Transcatheter aortic valve implantation (TAVI) is a minimally invasive procedure to treat severe aortic stenosis in patients with a high risk for conventional surgery. In-silico experiments of stent deployment within patient-specific models of the aortic root have created an opportunity to predict stent behavior during the intervention. Current limitations in procedure planning are a primary motivator for these simulations. The virtual stent placement preceding the deployment phase of such experiments has major influence on the outcome of the simulation, but only received little attention in literature up to now. This work presents a methodical approach to patient-specific planning of placement of self-expanding stent models by analyzing experimental outcomes of different sets of boundary conditions constraining the stent. As a results, different paradigms for automated or expert guided stent placement are evaluated, which demonstrate the benefits of virtual stent deployment for intervention planning. To build a predictive planning pipeline for TAVI we use an automatic segmentation of the aorta, aortic root and left ventricle, which is converted to a finite element mesh. The virtual stent is then placed along a guide wire model and deployed at multiple locations around the aortic root. The simulation has been evaluated using pre- and post-interventional CT scans with an average relative circumferential error of 4.0% (±2.5%), which is less than half of the average difference in circumference between individual stent sizes (8.6%). Our methods are therefore enabling patient-specific planning and provide better guidance during the intervention.
Effective and safe performance of cardiovascular interventions requires excellent catheter / guidewire manipulation skills. These skills are mainly gained through an apprenticeship on real patients, which may not be safe or cost-effective. Computer simulation offers an alternative for core skills training. However, replicating the physical behaviour of real instruments navigated through blood vessels is a challenging task.We use an inextensible Cosserat rod and impulse-based techniques to model virtual catheters and guidewires. This allows an efficient recreation of bending, stretching and twisting phenomena of the material in real-time. It also guarantees an immediate response to user manipulations even for long instruments. The mechanical parameters of six guidewires and three catheters were optimized with respect to their real counterparts scanned in a silicone phantom using CT.The validation results show near sub-millimetre accuracy with an average distance error between the trajectories of the simulated and scanned instruments of 1.34mm (standard deviation: 0.95mm, RMS: 1.66mm). Our implementation requires just 0.2ms per time step to process 200 Cosserat elements on an off-the-shelf laptop, enabling simulation of 40cm long instruments at 4 kHz, thus significantly exceeding the minimum required haptic interactive rate (1 kHz).
Modeling the flowing blood in vascular structures is crucial to perform in silico simulations in various clinical contexts. This remains however an emerging and challenging research field, that raises several open issues. In particular, a compromise is generally made between the completeness of the simulation and the complicated architecture of the vasculature: reduced order simulations (lumped parameter models) represent vascular networks, whereas detailed models are devoted to small regions of interest. However, technical improvements enable targeting of compartments of the blood circulation rather than focusing on vascular branched segments. This article aims at investigating the cerebral flow in the entire venous drainage that can be reconstructed from medical imaging.
We present the design, development and initial user testing of a virtual reality simulator to train orthopaedic surgeons in the optimal placement of K-wires for fixation of distal radius fractures. Our platform includes 5 DOF haptic feedback to recreate the manual skill aspects of the drilling process, a 3D view of the anatomy and a controllable x-ray image. Once complete, the user is given an overview of their performance compared with the 'ideal placement' defined by an expert orthopaedic surgeon. The design goals based on analysis of the core steps in the procedure are presented, along with the technical implementation in terms of both haptic and graphical feedback. Preliminary user testing results are discussed, together with current limitations and planned future development.
Anatomically detailed modeling of soft tissue structures such as the forehead plays an important role in physics based simulations of facial expressions, for surgery planning, and implant design. We present ultrasound measurements of through-layer tissue deformation in different regions of the forehead. These data were used to determine the local dependence of tissue interaction properties in terms of variations in the relative deformation between individual layers. A physically based finite element model of the forehead is developed and simulations are compared with measurements in order to validate local tissue interaction properties. The model is used for simulation of forehead wrinkling during frontalis muscle contraction.
Successful bone sawing requires a high level of skill and experience, which could be gained by the use of Virtual Reality-based simulators. A key aspect of these medical simulators is realistic force feedback. The aim of this paper is to model the bone sawing process in order to develop a valid training simulator for the bilateral sagittal split osteotomy, the most often applied corrective surgery in case of a malposition of the mandible. Bone samples from a human cadaveric mandible were tested using a designed experimental system. Image processing and statistical analysis were used for the selection of four models for the bone sawing process. The results revealed a polynomial dependency between the material removal rate and the applied force. Differences between the three segments of the osteotomy line and between the cortical and cancellous bone were highlighted.
