Intraoperative fusion of models and data for robust distance sensing

Area of Science:

• Microsurgery
• Surgical Robotics
• Optical Sensing

Background:

• Instrument-integrated optical sensors are crucial for precise microsurgery.
• Accurate instrument-to-tissue distance measurement is key for instrument control.
• Obstructions in the optical path can lead to significant measurement errors.

Purpose of the Study:

• To develop a robust method for improving distance information from sensorized microsurgical instruments.
• To overcome measurement errors caused by obstructions in the optical path.
• To enhance the reliability of instrument-integrated optical sensors in microsurgery.

Main Methods:

• Integration of a rapid search algorithm for identifying neighboring data points.
• Utilizing geometric and non-geometric techniques to model local tissue structure.
• Implementing sensor fusion combining measurement and model-based approaches to identify and mitigate disturbances.

Main Results:

• Simulations demonstrated a mean absolute error of less than 0.02 mm using the local spline fit.
• Experiments in ex vivo human eyes showed up to an 89% reduction in error compared to sensor-only methods.
• The developed pipeline effectively identifies and overcomes disturbances like obstructions and segmentation errors.

Conclusions:

• The proposed method significantly enhances the reliability of instrument-integrated optical sensors.
• This advancement enables distance-based instrument control in challenging surgical conditions, improving precision in ophthalmic procedures.
• The approach is generalizable to various surgical applications involving sensorized instruments.