Guide-wire tracking during endovascular interventions

Advancing endovascular neurosurgery training with extended reality: opportunities and obstacles for the next decade

Background

Extended reality (XR) includes augmented reality (AR), virtual reality (VR), and mixed reality (MR). Endovascular neurosurgery is uniquely positioned to benefit from XR due to the complexity of cerebrovascular imaging. Given the different XR modalities available, as well as unclear clinical utility and technical capabilities, we clarify opportunities and obstacles for XR in training vascular neurosurgeons.

Methods

A systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines was conducted. Studies were critically appraised using ROBINS-I.

Results

19 studies were identified. 13 studies used VR, while 3 studies used MR, and 3 studies used AR. Regarding specific educational applications, VR was used for simulation in 10 studies and anatomical modeling in 3 studies. AR was only used for live intra-operative guidance (n = 3 studies). MR was only used for modeling and intra-operative teaching. Considering disease-specific uses, XR enhanced trainee understanding of intracranial aneurysms (n = 12 studies) and stroke (n = 7). XR trained surgeons in diverse neurosurgical procedures, including aneurysm coiling (n = 5 studies), diagnostic angiography (n = 5), and thrombectomy (n = 5).

Conclusions

Anatomical modeling with VR and MR enhances neurovascular anatomy education with patient-specific, 3-D models from imaging data. AR and MR enable live intra-operative guidance, allowing experienced surgeons to remotely instruct novices, potentially improving patient care and reducing geographic disparities. AR overlays enhance instruction by allowing the surgeon to highlight key procedural aspects during training. Inaccurate tracking of surgical tools is an XR technological barrier for modeling and intra-operative training. Importantly, the most reported application of XR is VR for simulation-using platforms like the Mentice VIST and Angio Mentor. 10 studies examine VR for simulation, showing enhanced procedural performance and reduced fluoroscopy use after short training, although long-term outcomes have not been reported. Early-stage trainees benefited the most. Simulation improved collaboration between neurosurgeons and the rest of the surgical team, a promising role in interprofessional teamwork. Given the strength of VR for simulation, MR for simulation is an important gap in the literature for future studies. In conclusion, XR holds promise for transforming neurosurgical education and practice for simulation, but technological research is needed in modeling and intra-procedural training.

High-resolution feature based central venous catheter tip detection network in X-ray images

Hospital patients can have catheters and lines inserted during the course of their admission to give medicines for the treatment of medical issues, especially the central venous catheter (CVC). However, malposition of CVC will lead to many complications, even death. Clinicians always detect the malposition based on position detection of CVC tip via X-ray images. To reduce the workload of the clinicians and the percentage of malposition occurrence, we propose an automatic catheter tip detection framework based on a convolutional neural network (CNN). The proposed framework contains three essential components which are modified HRNet, segmentation supervision module, and deconvolution module. The modified HRNet can retain high-resolution features from start to end, ensuring the maintenance of precise information from the X-ray images. The segmentation supervision module can alleviate the presence of other line-like structures such as the skeleton as well as other tubes and catheters used for treatment. In addition, the deconvolution module can further increase the feature resolution on the top of the highest-resolution feature maps in the modified HRNet to get a higher-resolution heatmap of the catheter tip. A public CVC Dataset is utilized to evaluate the performance of the proposed framework. The results show that the proposed algorithm offering a mean Pixel Error of 4.11 outperforms three comparative methods (Ma's method, SRPE method, and LCM method). It is demonstrated to be a promising solution to precisely detect the tip position of the catheter in X-ray images.

Robotics in Interventional Radiology: Review of Current and Future Applications

This review is a brief overview of the current status and the potential role of robotics in interventional radiology. Literature published in the last decades, with an emphasis on the last 5 years, was reviewed and the technical developments in robotics and navigational systems using CT-, MR- and US-image guidance were analyzed. Potential benefits and disadvantages of their current and future use were evaluated. The role of fusion imaging modalities and artificial intelligence was analyzed in both percutaneous and endovascular procedures. A few hundred articles describing results of single or several systems were included in our analysis.

