Robotic assistance for ultrasound guided prostate brachytherapy
Transrectal ultrasound (TRUS) guided brachytherapy is an effective treatment for low-risk prostate cancer, but many implants continue to fail or cause adverse side effects. The procedure entails permanently implanting radioactive seeds into the prostate. It is commonly believed that pinpoint accuracy in executing a pre-operative implant plan should lead to good dosimetry. However, as two decades of practice have demonstrated, this is not achieved by all clinicians. Instead of enforcing a pre-operative plan, intra-operative dosimetry and in-situ optimization have been receiving increasing attention. This approach, however, demands precise localization of the implanted needles and seeds, which assumes exquisite spatial and temporal synchronization of the needle insertion and imaging tasks. The needles and seeds can be localized in TRUS, the dose field analyzed, and finally the remainder of the implant can be optimized. Needle positions are often rearranged to avoid overdosing and/or seeds added to fill cold spots. This, however, is a repetitive manual process that is prone to human operator errors and consumes valuable time in the operating room. Time delays may allow for increased edema that may change the anatomy and thereby negatively impact outcome. It is expected that intra-operative dosimetry can resolve these problems. This function, however, requires spatial and temporal synchronization of the actions of imaging, needle insertion, and needle/seed tracking.
The robotic assistant provides needle placement accuracy equivalent to that of conventional templates while offering much greater flexibility, owing to its biaxial needle angulation and continuum Cartesian needle spacing. It is a digitally encoded system that allows for synchronized imaging and image-based needle/seed tracking, thereby opening the way for online dosimetry and implant optimization. These features were achieved without causing interference with established clinical hardware, workflow, or calibration standards. This is especially important as commercial potential and clinical viability in contemporary medicine are inseparable issues. Engineering development will continue with motorizing the TRUS base which already performs optical encoding of the stepper, making such a process relatively straightforward. Note that the system is functional without such motorization of the TRUS probe, though it requires some degree of manual adjustment during needle insertion and seed release, which from the dosimetric point of view is only an issue of convenience.
Needle steering for prostate brachytherapy
The efficacy of Brachytherapy depends on the accuracy of seed placement. However, prostate motion and deformation, limited maneuverability of the needle due to the presence of the template and lack of sufficient visual feedback hinder accurate seed delivery.
Robotic assistant systems can replace the guiding template, accommodate insertion with arbitrary angles, provide higher resolution of insertion point and hence, compensate for prostate deformation. Although, robotic insertion systems are available, they are not used in clinical settings due to lack of a needle steering algorithm which can accommodate the clinical limitations such as lack of real time visual feedback.
Prostate deformation models can be used to estimate the initial insertion parameters. Biomechanic-based prostate models require patient specific parameters such as elasticity modulus and friction force which are not readily available.
We propose a statistical image-based model to predict the prostate deformation during puncture of the prostate capsule. This model will be constructed using pre- and intra-operatively acquired ultrasound images of the prostate. This model can be used to set the initial insertion parameters and also modify the needle path during the insertion using a robot.