Fluoroscopy-Guided Robotic System for Transforaminal Lumbar Epidural Injections

Robot-Assisted Reduction of the Ankle Joint via Multi-Body 3D-2D Image Registration

Purpose

Robot-assisted orthopaedic joint reduction offers enhanced precision and control across multiple axes of motion, enabling precise realignment according to predefined plans. However, the high levels of forces encountered may induce unintended anatomical motion and flex mechanical components. To address this, this work presents an approach that uses 2D fluoroscopic imaging to verify and readjust the 3D reduction path by tracking deviations from the planned trajectory.

Methods

The proposed method involves a 3D-2D registration algorithm using a pair of fluoroscopic images, along with prior models of each body in the radiographic scene. This objective is formulated to couple and constrain multiple object poses (fibula, tibia, talus, and robot end effector), and incorporate novel methods for automatic view and hyperparameter selection to improve robustness. The algorithms were refined through cadaver studies and evaluated in a preclinical trial, employing a robotic system to manipulate a dislocated fibula.

Results

Studies with cadaveric specimens highlighted the joint-specific formulation's high registration accuracy (Δ x = 0.3 ± 1.5 mm ), further improved with the use of automatic view and hyperparameter selection (Δ x = 0.2 ± 0.8 mm ). Preclinical studies demonstrated a high deviation between the intended and the actual path of the robotic system, which was accurately captured (Δ x < 1 mm ) using the proposed techniques.

Conclusions

The solution offers to close the loop on image-based guidance of robot-assisted joint reduction by tracking the robot and bones to dynamically correct the course. The approach uses standard clinical images and is expected to lower radiation exposure by providing 3D information and allowing the staff to stay clear of the x-ray beam.

Advancements in Robotic-Assisted Spine Surgery

Applications and workflows around spinal robotics have evolved since these systems were first introduced in 2004. Initially approved for lumbar pedicle screw placement, the scope of robotics has expanded to instrumentation across different regions. Additionally, precise navigation can aid in tumor resection or spinal lesion ablation. Robot-assisted surgery can improve accuracy while decreasing radiation exposure, length of hospital stay, complication, and revision rates. Disadvantages include increased operative time, dependence on preoperative imaging among others. The future of robotic spine surgery includes automated surgery, telerobotic surgery, and the inclusion of machine learning or artificial intelligence in preoperative planning.

A Fully Differentiable Framework for 2D/3D Registration and the Projective Spatial Transformers

Image-based 2D/3D registration is a critical technique for fluoroscopic guided surgical interventions. Conventional intensity-based 2D/3D registration approa- ches suffer from a limited capture range due to the presence of local minima in hand-crafted image similarity functions. In this work, we aim to extend the 2D/3D registration capture range with a fully differentiable deep network framework that learns to approximate a convex-shape similarity function. The network uses a novel Projective Spatial Transformer (ProST) module that has unique differentiability with respect to 3D pose parameters, and is trained using an innovative double backward gradient-driven loss function. We compare the most popular learning-based pose regression methods in the literature and use the well-established CMAES intensity-based registration as a benchmark. We report registration pose error, target registration error (TRE) and success rate (SR) with a threshold of 10mm for mean TRE. For the pelvis anatomy, the median TRE of ProST followed by CMAES is 4.4mm with a SR of 65.6% in simulation, and 2.2mm with a SR of 73.2% in real data. The CMAES SRs without using ProST registration are 28.5% and 36.0% in simulation and real data, respectively. Our results suggest that the proposed ProST network learns a practical similarity function, which vastly extends the capture range of conventional intensity-based 2D/3D registration. We believe that the unique differentiable property of ProST has the potential to benefit related 3D medical imaging research applications. The source code is available at https://github.com/gaocong13/Projective-Spatial-Transformers.

Full-Perception Robotic Surgery Environment with Anti-Occlusion Global-Local Joint Positioning

The robotic surgery environment represents a typical scenario of human-robot cooperation. In such a scenario, individuals, robots, and medical devices move relative to each other, leading to unforeseen mutual occlusion. Traditional methods use binocular OTS to focus on the local surgical site, without considering the integrity of the scene, and the work space is also restricted. To address this challenge, we propose the concept of a fully perception robotic surgery environment and build a global-local joint positioning framework. Furthermore, based on data characteristics, an improved Kalman filter method is proposed to improve positioning accuracy. Finally, drawing from the view margin model, we design a method to evaluate positioning accuracy in a dynamic occlusion environment. The experimental results demonstrate that our method yields better positioning results than classical filtering methods.