Catheter segmentation in X-ray fluoroscopy using synthetic data and transfer learning with light U-nets

Background and objectivesAutomated segmentation and tracking of surgical instruments and catheters under X-ray fluoroscopy hold the potential for enhanced image guidance in catheter-based endovascular procedures. This article presents a novel method for real-time segmentation of catheters and guidewires in 2d X-ray images. We employ Convolutional Neural Networks (CNNs) and propose a transfer learning approach, using synthetic fluoroscopic images, to develop a lightweight version of the U-Net architecture. Our strategy, requiring a small amount of manually annotated data, streamlines the training process and results in a U-Net model, which achieves comparable performance to the state-of-the-art segmentation, with a decreased number of trainable parameters. MethodsThe proposed transfer learning approach exploits high-fidelity synthetic images generated from real fluroscopic backgrounds. We implement a two-stage process, initial end-to-end training and fine-tuning, to develop two versions of our model, using synthetic and phantom fluoroscopic images independently. A small number of manually annotated in-vivo images is employed to fine-tune the deepest 7 layers of the U-Net architecture, producing a network specialized for pixel-wise catheter/guidewire segmentation. The network takes as input a single grayscale image and outputs the segmentation result as a binary mask against the background. ResultsEvaluation is carried out with images from in-vivo fluoroscopic video sequences from six endovascular procedures, with different surgical setups. We validate the effectiveness of developing the U-Net models using synthetic data, in tests where fine-tuning and testing in-vivo takes place both by dividing data from all procedures into independent fine-tuning/testing subsets as well as by using different in-vivo sequences. Accurate catheter/guidewire segmentation (average Dice coefficient of  ~ 0.55,  ~ 0.26 and  ~ 0.17) is obtained with both U-Net models. Compared to the state-of-the-art CNN models, the proposed U-Net achieves comparable performance ( ± 5% average Dice coefficients) in terms of segmentation accuracy, while yielding a 84% reduction of the testing time. This adds flexibility for real-time operation and makes our network adaptable to increased input resolution. ConclusionsThis work presents a new approach in the development of CNN models for pixel-wise segmentation of surgical catheters in X-ray fluoroscopy, exploiting synthetic images and transfer learning. Our methodology reduces the need for manually annotating large volumes of data for training. This represents an important advantage, given that manual pixel-wise annotations is a key bottleneck in developing CNN segmentation models. Combined with a simplified U-Net model, our work yields significant advantages compared to current state-of-the-art solutions.

Pyramid attention recurrent networks for real-time guidewire segmentation and tracking in intraoperative X-ray fluoroscopy

In endovascular and cardiovascular surgery, real-time and accurate segmentation and tracking of interventional instruments can aid in reducing radiation exposure, contrast agent and processing time. Nevertheless, this task often comes with the challenges of the elongated deformable structures with low contrast in noisy X-ray fluoroscopy. To address these issues, a novel efficient network architecture, termed pyramid attention recurrent networks (PAR-Net), is proposed for real-time guidewire segmentation and tracking. The proposed PAR-Net contains three major modules, namely pyramid attention module, recurrent residual module and pre-trained MobileNetV2 encoder. Specifically, a hybrid loss function of both reinforced focal loss and dice loss is proposed to better address the issues of class imbalance and misclassified examples. Quantitative and qualitative evaluations on clinical intraoperative images demonstrate that the proposed approach significantly outperforms simpler baselines as well as the best previously published result for this task, achieving the state-of-the-art performance.

A GPU-Based Automatic Approach for Guide Wire Tracking in Fluoroscopic Sequences

Guide wire tracking in fluoroscopic images has done a significant task in assisting the physicians during radiology-aided interventions. Many groups have tried to detect the guide wire from the fluoroscopic images based on the image properties. The main challenge is that manual intervention is required during the detection. Other groups try to introduce localizers to track guide wires during intervention, which requires additional hardware equipment, and may intervene with the traditional clinical routines. Machine learning methods are also exploited. Although such methods may provide accurate tracking, they often require large amount of data and training time.

In this paper, we propose a GPU-based fast and automatic approach to track guide wires in fluoroscopic sequences. We propose a multi-scale filtering and gradient vector field-based real-time tracking method for guide wire tracking from fluoroscopic images. To improve calculation efficiency and meet real-time application requirement, we propose a GPU-based acceleration scheme, and also a Bayesian filter-like motion tracking method to limit the guide wire tracking to a smaller range to improve calculation efficiency.

We test our proposed method on two test data sets of fluoroscopic sequences of 102 frames and 72 frames. We achieve an average guide wire detection rate of 96.7%, a false detection rate of 0.0011% and an error distance measure of 0.83 pixels for the first sequence, and 98.8%, 0.000069% and 0.85 pixels, respectively, for the second sequence. With the proposed acceleration method, we finish calculation for the first sequence in nine seconds, thus, efficiency is enhanced by 100 times with the unaccelerated algorithm.

