Bone density optimized pedicle screw instrumentation improves screw pull-out force in lumbar vertebrae

SafeRPlan: Safe deep reinforcement learning for intraoperative planning of pedicle screw placement

Spinal fusion surgery requires highly accurate implantation of pedicle screw implants, which must be conducted in critical proximity to vital structures with a limited view of the anatomy. Robotic surgery systems have been proposed to improve placement accuracy. Despite remarkable advances, current robotic systems still lack advanced mechanisms for continuous updating of surgical plans during procedures, which hinders attaining higher levels of robotic autonomy. These systems adhere to conventional rigid registration concepts, relying on the alignment of preoperative planning to the intraoperative anatomy. In this paper, we propose a safe deep reinforcement learning (DRL) planning approach (SafeRPlan) for robotic spine surgery that leverages intraoperative observation for continuous path planning of pedicle screw placement. The main contributions of our method are (1) the capability to ensure safe actions by introducing an uncertainty-aware distance-based safety filter; (2) the ability to compensate for incomplete intraoperative anatomical information, by encoding a-priori knowledge of anatomical structures with neural networks pre-trained on pre-operative images; and (3) the capability to generalize over unseen observation noise thanks to the novel domain randomization techniques. Planning quality was assessed by quantitative comparison with the baseline approaches, gold standard (GS) and qualitative evaluation by expert surgeons. In experiments with human model datasets, our approach was capable of achieving over 5% higher safety rates compared to baseline approaches, even under realistic observation noise. To the best of our knowledge, SafeRPlan is the first safety-aware DRL planning approach specifically designed for robotic spine surgery.

Optimization of S2-alar-iliac screw (S2AI) fixation in adult spine deformity using a comprehensive genetic algorithm and finite element model personalized to patient geometry and bone mechanical properties

PURPOSE

To optimize the biomechanical performance of S2AI screw fixation using a genetic algorithm (GA) and patient-specific finite element analysis integrating bone mechanical properties.

METHODS

Patient-specific pelvic finite element models (FEM), including one normal and one osteoporotic model, were created from bi-planar multi-energy X-rays (BMEXs). The genetic algorithm (GA) optimized screw parameters based on bone mass quality (BM method) while a comparative optimization method maximized the screw corridor radius (GEO method). Biomechanical performance was evaluated through simulations, comparing both methods using pullout and toggle tests.

RESULTS

The optimal screw trajectory using the BM method was more lateral and caudal with insertion angles ranging from 49° to 66° (sagittal plane) and 29° to 35° (transverse plane). In comparison, the GEO method had ranges of 44° to 54° and 24° to 30° respectively. Pullout forces (PF) using the BM method ranged from 5 to 18.4 kN, which were 2.4 times higher than the GEO method (2.1-7.7 kN). Toggle loading generated failure forces between 0.8 and 10.1 kN (BM method) and 0.9-2.9 kN (GEO method). The bone mass surrounding the screw representing the fitness score and PF of the osteoporotic case were correlated (R2 > 0.8).

CONCLUSION

Our study proposed a patient-specific FEM to optimize the S2AI screw size and trajectory using a robust BM approach with GA. This approach considers surgical constraints and consistently improves fixation performance.

Pedicle screw path planning for multi-level vertebral fixation

BACKGROUND

For the spinal internal fixation procedures, connecting rods to the pedicle screws are commonly used in all spinal segments from the cervical to sacral spine. So far, we have only seen single vertebral screw trajectory planning methods in literatures. Joint screw placements in multi-level vertebrae with the constraint of an ipsilateral connecting rod are not considered.

PURPOSE

In this paper, a screw trajectory planning method that considers screw-rod joint system with both multi-level vertebral constraints and individual vertebral safety tolerance are proposed.

METHODS

The proposed method addresses three challenging constraints jointly for multi-level vertebral fixation with pedicle screws. First, a cylindrical screw safe passage model is suggested instead of a unique mathematical optimal trajectory for a single pedicle. Second, the flexible screw cap accessibility model is also included. Third, the connecting rod is modeled to accommodate the spine contour and support the needed gripping capacity. The retrospective clinical data of relative normal shape spines from Beijing Jishuitan hospital were used in the testing. The screw trajectories from the existing methods based on single vertebra and the proposed method based on multi-level vertebrae optimization are calculated and compared.

RESULTS

The results showed that the calculated screw placements by the proposed method can achieve 88% success rate without breaking the pedicle cortex and 100% in clinical class A quality (allow less than 2 mm out of the pedicle cortex) compared to 86.1% and 99.1%, respectively, with the existing methods. Expert evaluation showed that the screw path trajectories and the connecting rod calculated by the new method satisfied the clinical implantation requirements.

CONCLUSIONS

The new screw planning approach that seeks an overall optimization for multi-level vertebral fixation is feasible and more advantageous for clinical use than the single vertebral approaches.

Correlation of CT-based bone mineralization with drilling-force measurements in anatomical specimens is suitable to investigate planning of trans-pedicular spine interventions

This interdisciplinary study examined the relationship between bone density and drilling forces required during trans-pedicular access to the vertebra using fresh-frozen thoraco-lumbar vertebrae from two female body donors (A, B). Before and after biomechanical examination, samples underwent high-resolution CT-quantification of total bone density followed by software-based evaluation and processing. CT density measurements (n = 4818) were calculated as gray values (GV), which were highest in T12 for both subjects (GVmaxA = 3483.24, GVmaxB = 3160.33). Trans-pedicular drilling forces F (Newton N) were highest in L3 (FmaxB = 5.67 N) and L4 (FmaxA = 5.65 N). In 12 out of 13 specimens, GVs significantly (p < 0.001) correlated with force measurements. Among these, Spearman correlations r were poor in two lumbar vertebrae, fair in five specimens, and moderately strong in another five specimens, and highest for T11 (rA = 0.721) and L5 (rB = 0.690). Our results indicate that CT-based analysis of vertebral bone density acquired in anatomical specimens is a promising approach to predict the drilling force appearance as surrogate parameter of its biomechanical properties by e.g., linear regression analysis. The study may be of value as basis for biomechanical investigations to improve planning of the optimal trajectory and to define safety margins for drilling forces during robotic-assisted trans-pedicular interventions on the spine in the future.

