Surface-based registration of CT images to physical space for image-guided surgery of the spine: a sensitivity study

Computer-Guided Navigation System Efficiency Evaluation Using Surgical Instruments for Spinal Fusion

BACKGROUND

Navigation surgical systems have been widely used in spinal fusion to ensure accuracy and safety during pedicle screw insertion.

METHODS

The research was performed under laboratory conditions, using stereotactic navigation, surgical instruments for spinal fusion, development of additional devices and software. During the experiments, all stages of the computed tomography-guided navigation system use were performed-preoperative preparation of patient data and planning to provide visual control of the navigation of surgical instruments during the insertion of screws.

RESULTS

The 20 screws were inserted into the vertebrae of the lumbar spine phantom with an average pedicle width of 12.6 ± 1.29 mm with 100% clinical accuracy. The 30 screws were inserted into cadaveric sheep vertebrae with an average pedicle width of 6.56 ± 0.58 mm with 96.67% efficiency.

CONCLUSIONS

The proposed navigation technique of the CT-guided navigation system is highly effective in the navigation process of surgical instruments and pedicle screws during spinal fusion.

Ultrasound-based navigated pedicle screw insertion without intraoperative radiation: feasibility study on porcine cadavers

BACKGROUND

Navigation systems for spinal fusion surgery rely on intraoperative computed tomography (CT) or fluoroscopy imaging. Both expose patient, surgeons and operating room staff to significant amounts of radiation. Alternative methods involving intraoperative ultrasound (iUS) imaging have recently shown promise for image-to-patient registration. Yet, the feasibility and safety of iUS navigation in spinal fusion have not been demonstrated.

PURPOSE

To evaluate the accuracy of pedicle screw insertion in lumbar and thoracolumbar spinal fusion using a fully automated iUS navigation system.

STUDY DESIGN

Prospective porcine cadaver study.

METHODS

Five porcine cadavers were used to instrument the lumbar and thoracolumbar spine using posterior open surgery. During the procedure, iUS images were acquired and used to establish automatic registration between the anatomy and preoperative CT images. Navigation was performed with the preoperative CT using tracked instruments. The accuracy of the system was measured as the distance of manually collected points to the preoperative CT vertebral surface and compared against fiducial-based registration. A postoperative CT was acquired, and screw placements were manually verified. We report breach rates, as well as axial and sagittal screw deviations.

RESULTS

A total of 56 screws were inserted (5.50 mm diameter n=50, and 6.50 mm diameter n=6). Fifty-two screws were inserted safely without breach. Four screws (7.14%) presented a medial breach with an average deviation of 1.35±0.37 mm (all <2 mm). Two breaches were caused by 6.50 mm diameter screws, and two by 5.50 mm screws. For vertebrae instrumented with 5.50 mm screws, the average axial diameter of the pedicle was 9.29 mm leaving a 1.89 mm margin in the left and right pedicle. For vertebrae instrumented with 6.50 mm screws, the average axial diameter of the pedicle was 8.99 mm leaving a 1.24 mm error margin in the left and right pedicle. The average distance to the vertebral surface was 0.96 mm using iUS registration and 0.97 mm using fiducial-based registration.

CONCLUSIONS

We successfully implanted all pedicle screws in the thoracolumbar spine using the ultrasound-based navigation system. All breaches recorded were minor (<2 mm) and the breach rate (7.14%) was comparable to existing literature. More investigation is needed to evaluate consistency, reproducibility, and performance in surgical context.

CLINICAL SIGNIFICANCE

Intraoperative US-based navigation is feasible and practical for pedicle screw insertion in a porcine model. It might be used as a low-cost and radiation-free alternative to intraoperative CT and fluoroscopy in the future.

A Novel Precise Optical Navigation System for Craniomaxillofacial Surgery Registered With an Occlusal Splint

BACKGROUND

An augmented reality tool allows visual tracking of real anatomical structures and superimposing virtual images, so it can be used for navigation of important structures during surgery.

OBJECTIVES

The authors have developed a new occlusal splint-based optical navigation system for craniomaxillofacial surgery. In this study, the authors aim to measure the accuracy of the system and further analyze the main factors influencing precision.

METHODS

Ten beagle dogs were selected and a three-dimensional model was established through computed tomography scanning, dental model making, and laser scanning, and then registration was performed according to the tooth marking points. The bilateral mandibular osteotomy was performed on Beagle dogs under navigation system based on the occlusal splint. The left side was taken to compare the deviation between the preoperative plan and the surgical results, and the accuracy of distance and angle and the stability of the system were analyzed.