A novel real-time computational framework for detecting catheters and rigid guidewires in cardiac catheterization procedures

Purpose

Catheters and guidewires are used extensively in cardiac catheterization procedures such as heart arrhythmia treatment (ablation), angioplasty, and congenital heart disease treatment. Detecting their positions in fluoroscopic X‐ray images is important for several clinical applications, for example, motion compensation, coregistration between 2D and 3D imaging modalities, and 3D object reconstruction.

Methods

For the generalized framework, a multiscale vessel enhancement filter is first used to enhance the visibility of wire‐like structures in the X‐ray images. After applying adaptive binarization method, the centerlines of wire‐like objects were extracted. Finally, the catheters and guidewires were detected as a smooth path which is reconstructed from centerlines of target wire‐like objects. In order to classify electrode catheters which are mainly used in electrophysiology procedures, additional steps were proposed. First, a blob detection method, which is embedded in vessel enhancement filter with no additional computational cost, localizes electrode positions on catheters. Then the type of electrode catheters can be recognized by detecting the number of electrodes and also the shape created by a series of electrodes. Furthermore, for detecting guiding catheters or guidewires, a localized machine learning algorithm is added into the framework to distinguish between target wire objects and other wire‐like artifacts. The proposed framework were tested on total 10,624 images which are from 102 image sequences acquired from 63 clinical cases.

Results

Detection errors for the coronary sinus (CS) catheter, lasso catheter ring and lasso catheter body are 0.56 ± 0.28 mm, 0.64 ± 0.36 mm, and 0.66 ± 0.32 mm, respectively, as well as success rates of 91.4%, 86.3%, and 84.8% were achieved. Detection errors for guidewires and guiding catheters are 0.62 ± 0.48 mm and success rates are 83.5%.

Conclusion

The proposed computational framework do not require any user interaction or prior models and it can detect multiple catheters or guidewires simultaneously and in real‐time. The accuracy of the proposed framework is sub‐mm and the methods are robust toward low‐dose X‐ray fluoroscopic images, which are mainly used during procedures to maintain low radiation dose.

Robust guidewire tracking under large deformations combining segment-like features (SEGlets)

Robust tracking of interventional tools, such as guidewires and catheters, in X-ray fluoroscopic video sequences has a wide range of clinical applications for endovascular procedures. Thus far, the tracking is usually achieved by finding the optimal displacement of the control points of a spline, which models the guidewire, between consecutive frames. The displacement of the control points is typically driven by a data term and smoothed by a regularization term. In the presence of large deformation and changes in length of the tool, the current tracking methods may fail to recover the guidewire motion. This can occur because of the limitation of the data and regularization terms, and the absence of an explicit solution for coping with elongations of the guidewire. The purpose of this paper is to present an algorithm that can robustly track guidewires under these challenging conditions. The algorithm is based on two main contributions: (a) new robust features termed SEGlets for segment-like features are introduced to overcome the limitations of the current data terms; (b) a tracking formulation based on the generation of tracking hypotheses by organizing the SEGlets in plausible guidewire shapes. The proposed method allows high flexibility of the guidewire between consecutive frames in contrast to the spline model, which can suffer from the limitations of the regularization terms. Furthermore, the technique models elongations of the guidewire which makes it possible for robust tracking under motion. A tool model which is recursively updated by employing a Kalman filter, is also proposed for modelling the regularization term. A detailed evaluation and a comparative study with three state-of-the-art guidewire tracking methods have been performed to demonstrate the potential clinical value of the technique. The proposed method achieves an overall guidewire tracking precision of 2.40 pixels, tip precision of 25.55 pixels, false tracking rate of 5.73%, missing tracking rate of 9.69%, and F₁ score of 0.92. The implementation of the proposed technique and the three tracking methods will be made publicly available as software libraries.

4D interventional device reconstruction from biplane fluoroscopy

PURPOSE

Biplane angiography systems provide time resolved 2D fluoroscopic images from two different angles, which can be used for the positioning of interventional devices such as guidewires and catheters. The purpose of this work is to provide a novel algorithm framework, which allows the 3D reconstruction of these curvilinear devices from the 2D projection images for each time frame. This would allow creating virtual projection images from arbitrary view angles without changing the position of the gantries, as well as virtual endoscopic 3D renderings.

