed accurate by the expert in 100% of cases.

Conclusions:

• The SiviewExam Expert system demonstrates comparable clinical accuracy to expert manual subjective refraction.
• The algorithm-based approach offers a reliable method for subjective refraction across a broad spectrum of ametropias.
• This technology has the potential to standardize and enhance the efficiency of refractive error assessment.