RESULTS

The average position deviation between the preoperative design and intraoperative navigation was: 0.01 ± 0.73 mm on the lateral height of the mandibular ramus, 0.26 ± 0.57 mm on the inner height of the mandibular ramus, and 0.20 ± 0.51 mm on the osteotomy length. The average angle deviation is 0.94° ± 1.38° on the angle between the mandibular osteotomy plane and ramus plane and 0.66° ± 0.97° on the angle of the retained mandibular angle. And most of the data showed good consistency.

CONCLUSIONS

In summary, the accuracy of the system can meet clinical requirements and can be used as a useful tool to improve the accuracy of craniomaxillofacial surgery.

The state-of-the-art in ultrasound-guided spine interventions

During the last two decades, intra-operative ultrasound (iUS) imaging has been employed for various surgical procedures of the spine, including spinal fusion and needle injections. Accurate and efficient registration of pre-operative computed tomography or magnetic resonance images with iUS images are key elements in the success of iUS-based spine navigation. While widely investigated in research, iUS-based spine navigation has not yet been established in the clinic. This is due to several factors including the lack of a standard methodology for the assessment of accuracy, robustness, reliability, and usability of the registration method. To address these issues, we present a systematic review of the state-of-the-art techniques for iUS-guided registration in spinal image-guided surgery (IGS). The review follows a new taxonomy based on the four steps involved in the surgical workflow that include pre-processing, registration initialization, estimation of the required patient to image transformation, and a visualization process. We provide a detailed analysis of the measurements in terms of accuracy, robustness, reliability, and usability that need to be met during the evaluation of a spinal IGS framework. Although this review is focused on spinal navigation, we expect similar evaluation criteria to be relevant for other IGS applications.

Deep learning-based liver segmentation for fusion-guided intervention

PURPOSE

Tumors often have different imaging properties, and there is no single imaging modality that can visualize all tumors. In CT-guided needle placement procedures, image fusion (e.g. with MRI, PET, or contrast CT) is often used as image guidance when the tumor is not directly visible in CT. In order to achieve image fusion, interventional CT image needs to be registered to an imaging modality, in which the tumor is visible. However, multi-modality image registration is a very challenging problem. In this work, we develop a deep learning-based liver segmentation algorithm and use the segmented surfaces to assist image fusion with the applications in guided needle placement procedures for diagnosing and treating liver tumors.

METHODS

The developed segmentation method integrates multi-scale input and multi-scale output features in one single network for context information abstraction. The automatic segmentation results are used to register an interventional CT with a diagnostic image. The registration helps visualize the target and guide the interventional operation.

RESULTS

The segmentation results demonstrated that the developed segmentation method is highly accurate with Dice of 96.1% on 70 CT scans provided by LiTS challenge. The segmentation algorithm is then applied to a set of images acquired for liver tumor intervention for surface-based image fusion. The effectiveness of the proposed methods is demonstrated through a number of clinical cases.

CONCLUSION

Our study shows that deep learning-based image segmentation can obtain useful results to help image fusion for interventional guidance. Such a technique may lead to a number of other potential applications.

Vector field analysis for surface registration in computer-assisted ENT surgery

BACKGROUND

Manual paired-point registration for navigated ENT-surgery is prone to human errors; automatic surface registration is often caught in local minima.

METHODS

Anatomical features of the human occiput are integrated into an algorithm for surface registration. A vector force field is defined between the patient and operating room datasets; registration is facilitated through gradient-based vector field analysis optimization of an energy function. The method is validated exemplarily on patient surface data provided by a mechanically positioned A-mode ultrasound sensor.

RESULTS

Successful registrations were achieved within the entire parameter space, as well as from positions of local minima that were found by the Gaussian fields algorithm for surface registration. Sub-millimetric registration error was measured in clinically relevant anatomical areas on the anterior skull and within the generally accepted margin of 1.5 mm for the entire head.

CONCLUSION

The satisfactory behavior of this approach potentially suggests a wider clinical integration.