METHODS

The first frame of each time sequence is registered to and subtracted from the following frame using an elastic grid registration technique. The images are then preprocessed by a noise reduction algorithm using directional adaptive filter kernels and a ridgeness filter that emphasizes curvilinear structures. A threshold based segmentation of the device is then performed, followed by a flux driven topology preserving thinning algorithm to extract the segments of the device centerline. The exact device path is determined using Dijkstra's algorithm to minimize the curvature and distance between adjacent segments as well as the difference to the device path of the previous frame. The 3D device centerline is then reconstructed using epipolar geometry.

RESULTS

The accuracy of the reconstruction was measured in a vascular head phantom as well as in a cadaver head and a canine study. The device reconstructions are compared to rotational 3D acquisitions. In the phantom experiments, an average device tip accuracy of 0.35 ± 0.09 mm, a Hausdorff distance of 0.65 ± 0.32 mm, and a mean device distance of 0.54 ± 0.33 mm were achieved. In the cadaver head and canine experiments, the device tip was reconstructed with an average accuracy of 0.26 ± 0.20 mm, a Hausdorff distance of 0.62 ± 0.08 mm, and a mean device distance of 0.41 ± 0.08 mm. Additionally, retrospective reconstruction results of real patient data are presented.

CONCLUSIONS

The presented algorithm is a novel approach for the time resolved 3D reconstruction of interventional devices from biplane fluoroscopic images, thus allowing the creation of virtual projection images from arbitrary view angles as well as virtual endoscopic 3D renderings. Availability of this technique would enhance the ability to accurately position devices in minimally invasive endovascular procedures.

Real-time 3D position reconstruction of guidewire for monoplane X-ray

We present a novel real-time method for the 3D reconstruction of the guidewire using a monoplane X-ray. The method consists of two steps: (1) the backprojection step to reconstruct a 3D surface that contains the guidewire and (2) the optimization step to select a curve on the surface that is the best match under the pre-specified constraints. The proposed method utilizes a priori knowledge in the form of a volume that indicates positions of the blood vessels and thus restricts the reconstruction. The reconstruction precision is limited by the local thickness of the vessels. The method is quantitatively evaluated on five phantom datasets and qualitatively on two patient datasets. For the phantom datasets the average reconstruction error is resolution limited to 1-2 voxels and is biased in the depth direction. The worst-case reconstruction error for any point, including the guidewire tip, is not larger than the local vessel thickness. A visual inspection of results for the patient datasets shows the guidewire is always placed in the proper vessel and is aligned with the 2D image, which is sufficient for the guidewire navigation. The developed implementation achieves the processing speed of 12 fps using Core™i7 CPU 920 at 2.67 GHz.

Current and emerging robot-assisted endovascular catheterization technologies: a review

Endovascular techniques have been embraced as a minimally-invasive treatment approach within different disciplines of interventional radiology and cardiology. The current practice of endovascular procedures, however, is limited by a number of factors including exposure to high doses of X-ray radiation, limited 3D imaging, and lack of contact force sensing from the endovascular tools and the vascular anatomy. More recently, advances in steerable catheters and development of master/slave robots have aimed to improve these practices by removing the operator from the radiation source and increasing the precision and stability of catheter motion with added degrees-of-freedom. Despite their increased application and a growing research interest in this area, many such systems have been designed without considering the natural manipulation skills and ergonomic preferences of the operators. Existing studies on tool interactions and natural manipulation skills of the operators are limited. In this manuscript, new technical developments in different aspects of robotic endovascular intervention including catheter instrumentation, intra-operative imaging and navigation techniques, as well as master/slave based robotic catheterization platforms are reviewed. We further address emerging trends and new research opportunities towards more widespread clinical acceptance of robotically assisted endovascular technologies.