Digital Rectal Examination (DRE) plays a crucial role for diagnosing anorectal and prostate abnormalities. Despite its importance, training and learning is limited due to their unsighted nature. Haptics and simulation offer a viable alternative for enhancing the learning experience by allowing the trainees to train in safety whilst trainers are able to assess competency. We present results of our geometrical, deformation and haptics modelling for two key anatomical structures obtained from patient specific MRI scans, namely the rectum and the prostate. Rectum mobility and hardness are modelled via a centreline consisting of control and structure points that are ruled by a mass-spring model based on elastic energy. Prostate mobility, hardness, deformability and friction are modelled via a surface model consisting of colliding spheres interconnected by springs with elongation, flexion and torsion properties. Clinical input and model fine-tuning was provided by three consultants from clinical disciplines that routinely perform DREs. Our approach is modular with scope to support additional palpable anatomical structures and the potential to be used as a teaching and learning tool for DRE.
We present here an analysis of the possible advantages of using a priori model order reduction techniques for real-time simulation in computational surgery. Special attention will be paid to methods based upon Proper Generalized Decomposition techniques. These techniques allow for impressive savings in on-line computations by obtaining offline a reduced-order approach to the problem at hand. This approach can be seen as a particular instance of meta-model, response surface or - as we have coined it-a sort of computational vademecum. In this work we detail the approach followed for the implementation of essential aspects in computational surgery, such as solid dynamics and contact detection.
Morphological changes of the brain lateral ventricles are known to be a marker of brain atrophy. Anatomically, each lateral ventricle has three horns, which extend into the different parts (i.e. frontal, occipital and temporal lobes) of the brain; their deformations can be associated with morphological alterations of the surrounding structures and they are revealed as complex patterns of their shape variations across subjects. In this paper, we propose a novel approach for the ventricular morphometry using structural feature descriptors, defined on the 3D shape model of the lateral ventricles, to characterize its shape, namely width, length and bending of individual horns and relative orientations between horns. We also demonstrate the descriptive ability of our feature-based morphometry through statistical analyses on a clinical dataset from a study of aging.
The aim of this work is to build a 3D geometric and mechanical model of the skin/subcutaneous complex (SSC) which could be adapted to the different parts of the body and to the morphological parameters of the patient. We present first the anatomical pattern of the SSC. Then, we propose a hybrid model which combines volume, membranous and unidimensional models. The complex internal structure of the SSC is automatically created by a procedural process. All the models are defined by some parameters which can be easily measured by medical imaging. We describe several preliminary experiments which show how this hybrid method models realistic geometrical deformations and physical behaviors and could be used for surgery simulation and planning.
This paper presents a method to interactively deform volume images with heterogeneous structural content, using coarse tetrahedral meshes. It rests on two major components: a massively parallel algorithm for the rasterization of tetrahedral meshes, and a method to define a coarse deformable tetrahedral mesh from the homogenization of a fine heterogeneous mesh. We show the potential of the method for training and planning applications through two examples: an abdominal CT exploration and the alignment of breast CT and MRIs.
We propose in this paper a new method for tongue tracking in ultrasound images which is based on a biomechanical model of the tongue. The deformation is guided both by points tracked at the surface of the tongue and by inner points of the tongue. Possible uncertainties on the tracked points are handled by this algorithm. Experiments prove that the method is efficient even in case of abrupt movements.
During stereotactic neurosurgery, the brain shift could affect the accuracy of the procedure. However, this deformation of the brain is not often considered in the pre-operative planning step or intra-operatively, and may lead to surgical complications, side effects or ineffectiveness. In this paper, we present a method to update the pre-operative planning based on a physical simulation of the brain shift. Because the simulation requires unknown input parameters, the method relies on a parameter estimation process to compute the intracranial state that matches the partial data taken from intra-operative modalities. The simulation is based on a biomechanical model of the brain and the cerebro-spinal fluid. In this paper, we show on an anatomical atlas that the method is numerically sound.
Periacetabular osteotomy (PAO) is an effective approach for surgical treatment of hip dysplasia. The aim of PAO is to increase acetabular coverage of the femoral head and to reduce contact pressures by reorienting the acetabulum fragment after PAO. The success of PAO significantly depends on the surgeon’s experience. Previously, we have developed a computer-assisted planning and navigation system for PAO, which allows for not only quantifying the 3D hip morphology for a computer-assisted diagnosis of hip dysplasia but also a virtual PAO surgical planning and simulation. In this paper, based on this previously developed PAO planning and navigation system, we developed a 3D finite element (FE) model to investigate the optimal acetabulum reorientation after PAO. Our experimental results showed that an optimal position of the acetabulum can be achieved that maximizes contact area and at the same time minimizes peak contact pressure in pelvic and femoral cartilages. In conclusion, our computer-assisted planning and navigation system with FE modeling can be a promising tool to determine the optimal PAO planning strategy.