Improved Surface-Based Registration of CT and Intraoperative 3D Ultrasound of Bones

The intraoperative registration of preoperative CT volumes is a key process of most computer-assisted orthopedic surgery (CAOS) systems. In this work, is reported a new method for automatic registration of long bones, based on the segmentation of the bone cortical in intraoperative 3D ultrasound images. A bone classifier was developed based on features, obtained from the principal component analysis of the Hessian matrix, of every voxel in an intraoperative ultrasound volume. 3D freehand ultrasound was used for the acquisition of the intraoperative ultrasound volumes. Corresponding bone surface segmentations in ultrasound and preoperative CT imaging were used for the intraoperative registration. Validation on a phantom of the tibia produced encouraging results, with a maximum mean segmentation error of 0.34⁡mm (SD=0.26⁡mm) and a registration accuracy error of 0.64⁡mm (SD=0.49⁡mm).

Segmental Surface Referencing during Intraoperative Three-dimensional Image-Guided Spine Navigation: An Early Validation with Comparison to Automated Referencing

Study Design Interventional human cadaver study. Objective Intraoperative three-dimensional (3-D)-guided navigation improves spine instrumentation accuracy. However, image acquisition may need to be repeated with segment hypermobility or distant target from reference frame (RF). The current study evaluates the usefulness of internal metal fiducials (IMFs) as surface references in enhancing registration accuracy and avoiding repeating imaging. Methods Six fresh-frozen cadaveric human torsos were utilized. Posterior C1-T2 exposure was done, and three IMFs were inserted per level; intraoperative 3-D images were then acquired. Two registration methods were utilized: autoregistration (AR, group 1) and point registration using IMF (IMFR, group 2). Registration accuracy was checked by identifying IMFs in both groups. Pedicle screws inserted into C2, C4, C5, and C7 based on the two registration methods (three cadavers each) with RF on C7 and then on C2. Results The mean registration error was lower with IMFR compared with AR (0.35 ± 0.5 mm versus 2.02 ± 0.85 mm, p = 0.0001). Overall, 34 pedicle screws were inserted (AR, 18; IMFR, 16). Final screw placement was comparable using both techniques (p = 0.58). Lateral screws violations were observed in four IMFR screws (1 to 2 mm) as compared with five in AR group (2 to 3 mm). Reregistration after moving RF to C2 was possible using surface screws in IMFR group, thus avoiding new 3-D image acquisition. Conclusion During intraoperative 3-D navigation in spine procedures, surface fiducial registration using IMF provided superior accuracy over automated registration. It allowed repeat registration without repeating radiation during long spine segment instrumentations. More studies are needed to clarify both practical and clinical application of this method.

Generalized iterative most likely oriented-point (G-IMLOP) registration

PURPOSE

The need to align multiple representations of anatomy is a problem frequently encountered in clinical applications. A new algorithm for feature-based registration is presented that solves this problem by aligning both position and orientation information of the shapes being registered.

METHODS

The iterative most likely oriented-point (IMLOP) algorithm and its generalization (G-IMLOP) to the anisotropic noise case are described. These algorithms may be understood as probabilistic variants of the popular iterative closest point (ICP) algorithm. A probabilistic model provides the framework, wherein both position information and orientation information are simultaneously optimized. Like ICP, the proposed algorithms iterate between correspondence and registration subphases. Efficient and optimal solutions are presented for implementing each subphase of the proposed methods.

RESULTS

Experiments based on human femur data demonstrate that the IMLOP and G-IMLOP algorithms provide a strong accuracy advantage over ICP, with G-IMLOP providing additional accuracy improvement over IMLOP for registering data characterized by anisotropic noise. Furthermore, the proposed algorithms have increased ability to robustly identify an accurate versus inaccurate registration result.

CONCLUSION

The IMLOP and G-IMLOP algorithms provide a cohesive framework for incorporating orientation data into the registration problem, thereby enabling improvement in accuracy as well as increased confidence in the quality of registration outcomes. For shape data having anisotropic uncertainty in position and/or orientation, the anisotropic noise model of G-IMLOP enables further gains in registration accuracy to be achieved.

Registration accuracy in computer-assisted pelvic surgery

INTRODUCTION

An in vitro study was performed to assess the global registration accuracy of a computer-assisted system in pelvic orthopaedic surgery. The system was applied to a putative tumor resection in a pelvic sawbone.

METHODS

Twenty landmarks were created on the surface of the pelvis, and a virtual model of the sawbone was constructed based on surface extraction from computed tomography. The coordinates of the landmarks were defined in the CT-scan coordinate system, and registration of the sawbone with the virtual model was achieved using a surface-based matching algorithm. The landmarks were considered as control points, and deviations between their physical locations and their locations in the virtual model were calculated, thereby quantifying the global accuracy error.