Interventional tool tracking using discrete optimization

This work presents a novel scheme for tracking of motion and deformation of interventional tools such as guide-wires and catheters in fluoroscopic X-ray sequences. Being able to track and thus to estimate the correct positions of these tools is crucial in order to offer guidance enhancement during interventions. The task of estimating the apparent motion is particularly challenging due to the low signal-to-noise ratio (SNR) of fluoroscopic images and due to combined motion components originating from patient breathing and tool interactions performed by the physician. The presented approach is based on modeling interventional tools with B-splines whose optimal configuration of control points is determined through efficient discrete optimization. Each control point corresponds to a discrete random variable in a Markov random field (MRF) formulation where a set of labels represents the deformation space. In this context, the optimal curve corresponds to the maximum a posteriori (MAP) estimate of the MRF energy. The main motivation for employing a discrete approach is the possibility to incorporate a multi-directional search space which is robust to local minima. This is of particular interest for curve tracking under large deformation. This work analyzes feasibility of employing efficient first-order MRFs for tracking. In particular it shows how to achieve a good compromise between energy approximations and computational efficiency. Experimental results suggest to define both the external and internal energy in terms of pairwise potential functions. The method was successfully applied to the tracking of guide-wires in fluoroscopic X-ray sequences of several hundred frames which requires extremely robust techniques. Comparisons with state-of-the-art guide-wire tracking algorithms confirm the effectiveness of the proposed method.

Toward multiple catheters detection in fluoroscopic image guided interventions

Catheters are routinely inserted via vessels to cavities of the heart during fluoroscopic image guided interventions for electrophysiology (EP) procedures such as ablation. During such interventions, the catheter undergoes nonrigid deformation due to physician interaction, patient's breathing, and cardiac motions. EP clinical applications can benefit from fast and accurate automatic catheter tracking in the fluoroscopic images. The typical low quality in fluoroscopic images and the presence of other medical instruments in the scene make the automatic detection and tracking of catheters in clinical environments very challenging. Toward the development of such an application, a robust and efficient method for detecting and tracking the catheter sheath is developed. The proposed approach exploits the clinical setup knowledge to constrain the search space while boosting both tracking speed and accuracy, and is based on a computationally efficient framework to trace the sheath and simultaneously detect one or multiple catheter tips. The algorithm is based on a modification of the fast marching weighted distance computation that efficiently calculates, on the fly, important geodesic properties in relevant regions of the image. This is followed by a cascade classifier for detecting the catheter tips. The proposed technique is validated on 1107 fluoroscopic images acquired on multiple patients across four different clinics, achieving multiple catheter tracking at a rate of 10 images/s with a very low false positive rate of 1.06.

Classification-based summation of cerebral digital subtraction angiography series for image post-processing algorithms

X-ray-based 2D digital subtraction angiography (DSA) plays a major role in the diagnosis, treatment planning and assessment of cerebrovascular disease, i.e. aneurysms, arteriovenous malformations and intracranial stenosis. DSA information is increasingly used for secondary image post-processing such as vessel segmentation, registration and comparison to hemodynamic calculation using computational fluid dynamics. Depending on the amount of injected contrast agent and the duration of injection, these DSA series may not exhibit one single DSA image showing the entire vessel tree. The interesting information for these algorithms, however, is usually depicted within a few images. If these images would be combined into one image the complexity of segmentation or registration methods using DSA series would drastically decrease. In this paper, we propose a novel method automatically splitting a DSA series into three parts, i.e. mask, arterial and parenchymal phase, to provide one final image showing all important vessels with less noise and moving artifacts. This final image covers all arterial phase images, either by image summation or by taking the minimum intensities. The phase classification is done by a two-step approach. The mask/arterial phase border is determined by a Perceptron-based method trained from a set of DSA series. The arterial/parenchymal phase border is specified by a threshold-based method. The evaluation of the proposed method is two-sided: (1) comparison between automatic and medical expert-based phase selection and (2) the quality of the final image is measured by gradient magnitudes inside the vessels and signal-to-noise (SNR) outside. Experimental results show a match between expert and automatic phase separation of 93%/50% and an average SNR increase of up to 182% compared to summing up the entire series.

Towards a new image guidance system for assisting transapical minimally invasive aortic valve implantation

We propose a new image guidance system for assisting transapical minimally invasive aortic valve implantation. The goal is to define the exact positioning of aortic valve prosthesis, preventing the misplacement of the valve. The proposed system consists of two stand-alone modules. First, preoperative planning software uses DynaCT images with manual anatomical landmarks to calculate the size and optimal position of the prosthesis. Second, an intraoperative system is developed for tracking of the prosthesis and the coronary ostia in 2-D fluoroscopic images. Then the safe area of implantation is defined. The preliminary experimental results of preoperative planning and intraoperative tracking system are promising.

