Xiao-Liang Xie

Haptic interfaces for the rescue walking robots motion in the disaster areas

In the recent years haptic interfaces became a reliable solution in order to solve problems which arise when humans interact with the environment. If in the research area of the haptic interaction between human and environment there are important researches, a innovative approach for the interaction between the robot and the environment using haptic interfaces and virtual projection method is presented in this paper. In order to control this interaction we used the Virtual Projection Method where haptic control interfaces of impedance and admittance will be embedded. The obtained results, validated by simulations assure stability, stiffness, high maneuverability and adaptability for rescue walking robots in order to move in disaster, dangerous and hazardous areas.

A 3-DOF compact haptic interface for endoscopic endonasal approach surgery simulation

Endoscopic endonasal approach surgery is now the preferred treatment for most pituitary and related skull base tumors. However, this procedure requires a high level of hands-on skills and rich clinical experience. During the operation, haptic feedback, as the only one sense of bidirectional information interaction, plays an important role in surgical decision-making especially for bone-drilling. Existing surgical simulators provide either no haptic device or multipurpose haptic devices, which is difficult to reproduce the characteristics of surgical tool handling. In this paper, a custom-designed 3-DOF (pitch, yaw, radial) compact haptic interface for this surgery simulation is presented. It is dedicated to mimicking the touch sense of the surgical tools inserted through the nostril. Its main innovation is the mechanism design to maintain as much fidelity of the tool handling in the surgical training as in a real operation. The mechanism design is presented in detail as well as the kinematics and the force transmission. The mechanical characteristics of this haptic interface are also analyzed and presented.

FEM-based guide wire simulation and interaction for a minimally invasive vascular surgery training system

Minimally invasive vascular surgery is an important surgical procedure which requires specialized skills to manipulate surgical instruments. However, the insufficient conventional training paradigm of “see one, do one, teach one” cannot meet the urgent need for training surgeons. In this paper we present a novel computer-based training system which aims at helping the trainees with acquiring manipulation skills and gaining simulated surgical experience. This system integrates a physics model of guide wire, a 3D vasculature segmented from real patients and a self-designed haptic device. We use the finite element method (FEM) to model the guide wire based on its specific physical parameters. Trainees can advance and rotate the simulated instrument by the haptic device to interact with the human vasculature. The experimental results of delivering the guide wire to the entrance of the coronary in real time validated that the guide wire simulation is effective.

Guide-wire detecting using a modified cascade classifier in interventional radiology

Endovascular surgery is becoming a widespread procedure to treat cardiovascular diseases (CVDs) such as abdominal aortic aneurysm and peripheral artery disease. The guide-wire is a crucial surgical instrument inserted into vessels to offer guidance to physicians during the surgery. There are some approaches for tracking the guide-wire, most algorithms consist of two phases, namely, the initialization phase and the tracking phase. In the initialization phase, most algorithms use B-splines for modeling the guide-wire which requires manually annotated data. In the tracking phase, the guide-wire motion is non-linearity because it is deforming and changing its shape and size as a result of patients respiration, some algorithms decompose the non-linearity motion into rigid motion and non-rigid motion, while the computational complexity is high especially for the non-rigid motion. This paper mainly presents an approach to detect the guide-wire. The algorithm has two main advantages. First, without modeling the guide-wire, this approach uses a cascade classifier which can detect the guide-wire under arbitrary motion automatically. Second, by taking the guide-wire motion direction into consideration, the detection accuracy improves significantly. The presented work has been validated on a test set of 349 frames, and the mean tracking accuracy achieves more than 95% which proves the effectiveness of the proposed method.Endovascular surgery is becoming a widespread procedure to treat cardiovascular diseases (CVDs) such as abdominal aortic aneurysm and peripheral artery disease. The guide-wire is a crucial surgical instrument inserted into vessels to offer guidance to physicians during the surgery. There are some approaches for tracking the guide-wire, most algorithms consist of two phases, namely, the initialization phase and the tracking phase. In the initialization phase, most algorithms use B-splines for modeling the guide-wire which requires manually annotated data. In the tracking phase, the guide-wire motion is non-linearity because it is deforming and changing its shape and size as a result of patients respiration, some algorithms decompose the non-linearity motion into rigid motion and non-rigid motion, while the computational complexity is high especially for the non-rigid motion. This paper mainly presents an approach to detect the guide-wire. The algorithm has two main advantages. First, without modeling the guide-wire, this approach uses a cascade classifier which can detect the guide-wire under arbitrary motion automatically. Second, by taking the guide-wire motion direction into consideration, the detection accuracy improves significantly. The presented work has been validated on a test set of 349 frames, and the mean tracking accuracy achieves more than 95% which proves the effectiveness of the proposed method.