RESULTS

The location of the initialization points was unimportant. The dynamic reference base gave the best results when placed far from the working area. Accuracy was improved when the sampling area was increased, but was decreased by its excessive expansion.

CONCLUSIONS

It is recommended that the DRB be located on the contralateral side of the pelvis. Extending the approach posteriorly and including the entire working area in the sampling surface area, if possible, will also help increase accuracy in computer-assisted pelvic surgery.

Bone alignment using the iterative closest point algorithm

Computer assisted surgical interventions and research in joint kinematics rely heavily on the accurate registration of three-dimensional bone surface models reconstructed from various imaging technologies. Anomalous results were seen in a kinematic study of carpal bones using a principal axes alignment approach for the registration. The study was repeated using an iterative closest point algorithm, which is more accurate, but also more demanding to apply. The principal axes method showed errors between 0.35 mm and 0.49 mm for the scaphoid, and between 0.40 mm and 1.22 mm for the pisiform. The iterative closest point method produced errors of less than 0.4 mm. These results show that while the principal axes method approached the accuracy of the iterative closest point algorithm in asymmetrical bones, there were more pronounced errors in bones with some symmetry. Principal axes registration for carpal bones should be avoided.

Image-based navigation improves the positioning of the humeral component in total elbow arthroplasty

HYPOTHESIS

Implant alignment in total elbow arthroplasty (TEA) is a challenging and error-prone process using conventional techniques. Identification of the flexion-extension (FE) axis is further complicated for situations of bone loss. This study evaluated the accuracy of humeral component alignment in TEA. We hypothesized that an image-based navigation system would improve humeral component positioning, with navigational errors less than or approaching 2.0 mm and 2.0 degrees .

MATERIALS AND METHODS

Implantation of a modified commercial TEA humeral component was performed with and without navigation on 11 cadaveric distal humeri. Navigated alignment was based on positioning the humeral component with the aid of a computed tomography (CT)-based preoperative plan registered to landmarks on the distal humerus. Alignment was performed under 2 scenarios of bone quality: (1) an intact distal humerus, and (2) a distal humerus without articular landmarks.

RESULTS

Navigation significantly improved implant alignment accuracy (P < .001). Navigated implant alignment was 1.2 +/- 0.3 mm in translation and 1.3 degrees +/- 0.3 degrees in rotation for the intact scenario. For the bone loss scenario, navigated alignment error was 1.1 +/- 0.5 mm and 2.0 degrees +/- 1.3 degrees . Non-navigated alignment was 3.1 +/- 1.3 mm and 5.0 degrees +/- 3.8 degrees for the intact scenario and 3.0 +/- 1.6 mm and 12.2 degrees +/- 3.3 degrees for the bone loss scenario.

DISCUSSION

Image-based navigation improves the accuracy and reproducibility of humeral component placement in TEA. Implant alignment errors for the navigated alignments were below the target of 2.0 degrees and 2 mm that is considered standard for most navigation systems. Non-navigated implant alignment error was significantly greater for the bone loss scenario compared with the intact scenario.

CONCLUSIONS

Implant malalignment may increase the likelihood of early implant wear, instability, and loosening. Improved implant positioning will likely lead to fewer complications and greater prosthesis longevity.

Determining the operative line of resection for image-guided emphysema surgery using a laser scanner and non-rigid registration

BACKGROUND

Although many diseases such as emphysema are diagnosed with preoperative imaging modalities, this information is rarely utilized in the operating room. A method that relates the preoperative images to the non-rigid organ in physical space would aid a surgeon to determine the line of resection.

METHODS

We used a three-dimensional (3D) laser scanner to obtain intraoperative images of the lung and overlayed it with preoperative CT images, using a non-rigid image registration method.

RESULTS

The non-overlapping registration error of the system was 1.91 +/- 0.28% without organ deformation and 2.69 +/- 0.28% with 9% organ deformation. When 83% of the organ was visible, the registration error was 2.99 +/- 0.42%.

CONCLUSION

A novel image overlay system using a 3D laser scanner and a non-rigid registration method was implemented and its accuracy evaluated. By using the proposed system, we successfully related the preoperative images with an open organ in the operating room.

Comparison of navigation accuracy in THA between the mini-anterior and -posterior approaches

BACKGROUND

The accuracy of a CT-based hip navigation might depend on surgical approaches, resulting in varying accuracy of implant alignment.