Variational guidewire tracking using phase congruency

We present a novel method to track a guidewire in cardiac x-ray video. Using variational calculus, we derive differential equations that deform a spline, subject to intrinsic and extrinsic forces, so that it matches the image data, remains smooth, and preserves an a priori length. We analytically derive these equations from first principles, and show how they include tangential terms, which we include in our model. To address the poor contrast often observed in x-ray video, we propose using phase congruency as an image-based feature. Experimental results demonstrate the success of the method in tracking guidewires in low contrast x-ray video.

An algorithm for tracking microcatheters in fluoroscopy

Currently, a large number of endovascular interventions are performed for treatment of intracranial aneurysms. For these treatments, correct positioning of microcatheter tips, microguide wire tips, or coils is essential. Techniques to detect such devices may facilitate endovascular interventions. In this paper, we describe an algorithm for tracking of microcatheter tips during fluoroscopically guided neuroendovascular interventions. A sequence of fluoroscopic images (1,024 x 1,024 x 12 bits) was acquired using a C-arm angiography system as a microcatheter was passed through a carotid phantom which was on top of a head phantom. The carotid phantom was a silicone cylinder containing a simulated vessel with the shape and curvatures of the internal carotid artery. The head phantom consisted of a human skull and tissue-equivalent material. To detect the microcatheter in a given fluoroscopic frame, a background image consisting of an average of the four previous frames is subtracted from the current frame, the resulting image is filtered using a matched filter, and the position of maximum intensity in the filtered image is taken as the catheter tip position in the current frame. The distance between the tracked position and the correct position (error distance) was measured in each of the fluoroscopic images. The mean and standard deviation of the error distance values were 0.277 mm (1.59 pixels) and 0.26 mm (1.5 pixels), respectively. The error distance was less than 3 pixels in the 93.0% frames. Although the algorithm intermittently failed to correctly detect the catheter, the algorithm recovered the catheter in subsequent frames.

Modeling Friction, Intrinsic Curvature, and Rotation of Guide Wires for Simulation of Minimally Invasive Vascular Interventions

Obtaining the expertise to perform minimally invasive vascular interventions requires thorough training. In this paper, an algorithm for simulating minimally invasive vascular interventions for training purposes is presented and evaluated. The algorithm enables the simulation of completely straight guide wires as well as intrinsically curved ones based on applied translations and rotations. Friction between the guide wire and the vasculature is incorporated in the model. Quantitative validation is performed by comparing the simulated guide-wire position with the actual position as assessed by 3-D rotational X-ray imaging in physical experiments on a variety of vascular phantoms that truthfully represent human anatomy. The results show that for proper settings of the model's parameters, accurate simulations of guide-wire motion can be obtained, with an average precision of the guide-wire position of around 1.0 mm.

Optimization of three-dimensional angiographic data obtained by self-calibration of multiview imaging

Stroke is one of the leading causes of death in the U.S. The treatment of stroke often involves vascular interventions in which devices are guided to the intervention site often through tortuous vessels based on two-dimensional (2-D) angiographic images. Three dimensional (3-D) vascular information may facilitate these procedures. Methods have been proposed for the self-calibrating determination of 3-D vessel trees from biplane and multiple plane images and the geometric relationships between the views (imaging geometries). For the biplane analysis, four or more corresponding points must be identified in the biplane images. For the multiple view technique, multiple vessels must be indicated and only the translation vectors relating the geometries are calculated. We have developed methods for the calculation of the 3-D vessel data and the full transformations relating the multiple views (rotations and translations) obtained during interventional procedures, and the technique does not require indication of corresponding points, but only the indication of a single vessel, e.g., the vessel of interest. Multiple projection views of vessel trees are obtained and transferred to the analysis computer. The vessel or vessels of interest are indicated by the user. Using the initial imaging geometry determined from the gantry information, 3-D vessel centerlines are calculated using the indicated centerlines in pairs of images. The imaging geometries are then iteratively adjusted and 3-D centerlines recalculated until the root-mean-square (rms) difference between the calculated 3-D centerlines is minimized. Simulations indicate that the 3-D centerlines can be accurately determined (to within 1 mm) even for errors in indication of the vessel endpoints as large as 5 mm. In phantom studies, the average rms difference between the pairwise calculated 3-D centerlines is approximately 7.5 mm prior to refinement (i.e., using the gantry information alone), whereas the average rms difference is usually below 1 mm after refinement. Accuracies and reliabilities of better than 1 mm were also determined by comparing centerlines determined using multiview and rotational angiography reconstruction and clinical data sets. These results indicate that the multiview approach will provide accurate and reliable 3-D centerlines for indicated vessel(s) without increasing the dose to the patient.

