Toward FBG-Sensorized Needle Shape Prediction in Tissue Insertions.

Area of Science:

• Robotics and Automation
• Medical Device Engineering
• Surgical Planning

Background:

• Accurate prediction of flexible needle shape is crucial for minimally invasive surgery.
• Existing models struggle with non-uniform curvatures, limiting their clinical applicability.
• Fiber Bragg grating (FBG) sensors offer a promising approach for real-time shape sensing.

Purpose of the Study:

• To validate a theoretical method for predicting the 3D shape of flexible needles during insertion.
• To extend a previous sensor-based model by incorporating non-uniform curvatures.
• To evaluate the model's shape sensing and prediction accuracy in isotropic tissue.

Main Methods:

• A theoretical model combining FBG sensor curvature measurements and inextensible elastic rod mechanics was used.
• Experiments were conducted using a four-active area, FBG-sensorized needle.
• Stereo vision provided 3D ground truth data of the needle's shape in single-layer isotropic tissues.

Main Results:

• The model demonstrated effective 3D needle shape prediction capabilities.
• Mean root-mean-square errors for shape sensing and prediction were 0.479 mm and 0.892 mm, respectively.
• Validation was performed across varying single-layer isotropic tissue types.

Conclusions:

• The validated theoretical method provides a viable approach for predicting non-uniform curvatures in flexible needles.
• This advancement can enhance the precision and safety of surgical interventions involving flexible needles.
• The model shows potential for real-time shape sensing and prediction in clinical settings.