METHODS

We performed primary cementless total hip arthroplasty (THA) with mini-incision surgery (MIS) to 40 well-matched patients (anterior or posterior approaches, 20 hips each), using navigation with surface registration. We investigated cup alignment using postoperative computed tomography (CT) and compared the navigation accuracy between the two approaches, i.e. the difference between intra-operative and postoperative alignments of the cup.

RESULTS

There was no significant difference between the two approaches. The mean navigation accuracies in abduction and anteversion were 2.0 degrees (SD 1.4 degrees) and 2.7 degrees (SD 1.9 degrees), respectively, in the anterior approach, and 2.4 degrees (SD 2.0 degrees) and 2.0 degrees (SD 1.4 degrees), respectively, in the posterior approach. All cup alignments were within 10 degrees of the target orientation.

CONCLUSIONS

This CT-based navigation for MIS-THA provides navigation accuracy without significant differences between the two approaches and with favourable alignment of the cup.

Image guided oral implantology and its application in the placement of zygoma implants

The application of zygoma implants proposes a successful treatment for functional reconstruction of maxillary defects. However, the placement of zygoma implants is not without risk due to anatomically complex operation sites. Aiming at minimizing the risks and improving the precision of the surgery, an image guided oral implantology system (IGOIS) is presented in this study to transfer the preoperative plan accurately to the operating theatre. The principle of IGOIS is introduced in detail, including the framework, 3D-reconstruction, preoperative planning, registration, and the motion tracking algorithm. The phantom experiment shows that fiducial registration error (FRE) and TRE (target registration error) of IGOIS are, respectively, 1.12mm and 1.35mm. With respect to the overall accuracy, the average distance deviations at the coronal and apical point of the implant are, respectively, 1.36+/-0.59mm and 1.57+/-0.59mm, while average angle deviation between the axes of the planned and the actual implant is 4.1 degrees +/-0.9 degrees . A clinical report for a patient with a severely atrophic maxilla demonstrates that the major advantage of this computer-aided navigation technology lies in its accuracy, reliability, and flexibility.

Feasibility study for image-guided kidney surgery: assessment of required intraoperative surface for accurate physical to image space registrations

A notable complication of applying current image-guided surgery techniques of soft tissue to kidney resections (nephrectomies) is the limited field of view of the intraoperative kidney surface. This limited view constrains the ability to obtain a sufficiently geometrically descriptive surface for accurate surface-based registrations. The authors examined the effects of the limited view by using two orientations of a kidney phantom to model typical laparoscopic and open partial nephrectomy views. Point-based registrations, using either rigidly attached markers or anatomical landmarks as fiducials, served as initial alignments for surface-based registrations. Laser range scanner (LRS) obtained surfaces were registered to the phantom's image surface using a rigid iterative closest point algorithm. Subsets of each orientation's LRS surface were used in a robustness test to determine which parts of the surface yield the most accurate registrations. Results suggest that obtaining accurate registrations is a function of the percentage of the total surface and of geometric surface properties, such as curvature. Approximately 28% of the total surface is required regardless of the location of that surface subset. However, that percentage decreases when the surface subset contains information from opposite ends of the surface and/or unique anatomical features, such as the renal artery and vein.

The effect of anatomic landmark selection of the distal humerus on registration accuracy in computer-assisted elbow surgery

Incorrect selection of the flexion-extension axis of the elbow may be an important cause of failure following total elbow arthroplasty. Axis selection can be improved by locating it on a pre-operative image and registering the image to the subject intra-operatively. However, registration is dependent on the availability of anatomic landmarks that may be distorted or absent because of tumors, arthritis, dislocations, or fractures. This study determined the anatomic landmarks required to register surface data accurately to a pre-operative image of the distal humerus. Registration error for landmarks unlikely to be compromised by severe bone loss was 1.1 +/- 0.2 mm in translation and 0.4 +/- 0.1 degrees in rotation. These results suggest that a close alignment of a pre-operative image with intra-operative surface data can be achieved using only a relatively small portion of the distal humerus that is readily available to the surgeon, and unlikely to be compromised, even in the setting of significant articular bone loss.

Verification of clinical precision after computer-aided reconstruction in craniomaxillofacial surgery

OBJECTIVE

Computer-aided surgery (CAS) has proved to be useful in reconstructive craniomaxillofacial surgery. Preoperative creation of virtual models by segmentation of the computerized tomography (CT) dataset and mirroring of the unaffected side allows for precise planning of complex reconstructive procedures. The aim of this study was to evaluate the accuracy of the preoperative planning and the postoperative result regarding the skeletal reconstruction.

