Keratoconus is defined by focal rather than uniform corneal change. Yet many established methods for measuring the cornea’s response to an air puff examine only the apex or a single meridian. Even systems that sample several regions sequentially may struggle with a transient event unfolding over milliseconds: by the time the scanner reaches a new location, the cornea has already changed.
Researchers from the International Centre for Translational Eye Research (ICTER) in Warsaw, Poland, have now developed a swept-source OCT platform designed to remove that temporal mismatch. The system tracks air-puff-induced deformation simultaneously at nine corneal locations – one central and eight peripheral – with an effective temporal resolution of 10 µs.
The approach combines space-division multiplexing with depth encoding. Each probe beam is assigned a different optical path length, placing its signal within a separate depth window of a single OCT acquisition. This allows all nine locations to be recorded using one detector channel, without the spatial and temporal mis-registration associated with sequential scanning.
The peripheral spots were positioned 1.1 mm from the corneal apex, based on biomechanical modelling of healthy and keratoconic corneas. From the resulting displacement-versus-time profiles, the investigators extracted several deformation parameters. Their principal analysis focused on the first displacement amplitude, which provided the strongest combination of directional consistency and discrimination.
To summarize the spatial response, the team introduced an “asymmetry vector.” Displacement differences between four pairs of opposing peripheral spots are converted into directional vectors and combined. The final vector’s magnitude indicates the strength of the asymmetry, while its angle points towards the region of greater deformation – potentially reflecting localized thinning, weakening or both.
The system was first tested in an ex vivo porcine model, in which one corneal region was softened enzymatically to mimic focal ectasia. Although the conventional polar plot contained multiple displacement maxima, the asymmetry vector pointed towards the treated region. Its measured orientation was 318°, closely matching the 321° position of the softened area.
The investigators then examined 13 healthy volunteers and three patients with keratoconus. Keratoconic eyes displayed substantially longer asymmetry vectors than the healthy example, and the vectors generally pointed towards the cone regions identified on commercial pachymetry, tangential curvature and posterior elevation maps. The correspondence was directional rather than exact – an expected distinction because topography describes geometry, whereas the air-puff response reflects a mixture of geometry, intraocular pressure and tissue mechanics.
Repeat measurements in two healthy eyes produced angular standard deviations of 2.7–3.7° and normalized magnitude errors of 5–8%. Temporal resolution also proved critical. Simulated downsampling from 10 µs to 250 µs generated vector-magnitude errors of up to 20% and angular errors of 17.5°. At 1 ms, errors approached 70% and 47.5°, respectively.
The study remains a technical proof of concept. Only three keratoconic eyes were included, displacement profiles were segmented manually, and the metric does not directly measure intrinsic stiffness or viscoelasticity. Precise alignment is also essential.
Nevertheless, the platform demonstrates how simultaneous sampling may expose biomechanical patterns missed by apex-only or sequential methods. With automated segmentation and validation in larger cohorts, its nine-point view could support keratoconus detection, treatment monitoring and the study of subtle, localized corneal change.