Subject-specific models of the musculoskeletal system are capable of accurately estimating function and loads and show promise for clinical use. However, creating subject-specific models is time-consuming and requires high levels of expertise. To address these issues, we have developed the open source Musculoskeletal Atlas Project (MAP) Client software. The MAP Client provides a user-friendly interface for creating musculoskeletal modelling workflows using community-created plug-ins. In this paper, we discuss the design of the MAP Client, its plug-in architecture and its integration with the Physiome Model Repository. We demonstrate the use of MAP Client with a subject-specific femur modelling workflow using a set of modular open source plug-ins for image segmentation, landmark prediction, model registration and customisation. Our long-term goal is to foster a community of MAP users and plug-in developers to accelerate the clinical use of computational models.
The purpose of our work is to develop tools for simulation of angiographic MRI images of cerebral vasculature. We present here our extension of the open-source software Jemris dedicated to this goal. Jemris is an advanced simulation software including most of MRI physical phenomena (concomitant gradient fields, molecular diffusion, patient move, etc). Our work now provides the additional ability to simulate fluids in motion, in addition to static tissues. These changes allow us to obtain velocimetric phase contrast images for simple Poiseuille flow, with multi-directional speed encoding. Comparison between our simulations and real MRI data gives promising first results.
Each year in Europe 50,000 new liver cancer cases are diagnosed for which hepatic surgery combined to chemotherapy is the most common treatment. In particular the number of laparoscopic liver surgeries has increased significantly over the past years. This type of minimally invasive procedure which presents many benefits for the patient is challenging for the surgeons due to the limited field of view. Recently new augmented reality techniques which merge preoperative data and intraoperative images and permit to visualize internal structures have been proposed to help surgeons during this type of surgery. One of the difficulties is to align preoperative data with the intraoperative images. We propose in this paper a semi-automatic approach for solving the ill-posed problem of initial alignment for Augmented Reality systems during liver surgery. Our registration method relies on anatomical landmarks extracted from both the laparoscopic images and three-dimensional model, using an image-based soft-tissue reconstruction technique and an atlas-based approach, respectively. The registration evolves automatically from a quasi-rigid to a non-rigid registration. Furthermore, the surface-driven deformation is induced in the volume via a patient specific biomechanical model. The experiments conducted on both synthetic and in vivo data show promising results with a registration error of 2 mm when dealing with a visible surface of 30% of the whole liver.
Recent progress in cardiac catheterization and devices allowed to develop new therapies for severe cardiac diseases like arrhythmias and heart failure. The skills required for such interventions are still very challenging to learn, and typically acquired over several years. Virtual reality simulators can reduce this burden by allowing to practice such procedures without consequences on patients. In this paper, we propose the first training system dedicated to cardiac electrophysiology, including pacing and ablation procedures. Our framework involves an efficient GPU-based electrophysiological model. Thanks to an innovative multithreading approach, we reach high computational performances that allow to account for user interactions in real-time. Based on a scenario of cardiac arrhythmia, we demonstrate the ability of the user-guided simulator to navigate inside vessels and cardiac cavities with a catheter and to reproduce an ablation procedure involving: extra-cellular potential measurements, endocardial surface reconstruction, electrophysiology mapping, radio-frequency (RF) ablation, as well as electrical stimulation. This works is a step towards computerized medical learning curriculum.
Realistic medical procedure simulators improve the learning curve of the clinicians if they can reproduce real conditions and use. This paper describes the improvement of a transrectal ultrasound guided prostate biopsy simulator by adding the simulation of real-time prostate movements and deformations. A discrete bio-mechanical model is used to modify a 3D texture of an ultrasound image volume in order to quickly simulate the actual displacements and deformations. This paper describes this model and presents how the mesh deformation is used to induce the UltraSound volume deformation. The validation of the method is based on both a quantitative and a qualitative assessment. Experimental images acquired on a phantom are compared using mutual information metrics to the resulting generated images. This comparison shows that the proposed method offers realistic deformed 3D ultrasound images at interactive time. The method was successfully integrated to improve the transrectal ultrasound simulator.
In this paper, we present Bender, an interactive and freely available software application for changing the pose of anatomical models that are represented as labeled, voxel-based volumes.Voxelized anatomical models are used in numerous applications including the computation of specific absorption rates associated with cell phone transmission energies, radiation therapy, and electromagnetic dosimetry simulation. Other applications range from the study of ergonomics to the design of clothing. Typically, the anatomical pose of a voxelized model is limited by the imaging device used to acquire the source anatomical data; however, absorption of emitted energies and the fit of clothes will change based on anatomic pose.Bender provides an intuitive, workflow-based user-interface to an extensible framework for changing the pose of anatomic models. Bender is implemented as a customized version of 3D Slicer, an image analysis and visualization framework that is widely used in the medical computing research community. The currently available repositioning methods in Bender are based on computer-graphics techniques for rigging, skinning, and resampling voxelized anatomical models. In this paper we present the software and compare two resampling methods: a novel extension to dual quaternions and finite element modeling (FEM) techniques. We show that FEM can be used to quickly and effectively resample repositioned anatomic models.