STUDY DESIGN

In a first step, the symmetry of unaffected human skulls and faces were evaluated by 20 midface CT data of skulls and 20 surface-scan data of healthy individuals. By mirroring and adjusting the original and mirrored datasets using a 3-dimensional modeling software, an automatic measurement procedure could evaluate the mean and the maximal modulus of the distances between both datasets. In a second step, 18 consecutive cases were selected which had been treated with CAS support. Group 1 consisted of orbital floor and/or medial wall fractures (n = 12), group 2 consisted of zygomatic bone fractures (n = 4), and group 3 included 2 patients who were treated by secondary orbital reconstruction including reosteotomy of the zygomatic bone (n = 2). To verify the surgical result, the preoperative CT dataset including the virtual planning and the postoperative CT dataset were compared by using image fusion. Additionally, postoperative surface scans and the clinical symptoms of the patients were evaluated.

RESULTS

No differences between the skull and face symmetry were found. Mean values for distances considering the skull symmetry were 0.83 mm for male and 0.71 mm for female and for the face symmetry 0.65 mm for male and 0.76 mm for female. Comparing the preoperative planning with the postoperative outcome, a mean accuracy of 1.49-4.12 mm with maximum modulus of 2.49-6.00 mm was achieved. Orbital true-to-original reconstructions and the secondary reconstructions were more precise than the reposition of the zygomatic bones. The postoperative acquired surface scans resulted in mean distances from 0.89 to 1.784 mm. Despite these deviations, all patients demonstrated satisfying clinical outcome.

CONCLUSION

The natural asymmetry in humans influences the accuracy of preoperative planning procedure, when the mirroring tool is used. The accuracy transforming the preoperative planning to the surgical reconstruction using CAS depends on location, surgical approach, and matter of reconstruction.

Comparison of 4 registration strategies for computer-aided maxillofacial surgery

PURPOSE

Surgeons have recently started to use computer-aided surgery (CAS) to assist with maxillofacial reconstructive surgery. This study evaluates four different CAS registration strategies in the maxillofacial skeleton.

MATERIALS AND METHODS

Fifteen fiducial markers were placed on each of four cadaveric heads. Four registration protocols were used: 1) group 1-invasive markers, 2) group 2-skin surface, 3) group 3-bony landmark, 4) group 4-intraoral splint. Two observers registered each head twice with each of the four protocols and measured the target registration error (TRE). The process was repeated on two different navigation systems for confirmation.

RESULTS

The mean TRE values were: invasive, 1.13 +/- 0.05 mm (P < 0.05); skin, 2.03 +/- 0.07 mm (P < 0.05); bone, 3.17 +/- 0.10 mm (P < 0.05); and splint, 3.79 +/- 0.13 mm (P < 0.05). The TRE values were consistent across CAS systems.

CONCLUSION

Of the techniques tested for CAS registration, invasive fiducial markers are the most accurate. Skin surface landmarks, bony landmarks, and an intraoral splint are incrementally less accurate.

A system for ultrasound-guided computer-assisted orthopaedic surgery

Current computer-assisted orthopedic surgery (CAOS) systems typically use preoperative computed tomography (CT) and intraoperative fluoroscopy as their imaging modalities. Because these imaging tools use X-rays, both patients and surgeons are exposed to ionizing radiation that may cause long-term health damage. To register the patient with the preoperative surgical plan, these techniques require tracking of the targeted anatomy by invasively mounting a tracking device on the patient, which results in extra pain and may prolong recovery time. The mounting procedure also leads to a major difficulty of using these approaches to track small bones or mobile fractures. Furthermore, it is practically impossible to mount a heavy tracking device on a small bone, which thus restricts the use of CAOS techniques. This article presents a novel CAOS method that employs 2D ultrasound (US) as the imaging modality. Medical US is non-ionizing and real-time, and our proposed method does not require any invasive mounting procedures. Experiments have shown that the proposed registration technique has sub-millimetric accuracy in localizing the best match between the intraoperative and preoperative images, demonstrating great potential for orthopedic applications. This method has some significant advantages over previously reported US-guided CAOS techniques: it requires no segmentation and employs only a few US images to accurately and robustly localize the patient. Preliminary laboratory results on both a radius-bone phantom and human subjects are presented.

