Fig 1.
Most of the 234 test points were arranged temporally within a 13° × 19° area with a central focus on (16°, -2°) in 1° steps. We determined that consecutive non-response areas represented the blind spot.
Fig 2.
Relationship between head-tilt angle and S-OCR amplitude.
The amplitude of the s-OCR tended to be large as the head-tilt angle increased in both eyes and in either direction of head tilting. This relationship was indicated by sinusoidal regression. The amplitudes of the s-OCRs differed among the subjects.
Table 1.
Coefficient of determination (R2) and Akaike’s information criterion (AIC) for linear and sinusoidal regression models.
Table 2.
Between-individual SDs at every head-tilt angle, indicating inter-individual variability.
Table 3.
Within-individual SDs at every head-tilt angle, indicating intra-individual variability.
Fig 3.
Blind spot detection results of a 33-year-old male subject.
The barycenter of the blind spot rotated by head tilting. At a 50° head tilt to the right, the barycenter rotated in a counterclockwise direction. At a 50° head tilt to the left, the barycenter rotated in a clockwise direction.
Fig 4.
Relationship between head-tilt angle and rotation angle of the blind spot.
The blind spot rotated in the direction opposite to the head tilt in all subjects at all head-tilt angles. The rotation angle of the blind spot tended to be large as the head-tilt angle increased in both eyes. This relationship showed sinusoidal regression.
Fig 5.
The correlation between S-OCR amplitude and rotation angle of the blind spot.
These values were strongly correlated in both eyes (right eye: R2 = 0.94, p<0.0001, left eye: R2 = 0.94, p<0.0001), which strongly suggested that the rotated VF was correlated with the s-OCR amplitude in perimetry with imo.
Table 4.
Acceptable maximum VF rotation angle based on the degree of eccentricity and the head-tilt angle that generated the S-OCR amplitude equal to the VF rotation angle.