Microcatheter tip enhancement in fluoroscopy: a comparison of techniques

We compared three techniques for enhancement of microcatheter tips in fluoroscopic images: conventional subtraction technique (CST); averaged image subtraction technique (AIST), which we have developed; and double average filtering (DAF) technique, which uses nonlinear background estimates. A pulsed fluoroscopic image sequence was obtained as a microcatheter was passed through a carotid phantom that was on top of a head phantom. The carotid phantom was a silicone cylinder containing a simulated vessel with the shape and curvatures of the internal carotid artery. The three techniques were applied to the images of the sequence, then the catheter tip was manually identified in each image, and 100 x 100 pixel images, centered at the indicated microcatheter tip positions, were extracted for the evaluations. The signal-to-noise ratio (SNR) was calculated in each of the extracted images from which the mean value of the SNR and its standard deviation (SD) were calculated for each technique. The mean values and the standard deviations were 4.36 (SD 3.40) for CST, 6.34 (SD 3.62) for AIST, and 3.55 (SD 1.27) for DAF. AIST had a higher SNR compared to CST in almost all frames. Although DAF yielded the smallest mean SNR value, it yielded the best SNR in those frames in which the microcatheter tip did not move between frames. We conclude that AIST provides the best SNR for a moving microcatheter tip and that DAF is optimal for a temporarily stationary microcatheter tip.

Tracking technique of a micro guide wire in sequential fluorograms

In this paper, we propose a tracking technique for a micro guide wire in a sequence of fluorograms. A technique in which the region-growing technique was embedded was developed. It gives the center line of a micro guide wire. A sequence of 1,024 x 1,024 x 16 bit fluorograms (111 frames) with a carotid phantom and head phantom was obtained with a CAS-8000V (Toshiba America Medical Systems, Inc., CA, USA) C-arm angiography system and a Hi-Torque Standard micro guide wire (Advanced Cardiovascular Systems, Inc., CA, USA). To evaluate the technique, we manually traced the guide wire in each sequence frame three times, and a "true" single-width micro guide wire was created from them. The number of pixels on the true guide wire and inside the two-pixel tolerance of the center line was counted in each fluorogram, and the percentage of that count based on the number of all pixels on the true guide wire was calculated as true positive (TP). In addition, the number of pixels on the center line and outside the two-pixel tolerance of the true guide wire was counted in each frame as false positive (FP). The tracking technique has a mean TP of 94.8% and a mean FP of 5.1 pixels/frame. Several frames have low TP and high FP, but the technique could continue to track the micro guide wire until the end of the sequence. We therefore concluded that we had developed an accurate automatic tracking technique for micro guide wires in fluoroscopic sequences.

FTRAC--a robust fluoroscope tracking fiducial

C-arm fluoroscopy is ubiquitous in contemporary surgery, but it lacks the ability to accurately reconstruct three-dimensional (3D) information. A major obstacle in fluoroscopic reconstruction is discerning the pose of the x-ray image, in 3D space. Optical/magnetic trackers tend to be prohibitively expensive, intrusive and cumbersome in many applications. We present single-image-based fluoroscope tracking (FTRAC) with the use of an external radiographic fiducial consisting of a mathematically optimized set of ellipses, lines, and points. This is an improvement over contemporary fiducials, which use only points. The fiducial encodes six degrees of freedom in a single image by creating a unique view from any direction. A nonlinear optimizer can rapidly compute the pose of the fiducial using this image. The current embodiment has salient attributes: small dimensions (3 x 3 x 5 cm); need not be close to the anatomy of interest; and accurately segmentable. We tested the fiducial and the pose recovery method on synthetic data and also experimentally on a precisely machined mechanical phantom. Pose recovery in phantom experiments had an accuracy of 0.56 mm in translation and 0.33 degrees in orientation. Object reconstruction had a mean error of 0.53 mm with 0.16 mm STD. The method offers accuracies similar to commercial tracking systems, and appears to be sufficiently robust for intraoperative quantitative C-arm fluoroscopy. Simulation experiments indicate that the size can be further reduced to 1 x 1 X 2 cm, with only a marginal drop in accuracy.

