Motion Planning Under Uncertainty for Image-guided Medical Needle Steering.

Area of Science:

• Robotics
• Medical Device Technology
• Computational Geometry

Background:

• Traditional medical needles are stiff and have limited maneuverability.
• Steerable needles offer enhanced flexibility for reaching complex anatomical targets.
• Uncertainty in needle motion and tissue interaction poses a significant challenge for precise navigation.

Purpose of the Study:

• To develop a novel motion planning algorithm for steerable needles.
• To account for uncertainties in needle motion and tissue interaction.
• To maximize the probability of reaching clinical targets with steerable needles.

Main Methods:

• Developed a motion planning algorithm for a Dubins car variant with binary steering.
• Formulated the planning problem as a Markov Decision Process (MDP) using state-space discretization.
• Modeled motion uncertainty with probability distributions and used Dynamic Programming (DP) for optimal steering.
• Integrated image-based parameter extraction for fast computation and intra-operative use.

Main Results:

• The algorithm generates optimal steering actions to maximize target reach probability.
• It enables fast computation of optimal needle entry points.
• The method allows for intra-operative needle steering using pre-computed DP look-up tables.
• Demonstrated the importance of incorporating uncertainty into motion planning.

Conclusions:

• The developed motion planning algorithm effectively guides steerable needles to challenging targets.
• Explicitly modeling uncertainty improves the reliability of needle steering in soft tissues.
• This approach facilitates precise, image-guided interventions with flexible medical needles.