Self-calibrating 3D-ultrasound-based bone registration for minimally invasive orthopedic surgery

Intraoperative freehand three-dimensional (3-D) ultrasound (3D-US) has been proposed as a noninvasive method for registering bones to a preoperative computed tomography image or computer-generated bone model during computer-aided orthopedic surgery (CAOS). In this technique, an US probe is tracked by a 3-D position sensor and acts as a percutaneous device for localizing the bone surface. However, variations in the acoustic properties of soft tissue, such as the average speed of sound, can introduce significant errors in the bone depth estimated from US images, which limits registration accuracy. We describe a new self-calibrating approach to US-based bone registration that addresses this problem, and demonstrate its application within a standard registration scheme. Using realistic US image data acquired from 6 femurs and 3 pelves of intact human cadavers, and accurate Gold Standard registration transformations calculated using bone-implanted fiducial markers, we show that self-calibrating registration is significantly more accurate than a standard method, yielding an average root mean squared target registration error of 1.6 mm. We conclude that self-calibrating registration results in significant improvements in registration accuracy for CAOS applications over conventional approaches where calibration parameters of the 3D-US system remain fixed to values determined using a preoperative phantom-based calibration.

Evaluation of achievable registration accuracy of the femur during minimally invasive total hip replacement

The aim of the paper was to investigate whether accurate, point-based registration of the intra-operative femur will be achieved within the context of minimally invasive surgery for total hip replacement. Computer tomography images, collected for pre-operative planning purposes, were used to simulate the intra-operative registration procedure using algorithms for various levels of measurement noise, different small areas of the femur available to the surgeon, and a limited number of collected data points (20-60). This helped with the choice of design variables to perform in vitro registration on a plastic bone model to validate the procedure, which included a multistart algorithm developed for intra-operative registration. The algorithm minimised the distance between the measured and image-derived surfaces and was able to cope with the presence of multiple local minima given sufficient computational effort, even with realistically large measurement noise. It was found that, if a small patch of the femur was used, accessible by a needle that could at times penetrate thin layers of soft tissue, errors in the order of 1.0 mm in translation and 0.5 degrees in rotation were achievable.

Surface-based registration accuracy of CT-based image-guided spine surgery

Registration is a critical and important process in maintaining the accuracy of CT-based image-guided surgery. The aim of this study was to evaluate the effects of the area of intraoperative data sampling and number of sampling points on the accuracy of surface-based registration in a CT-based spinal-navigation system, using an optical three-dimensional localizer. A cadaveric dry-bone phantom of the lumbar spine was used. To evaluate registration accuracy, three alumina ceramic balls were attached to the anterior and lateral aspects of the vertebral body. CT images of the phantom were obtained (1-mm slice thickness, at1-mm intervals) using a helical CT scanner. Twenty surface points were digitized from five zones defined on the basis of anatomical classification on the posterior aspects of the target vertebra. A total of 20 sets of sampling data were obtained. Evaluation of registration accuracy accounted for positional and rotational errors. Of the five zones, the area that was the largest and easiest to expose surgically and to digitize surface points was the lamina. The lamina was defined as standard zone. On this zone, the effect of the number of sampling points on the positional and rotational accuracy of registration was evaluated. And the effects of the additional area selected for intraoperative data sampling on the registration accuracy were evaluated. Using 20 surface points on the posterior side of the lamina, positional error was 0.96 mm +/- 0.24 mm root-mean-square (RMS) and rotational error was 0.91 degrees +/- 0.38 degrees RMS. The use of 20 surface points on the lamina usually allows surgeons to carry out sufficiently accurate registration to conduct computer-aided spine surgery. In the case of severe spondylosis, however, it might be difficult to digitize the surface points from the lamina, due to a hypertrophic facet joint or the deformity of the lamina and noisy sampling data. In such cases, registration accuracy can be improved by combining use of the 20 surface points on the lamina with surface points on other zones, such as on the both sides of the spinous process.