A comparison of a similarity-based and a feature-based 2-D-3-D registration method for neurointerventional use

Two-dimensional (2-D)-to-three-dimensional (3-D) registration can improve visualization which may aid minimally invasive neurointerventions. Using clinical and phantom studies, two state-of-the-art approaches to rigid registration are compared quantitatively: an intensity-based algorithm using the gradient difference similarity measure; and an iterative closest point (ICP)-based algorithm. The gradient difference approach was found to be more accurate, with an average registration accuracy of 1.7 mm for clinical data, compared to the ICP-based algorithm with an average accuracy of 2.8 mm. In phantom studies, the ICP-based algorithm proved more reliable, but with more complicated clinical data, the gradient difference algorithm was more robust. Average computation time for the ICP-based algorithm was 20 s per registration, compared with 14 min and 50 s for the gradient difference algorithm.

Guide wire reconstruction and visualization in 3DRA using monoplane fluoroscopic imaging

A method has been developed that, based on the guide wire position in monoplane fluoroscopic images, visualizes the approximate guide wire position in the three-dimensional (3-D) vasculature, that is obtained prior to the intervention with 3-D rotational X-ray angiography (3DRA). The method assumes the position of the guide wire in the fluoroscopic images is known. A two-dimensional feature image is determined from the 3DRA data. In this feature image, the guide wire position is determined in a two-step approach: a mincost algorithm is used to determine a suitable position for the guide wire, and subsequently a snake optimization technique is applied to move the guide wire to a better position. The resulting guide wire can then be visualized in 3-D in combination with the 3DRA dataset. The reconstruction accuracy of the method has been evaluated using a 3DRA image of a vascular phantom filled with contrast, and monoplane fluoroscopic images of the same phantom without contrast and with a guide wire inserted. The evaluation has been performed for different projection angles, and with different parameters for the method. The final result does not appear to be very sensitive to the parameters of the method. The average mean error of the estimated 3-D guide wire position is 1.5 mm, and the average tip distance is 2.3 mm. The effect of inaccurate C-arm geometry information is also investigated. Small errors in geometry information (up to 1 degrees) will slightly decrease the 3-D reconstruction accuracies, with an error of at most 1 mm. The feasibility of this approach on clinical data is demonstrated.

Endpoint localization in guide wire tracking during endovascular interventions1

RATIONALE AND OBJECTIVES

A method is presented to track guide wires during endovascular interventions under X-ray fluoroscopy. Accurate guide wire tracking can be used to improve guide wire visualization in the low quality fluoroscopic images, and to estimate the position of the guide wire in world coordinates for navigation purposes.

MATERIALS AND METHODS

A two-step procedure is used to track the guide wire in subsequent frames. First, the position of the guide wire is obtained by fitting a spline to the image. Subsequently, the spline is iteratively moved toward the tip of the guide wire for accurate tip localization. For both steps, a feature image is used in which line-like structures are enhanced. The method is validated using a reference standard, obtained by manual tracings of three observers.

RESULTS

The method is evaluated on 20 image sequences, 10 sequences with a J-tipped guide wire and 10 with a straight guide wire. The tracking success was 96% for J-tipped and 100% for straight guide wires, whereas accurate endpoint localization could be performed in 91.3% and 94.4% of the frames respectively, with a tip localization error of less than 1.5 mm.

CONCLUSIONS

Accurate endpoint localization can be performed for both J-tipped and straight guide wires and therefore the presented tracking method can be used for navigation purposes.

Three-dimensional guide-wire reconstruction from biplane image sequences for integrated display in 3-D vasculature

Using three-dimensional rotational X-ray angiography (3DRA), three-dimensional (3-D) information of the vasculature can be obtained prior to endovascular interventions. However, during interventions, the radiologist has to rely on fluoroscopy images to manipulate the guide wire. In order to take full advantage of the 3-D information from 3DRA data during endovascular interventions, a method is presented that yields an integrated display of the position of the guide wire and vasculature in 3-D. The method relies on an automated method that tracks the guide wire simultaneously in biplane fluoroscopy images. Based on the calibrated geometry of the C-arm, the 3-D guide-wire position is determined and visualized in the 3-D coordinate system of the vasculature. The method is evaluated in an intracranial anthropomorphic vascular phantom. The influence of the angle between projections, distortion correction of the projection images, and accuracy of geometry knowledge on the accuracy of 3-D guide-wire reconstruction from biplane images is determined. If the calibrated geometry information is used and the images are corrected for distortion, a mean distance to the reference standard of 0.42 mm and a tip distance of 0.65 mm is found, which means that accurate guide-wire reconstruction from biplane images can be performed.