Bone density optimized pedicle screw insertion

Background: Spinal fusion is the most common surgical treatment for the management of degenerative spinal disease. However, complications such as screw loosening lead to painful pseudoarthrosis, and are a common reason for revision. Optimization of screw trajectories to increase implant resistance to mechanical loading is essential. A recent optimization method has shown potential for determining optimal screw position and size based on areas of high bone elastic modulus (E-modulus). Aim: The aim of this biomechanical study was to verify the optimization algorithm for pedicle screw placement in a cadaveric study and to quantify the effect of optimization. The pull-out strength of pedicle screws with an optimized trajectory was compared to that of a traditional trajectory. Methods: Twenty-five lumbar vertebrae were instrumented with pedicle screws (on one side, the pedicle screws were inserted in the traditional way, on the other side, the screws were inserted using an optimized trajectory). Results: An improvement in pull-out strength and pull-out strain energy of the optimized screw trajectory compared to the traditional screw trajectory was only observed for E-modulus values greater than 3500 MPa cm3. For values of 3500 MPa cm3 or less, optimization showed no clear benefit. The median screw length of the optimized pedicle screws was significantly smaller than the median screw length of the traditionally inserted pedicle screws, p < 0.001. Discussion: Optimization of the pedicle screw trajectory is feasible, but seems to apply only to vertebrae with very high E-modulus values. This is likely because screw trajectory optimization resulted in a reduction in screw length and therefore a reduction in the implant-bone interface. Future efforts to predict the optimal pedicle screw trajectory should include screw length as a critical component of potential stability.

Improving pedicle screw path planning by vertebral posture estimation

Objective.Robot-assisted pedicle screw placement in spinal surgery can reduce the complications associated with the screw placement and reduce the hospital return counts due to malfunctions. However, it requires accurate planning for a high-quality procedure. The state-of-the-art technologies reported in the literature either ignore the anatomical variations across vertebrae or require substantial human interactions. We present an improved approach that achieves pedicle screw path planning through multiple projections of a numerically re-oriented vertebra with the estimated posture.Approach.We proposed an improved YOLO-type neural network model (YOLOPOSE3D) to estimate the posture of a vertebra before pedicle path planning. In YOLOPOSE3D, the vertebral posture is given as a rotation quaternion and 3D location coordinates by optimizing the intersection over union of the vertebra with the predicted posture and the actual posture. Then, a new local coordinate system is established for the vertebra based on the estimated posture. Finally, the optimal pedicle screw path trajectory is determined from the multiple projections of the vertebra in the local coordinates.Main results.The experimental results in difficult cases of scoliosis showed that the new YOLOPOSE3D network could accurately detect the location and posture of the vertebra with average translation and orientation errors as small as 1.55 mm and 2.55°. The screw path planning achieved 83.1% success rate without breaking the pedicle cortex for the lumbar vertebral L1-L5, which is better than that of a doctor's manual planning, 82.4%. With the clinical class A requirement to allow less than 2 mm out of the pedicle cortex, the success rate achieved nearly 100%.Significance.The proposed YOLOPOSED3D method can accurately determine the vertebral postures. With the improved posture prior, better clinical outcomes can be achieved for pedicle screw placement in spine internal fixation procedures.

Fast and versatile platform for pedicle screw insertion planning

PURPOSE

Computer-assisted surgical planning methods help to reduce the risks and costs in transpedicular fixation surgeries. However, most methods do not consider the speed and versatility of the planning as factors that improve its overall performance. In this work, we propose a method able to generate surgical plans in minimal time, within the required safety margins and accounting for the surgeon's personal preferences.

METHODS

The proposed planning module takes as input a CT image of the patient, initial-guess insertion trajectories provided by the surgeon and a reduced set of parameters, delivering optimal screw sizes and trajectories in a very reduced time frame.

RESULTS

The planning results were validated with quantitative metrics and feedback from surgeons. The whole planning pipeline can be executed at an estimated time of less than 1 min per vertebra. The surgeons remarked that the proposed trajectories remained in the safe area of the vertebra, and a Gertzbein-Robbins ranking of A or B was obtained for 95 % of them.

CONCLUSIONS

The planning algorithm is safe and fast enough to perform in both pre-operative and intra-operative scenarios. Future steps will include the improvement of the preprocessing efficiency, as well as consideration of the spine's biomechanics and intervertebral rod constraints to improve the performance of the optimisation algorithm.

Discussion on the possibility of multi-layer intelligent technologies to achieve the best recover of musculoskeletal injuries: Smart materials, variable structures, and intelligent therapeutic planning

Although intelligent technologies has facilitated the development of precise orthopaedic, simple internal fixation, ligament reconstruction or arthroplasty can only relieve pain of patients in short-term. To achieve the best recover of musculoskeletal injuries, three bottlenecks must be broken through, which includes scientific path planning, bioactive implants and personalized surgical channels building. As scientific surgical path can be planned and built by through AI technology, 4D printing technology can make more bioactive implants be manufactured, and variable structures can establish personalized channels precisely, it is possible to achieve satisfied and effective musculoskeletal injury recovery with the progress of multi-layer intelligent technologies (MLIT).