3-D/2-D registration of CT and MR to X-ray images

A crucial part of image-guided therapy is registration of preoperative and intraoperative images, by which the precise position and orientation of the patient's anatomy is determined in three dimensions. This paper presents a novel approach to register three-dimensional (3-D) computed tomography (CT) or magnetic resonance (MR) images to one or more two-dimensional (2-D) X-ray images. The registration is based solely on the information present in 2-D and 3-D images. It does not require fiducial markers, intraoperative X-ray image segmentation, or timely construction of digitally reconstructed radiographs. The originality of the approach is in using normals to bone surfaces, preoperatively defined in 3-D MR or CT data, and gradients of intraoperative X-ray images at locations defined by the X-ray source and 3-D surface points. The registration is concerned with finding the rigid transformation of a CT or MR volume, which provides the best match between surface normals and back projected gradients, considering their amplitudes and orientations. We have thoroughly validated our registration method by using MR, CT, and X-ray images of a cadaveric lumbar spine phantom for which "gold standard" registration was established by means of fiducial markers, and its accuracy assessed by target registration error. Volumes of interest, containing single vertebrae L1-L5, were registered to different pairs of X-ray images from different starting positions, chosen randomly and uniformly around the "gold standard" position. CT/X-ray (MR/ X-ray) registration, which is fast, was successful in more than 91% (82% except for L1) of trials if started from the "gold standard" translated or rotated for less than 6 mm or 17 degrees (3 mm or 8.6 degrees), respectively. Root-mean-square target registration errors were below 0.5 mm for the CT to X-ray registration and below 1.4 mm for MR to X-ray registration.

Robust registration for computer-integrated orthopedic surgery: laboratory validation and clinical experience

In order to provide navigational guidance during computer-integrated orthopedic surgery, the anatomy of the patient must first be registered to a medical image or model. A common registration approach is to digitize points from the surface of a bone and then find the rigid transformation that best matches the points to the model by constrained optimization. Many optimization criteria, including a least-squares objective function, perform poorly if the data include spurious data points (outliers). This paper describes a statistically robust, surface-based registration algorithm that we have developed for orthopedic surgery. To find an initial estimate, the user digitizes points from predefined regions of bone that are large enough to reliably locate even in the absence of anatomic landmarks. Outliers are automatically detected and managed by integrating a statistically robust M-estimator with the iterative-closest-point algorithm. Our in vitro validation method simulated the registration process by drawing registration data points from several sets of densely digitized surface points. The method has been used clinically in computer-integrated surgery for high tibial osteotomy, distal radius osteotomy, and excision of osteoid osteoma.

Segmentation of MR images for computer-assisted surgery of the lumbar spine

This paper describes a segmentation algorithm designed to separate bone from soft tissue in magnetic resonance (MR) images developed for computer-assisted surgery of the spine. The algorithm was applied to MR images of the spine of healthy volunteers. Registration experiments were carried out on a physical model of a spine generated from computed tomography (CT) data of a surgical patient. Segmented CT, manually segmented MR and MR images segmented using the developed algorithm were compared. The algorithm performed well at segmenting bone from soft tissue on images taken of healthy volunteers. Registration experiments showed similar results between the CT and MR data. The MR data, which were manually segmented, performed worse on visual verification experiments than both the CT and semi-automatic segmented data. The algorithm developed performs well at segmenting bone from soft tissue in MR images of the spine as measured using registration experiments.

The process and development of image-guided procedures

Medical imaging has been used primarily for diagnosis. In the past 15 years there has been an emergence of the use of images for the guidance of therapy. This process requires three-dimensional localization devices, the ability to register medical images to physical space, and the ability to display position and trajectory on those images. This paper examines the development and state of the art in those processes.

A 3D MRI sequence for computer assisted surgery of the lumbar spine

The aim of this research was to develop a magnetic resonance (MR) sequence capable of producing images suitable for use with computer assisted surgery (CAS) of the lumbar spine. These images needed good tissue contrast between bone and soft tissue to allow for image segmentation and generation of a 3D-surface model of the bone for surface registration. A 3D double echo fast gradient echo sequence was designed. Images were filtered for noise and non-uniformity and combined into a single data set. Registration experiments were carried out to directly compare segmentation of MR and computed tomography (CT) images using a physical model of a spine. These experiments showed the MR data produced adequate surface registration in 90% of the experiments compared to 100% with CT data. The MR images acquired using the sequence and processing described in this article are suitable to be used with CAS of the spine.

Automatic lumbar vertebral identification using surface-based registration

This work proposes the use of surface-based registration to automatically select a particular vertebra of interest during surgery. Manual selection of the correct vertebra can be a challenging task, especially for closed-back, minimally invasive procedures. Our method uses shape variations that exist among lumbar vertebrae to automatically determine the portion of the spinal column surface that correctly matches a set of physical vertebral points. In our experiments, we register vertebral points representing posterior elements of a single vertebra in physical space to spinal column surfaces extracted from computed tomography images of multiple vertebrae. After registering the set of physical points to each vertebral surface that is a potential match, we then compute the standard deviation of the surface error for each registration trial. The registration that corresponds to the lowest standard deviation designates the correct match. We have performed our current experiments on two plastic spine phantoms and two patients.