DEIT-Based Bone Position and Orientation Estimation for Robotic Support in Total Knee Arthroplasty-A Computational...

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Orthopaedic Surgery

Background:

• Total knee arthroplasty (TKA) is a common procedure for end-stage knee osteoarthritis, often utilizing robotic systems for enhanced precision.
• Current methods for determining limb axis in TKA rely on pinned trackers, which can cause adverse events like bone fractures due to drilled holes.
• There is a need for non-invasive methods to accurately determine bone positions during TKA.

Purpose of the Study:

• To investigate the feasibility of using time differential electrical impedance tomography (TDIET) for non-invasively locating femur positions during TKA.
• To assess the accuracy of TDIET in determining bone axis and position compared to traditional methods.

Main Methods:

• A computational feasibility study was conducted using simulated data.
• Time differential electrical impedance tomography was employed to reconstruct conductivity distribution differences between two states (s0 and s1).
• Simulations included five configurations of thigh shape and tissue conductivity, with verification using cylinder and realistic femur cross-section models.

Main Results:

• For simplified cylinder models, the reconstructed bone center deviation was approximately 1 mm.
• For realistic femur cross-section models, the deviation in bone position ranged from 7.9 mm to 24.8 mm.
• The reconstructed bone axis deviated by approximately 1.50° from its actual position across all models.

Conclusions:

• Time differential electrical impedance tomography shows potential as a non-invasive technique for locating femur positions in TKA.
• The accuracy of TDIET requires further investigation, particularly for complex anatomical structures, to overcome limitations observed in realistic models.
• This method could offer a safer alternative to pinned trackers, potentially reducing complications in robotic-assisted TKA.