Kinematics of the normal, injured, or prosthetically replaced knee joint are a complex combination of rolling, gliding and rotational motions which are significantly influenced by the activity undertaken, the integrity of the ligaments and capsular structures, muscle activity, and articular geometry. Accurate kinematic information is critical to understanding the function and pathogenesis of the knee, particularly during weight bearing dynamic activities. In addition, intraoperative kinematic assessment during knee reconstruction would permit surgeons to objectively optimize graft placement and tensioning. The present study was undertaken to characterize the accuracy of a non-invasive fluoroscopic technique for measuring dynamic three-dimensional (3D) knee motions in individuals whose knees have not been prosthetically replaced, and demonstrate the measurement on in vivo step-up data. This technique utilizes orthogonal planar radiographic views of the knee to create a 3D contour model of consistently identifiable bony features for both the tibia and femur. The measurement technique is implemented by projecting the contour model onto digitized fluoroscopic images of the moving knee, and determining the translations and rotations which give the best correspondence between the projected contour model and the radiographic projection of the bone. Controlled in vitro assessment of the technique resulted in an average rotational accuracy of 1.1 degrees and a sagittal plane translational accuracy of 1.2 nun.
The VISLAN system is based on one of the newest developments in optical localiser tracking, where the localiser reflects light shone on it (passive illumination) rather than emitting light. In experiments with a CT-compatible skull phantom we evaluated the VISLAN system in comparison with a widely used stereotactic system, the Cosman-Roberts-Wells frame (Radionics Inc). The mean Euclidean difference in localising points initially defined in CT images was 2.5mm across a wide range of target positions. We estimated the accuracy of the VISLAN system with the same phantom set up but independently of the stereotactic frame, and showed a mean Euclidean error of 1.0mm. We concluded that the majority of the discrepancy between the two systems was accounted for by the inaccuracy of the CRW frame. It would be inappropriate to regard the frame based systems as a satisfactory gold standard for estimating the accuracy of VISLAN. However, frame-based and frameless guidance may be needed in the same operation, and it is important for such systems to be compared.
Robot-assisted surgery in Total Hip Replacement, was evaluated using the Robodoc system. 490 patients were operated with Robodoc since November 1994, and a two arm trial was conducted on twenty greyhounds. Results in both studies showed no system-related complications, a tolerable longer OR-time and benefits regarding position and healing of the implant in the robot group. No fractures or malposition were seen in the robot-operated people and dogs. Study results support further clinical use and applications.
Percutaneous surgical procedures are rapidly growing in popularity as they significantly reduce patient morbidity and recovery time when compared to more traditional open techniques. However, percutaneous procedures are often difficult. Percutaneous renal surgery is one example. Gaining access to the renal collecting system requires targeting of a specific calyx while avoiding critical internal structures and requires experience in the interpretation of radiographic images. Currently, this procedure is essentially done manually by trial and error. We present a system which automates the task of image-guided percutaneous needle placement. It is generally applicable to procedures for which the precision and accuracy of a percutaneous needle insertion is of primary concern. It can also be used, with minor modification, as a sub-component of larger systems for more involved procedures, for instance, percutaneous treatment of liver cancer [5]. A prototype system has been implemented, and we present validation of its applicability to renal access with in vitro, ex vivo, and in situ studies. Additionally, we discuss what work remains on the path to a complete system.
One of the practical reasons for applying film-cassettes instead of digitized video images, is that for some applications the entrance plane of the Image Intensifier is too small to cover the relevant part of a patients anatomy in one single exposure. Imaging of the deformed spine is such an application.
A computer algorithm for determining optimal surgical paths in the brain is presented. The algorithm computes a cost function associated with each point on the outer brain boundary, which is treated as a candidate entry point. The cost function is determined partly based on a segmentation of the patients images into gray and white matter, and partly based on a spatially transformed atlas of the human brain registered to the patient's MR images. The importance of various structures, such as thalamic nuclei, optic nerve and radiations, and individual Brodman's areas, can be defined on the atlas and transferred onto the patient's images through the spatial transformation. The cost of a particular path associated with each critical structure, as well as the total cost of each path are computed and displayed, allowing the surgeon to define a low cost path, to visualize an arbitrary cross-section through the patient's MR images that contains this path, and to examine all the cross-sectional images orthogonal to that path.
The lack of robust and reproducible methods for object segmentation still impedes the introduction of image postprocessing as widely used routine tools in clinical environments. In this paper, we present new tools for the segmentation of two- and three-dimensional objects from multidimensional image data. Our strategy is twofold: After creating an extended graph description of contour fragments and a tessellation of the image plane which is a fully automatic process running in the background, a user can choose between an interactive and a model-based segmentation procedure. A contour grouping algorithm based on path optimization can be used when full user interaction is required. Interactivity is limited to a few simple and quick operations. Another, region-based method uses a split- and-merge strategy and discrete optimization with global shape criteria. Grouping of primitive region patches is invoked by a contour model and a comparison of shape features. In combination, the two procedures form an efficient slice-propagation technique for the segmentation of volumetric objects from three-dimensional image data.
