Magnification-Invariant Retinal Distance Estimation for Vitreoretinal Surgery Using a Laser Aiming Beam.

Area of Science:

• Robotics in Medicine
• Ophthalmic Surgery
• Surgical Navigation

Background:

• Robot-assisted retinal surgery requires precise tool-to-retina distance estimation for effective virtual fixtures.
• Retinal surface estimation is challenging due to limited features in microscope imagery and complex light paths.
• A laser aiming beam offers a detectable feature for inferring tool position.

Purpose of the Study:

• To develop a method for determining the distance between a surgical tool and the retina using a laser aiming beam.
• To ensure the method is independent of microscope magnification and robust to initial parameter errors.
• To enable accurate distance prediction for enhanced virtual fixture implementation in retinal surgery.

Main Methods:

• Utilized the area of a laser aiming beam, adjusted for tool width, to estimate retinal distance.
• Employed a dual Kalman filter integrating laser metric data with optical tracker information.
• The system updates state and parameters in parallel for robust distance prediction.

Main Results:

• Demonstrated the laser metric's independence from microscope magnification across various settings.
• Achieved median prediction errors of less than 100 μm for tool-to-retina distance.
• Validated the method's accuracy during freehand motion using a force sensor.

Conclusions:

• The laser aiming beam method accurately predicts surgical tool-to-retina distance, suitable for virtual fixtures in robot-assisted retinal surgery.
• The technique is independent of microscope magnification and initial parameter uncertainty.
• Future research will focus on in vivo adaptation and further error reduction.