We present here a technique for reconstructing an eye, to guide the protontherapy of ocular tumours. This approach is expected to considerably improve the safety of the therapy. Our method is twofold. First, we present the eye's reconstruction from scanner images. Second, we describe the retinography and demonstrate the projection equations of this modality. This last result allows us to combine retinography with the reconstructed eye to get a virtual eye.
Since the first introduction of tumor stereotaxis to neurological surgery in the early 1980's, computer assisted surgery (CAS) has been applied to several medical fields. An important common issue for all CAS systems is the registration between preoperative data (e.g. computed tomography (CT), magnetic resonance (MR) images, ...) and intraoperative data. In our application the objective is to obtain position information of medical tools with respect to CT images of bony structures. We describe in this paper a new approach for the registration of preoperative CT images with the patient in the operating room. The method, that we named ‘restricted surface matching’, is a fast, accurate, and robust surface matching based method.
Certain minimally displaced acetabular fracture patterns are amenable to treatment with computed tomography (CT) assisted percutaneous fixation using cannulated screws (Kahler DM, et al. Presented at AAOS Annual Meeting, Feb. 1996). Although the early results have shown promise in the treatment of transverse, anterior column, and anterior column/posterior hemitransverse fractures, there remain concerns regarding radiation exposure to the patient, operative time averaging 45 minutes per screw, and the questionable sterility of the CT suite. A feasibility study was therefore undertaken to evaluate a new technique of percutaneous acetabular fracture fixation utilizing a stored CT image of the pelvis in a computer integrated surgical system, consisting of an array of three CCD cameras and a computer workstation. Five human cadaver pelvis specimens with intact external soft tissues were obtained, and simulated transverse acetabular fractures with roof arcs of approximately 30 degrees were created bilaterally. An external fiducial array consisting of aluminum spheres affixed to carbon fiber rods was attached to the iliac wing of each specimen with two 4.5 mm Schanz pins. Computed tomography was performed to allow segmentation of the specimens and fiducial arrays. The digitized scans were then loaded onto a computer workstation, and ideal pin placement and length were planned. The fiducial arrays were registered using a probe to allow orientation of the pelvis, and the specimen and drill guide were then optically and dynamically tracked during the procedure. Internal fixation was simulated by the percutaneous placement of a 2.8 mm guide wire perpendicular to the fracture cephalad to the joint surface, and a second wire crossing the fracture into the anterior column. Plain radiographs and specimen dissection were performed after fixation for evaluation of the accuracy of pin placement. There was no sign of penetration of the hip joint or the cortical bone of the pelvic brim along the entire path of the wires. It was found that computer integrated percutaneous fixation of transverse acetabular fractures in vitro provides comparable accuracy to CT-guided procedures without the need for intraoperative CT or a formal surgical exposure. Additionally, operative time is greatly reduced in comparison with CT-guided technique, and the computer integrated surgical system allows greater variability in the angle of screw placement to compensate for variations in pelvic anatomy. Computer integrated surgery may allow accurate percutaneous placement of cannulated screws in acetabular fracture surgery, and may prove a useful adjunct to conventional internal fixation using standard surgical approaches.
Radiation therapy uses the energy imparted by radiation in order to treat tumours: this method offers the main advantage of being non-invasive. Because all parts of the body get a dose, it requires, however, a great prudence. The dosimetric planning is necessary for concentrating the dose into the tumour and saving healthy tissues as much as possible. This task is performed by radiophysicists, who determine a treatment plan after several tests. The aim of our work, which is focused on the treatment of the prostate, is to automate this stage. We describe a new mathematical model for this optimization problem, and present some numerical results including different field shapes.
The infant's femoral head form and position relative to the acetabulum are the major parameters for the diagnosis of a hip joint dysplasia. The analysis can be performed using conventional ultrasound 2D-images, which virtually cut the hip joint in a specific plane. The joint's shape is scored by geometric measures taken from the image. The imaging in a defined plane is tedious and demands some experience. Furthermore the reporting of a spatial geometry through 2D-images tends to be subjective and incomplete. To consider the entire joint geometry, a 3D-data set of conventional ultrasound images is recorded along with the transducer movement. The femoral head is measured automatically in the spatially arranged images by fitting a virtual sphere, which approximates the femoral head's contours. A pilot study based on 20 3D-ultrasound data sets of hip joints of infants aging 2 days–12 weeks shows highly accurate coincidence of the automatically calculated diameter and centre parameters with those of control measurements, i.e. virtual spheres interactively placed and parametrized.
Dislocation following total hip replacement surgery represents a significant cause of early failure, incurring additional medical costs. The causes of dislocation are multifactorial and are related to surgical approach, soft tissue tension, prosthetic design, and most important, orientation of components. This paper describes experimental verification of our analytical approach for predicting implant impingement and dislocation. Once fully developed and tested, this analytical methodology could be used as a preoperative simulation tool that will present surgeons with information about the “safe” range of motion and chance of dislocation based on selected component positions, allowing for the surgical plan to be optimized based on this criterion. Coupled with a computer-assisted clinical system for precise implant positioning, this approach could significantly reduce the postoperative risk of dislocation, maximize “safe” range of motion and minimize impingement.
This paper describes the analysis of micro-surgery operation to determine the specifications of a tele-micro-surgery system. The implemented system for the connection of micro blood vessels consists of multiple micro co-located operation point slave manipulators, macro rotational-force-feedback-free master manipulators and a vision system which is comprised of a fixed viewpoint microscope and a movable monitor system. The parameters of the vision system can be controlled in accordance with the system's understanding of the operator's intention by monitoring the posture change of the operator. The experimental results showed the effectiveness of the developed system. In the experiment, a blood vessel of a rat with a diameter less than 1 mm was successfully sutured using needle with a curvature radius of 2 mm.
Robots have the potential to assist in orthopædic surgery and improve the outcome of prosthetic implants. Robots have high positional accuracy, and so can achieve the geometrical precision necessary for implanting prostheses into the tibia and femur but poor tactile response, so are not good at assessing the forces required to apply to a cutter while resecting bone; while surgeons have a good tactile sense, and are able to sense changes in bone density, and adjust cutting forces to match. A system is described here that exploits the synergy between robot and surgeon. A force controlled pobot is used, guided from the end-effector by the surgeon. The robot exploits a software based motion constraint system to ensure that the surgeon cannot move a cutting device connected to the robot outside of a safe region, or resect more bone than is required. Thus, the surgeon retains his tactile sense of the bone, while the precision of the robot allows the bones to be cut accurately. Preliminary results are presented in this paper. More complete details of the robot and system performance will be presented at the symposium.
In this paper, we will describe a new 2-D–D registration algorithm, intended to solve a 3-D localization problem.
Individual templates for Orthopedic Surgery provide a precise intraoperative reproduction of the geometries of work on bone planned preoperatively on the base of CT-image data. The general feasibility concerning the adaptation to different surgical applications has already been demonstrated in various in-vitro studies as well as within clinical application1,2,3,4,5. Within the framework of the European IGOS-project a demonstrator for image guided pelvis surgery as an exemplary clinical application will be developed. In contrast to initial feasibility studies with our first labtyp systems, the introduction into clinical routine induces additional constraints. To provide an adequate level of reliability and usability additional efforts are necessary especially concerning quality assurance, ergonomic design and standardisation of the related chain of image acquisition and surgical planning, manufacturing of individual templates and finally the intraoperative execution. Aspects of the integration into clinical routine as well as laboratory investigations concerning accuracy and integrated manufacturing are discussed.
The Neurovisualization Lab at the University of Virginia is developing the Integrated Remote Neurosurgical System (IRNS) to allow mentoring of neurosurgical procedures in remote locations. The system allows a remote neurosurgeon to control a robotic microscope through the use of a 3-D input device, communicate with the operating room (0/R) team through live audio and video, and view presurgical imagery. The surgical team in the O/R will have access to the same images and communication facilities. The system will serve as a training tool through the use of a complete robotic simulation we have developed. We have also instituted safety precautions in the form of restriction of robot motion, monitoring, and protocols of system use. We have developed a registration system to assist in the implementation of these guidelines. A task analysis has led to the development of a prototype user interface, and the preliminary integration of available components has been completed. We report on the current state of the system and ongoing development with respect to the user interface and experimentation.
Image-guided, computer-assisted surgery systems base on a common reference between the pre-operative image data and the corresponding patient pathology. Therefore, accurate 3D patient registration and referencing is necessary. Considering that during image data acquisition patient misalignment and movement can arise in three dimensions, an automatic image registration method must base on a three-dimensional approach. Our method provides non-invasive 3D patient registration for correction of such movement errors during data acquisition and active referencing to update the position of the patient's head during surgery. Experimental data concerning the accuracy of repeated positionings of our patient registration and reference system demonstrated an accuracy with mean spatial errors of 0.82 mm ± 0.31 mm in a plastic skull and 1.39 mm ± 0.61 mm in patients, respectively. Results suggest that non-invasive 3D patient registration for image-guided surgery may be a precise and useful method for computer-assisted identification of anatomical structures. Clinical experiences for different pathologies are presented (30 patients).