Centerlines extraction for lumen model of human vasculature for computer-aided simulation of intravascular procedures

The computer-aided surgical simulation aims to provide an economic tool of effectiveness and convenience for the training process. In building this simulation system, the construction of the virtual anatomic environment is one of the major tasks. It provides the virtual tools with the scenario in which they are manipulated by the trainee. In intravascular surgery simulation, the surface model of the blood vessels is the most important part of the virtual environment. In order to achieve better performances in the simulation of path planning and navigation, the surface model based on real patients CTA data needs further process. We proposed in this paper an approach to extract the centerlines of each segment of the image-based surface model of the blood vessels. The surface model is firstly processed to check the connectivity of the consisting polygons in order to extract the largest connected region within the surface. Next, the resulting surface is smoothed by a windowed sinc function kernel with proper parameters. After the normal vectors of the smoothed surface are computed, the surface is subdivided and the centerlines of the surface model are computed by using the power crust algorithm. The experimental results show that the approach is capable of extracting the centerlines of the vessel model.

Prediction of natural guidewire rotation using an sEMG-based NARX neural network

For the treatment of cardiovascular diseases, clinical success of percutaneous coronary intervention is highly dependent on natural technical skills and dexterous manipulation strategies of surgeons. However, the increasing used robotic surgical systems have been designed without considering manipulation techniques, especially surgical behaviors and motion patterns. This has driven research towards exploitation of natural manipulation skills in recent years. In this paper, natural guidewire manipulations are analyzed and predicted using an sEMG-based nonlinear autoregressive neural network with exogenous inputs. The relationship between natural endovascular manipulation and guidewire rotation is built through the network. Two experiments at different rotational speed were performed to verify the effectiveness and robustness of the applied model. The experimental results show that the average predictive root mean error of five subjects is 15.61° at the low speed and 21.85° at the high speed. These favorable results could be of interest to improve existing robotic surgical systems.

Guide-wire detection using region proposal network for X-ray image-guided navigation

Detection of surgical devices, in particular of guide-wire detection, is prerequisite during image-guided navigation in percutaneous coronary intervention (PCI). Guide-wire detection is a challenging task for following reasons: (i) X-ray images have a low signal-to-noise rate (SNR); (ii) there is a high similarity between guide-wires and some other adjacent anatomical skeletons contours; (iii) guide-wires have various shapes and their motion is complex and nonlinear. Traditionally, guide-wires are detected using curve fitting method, and third-order B-spline curve model is always used to fit guide-wires, while B-spline fitting method has some obvious shortcomings such as it is a semi-automatic method which needs manual initialization, and it is not a real-time method because of high computational complexity. Recently, with the availability of large annotated datasets and the accessibility of hardware resources with GPUs, it is succeeded in detecting general objects with convolutional neural networks (ConvNet). In this paper, we present a novel image-based fully-automatic and real-time approach with ConvNet for guide-wires detection. ConvNet method is robust to guide-wires various poses and other structures effects. We evaluate our method on 22 different sequences of X-ray images. The detection accuracy evaluated by average precision (AP) reaches 89.2% and the detection speed achieves 40fps. Our experiment result shows a promising for accurate and real-time guide-wires detection in PCI navigation with ConvNet model.

An HMM-based recognition framework for endovascular manipulations

Robotic surgical systems are becoming increasingly popular for the treatment of cardiovascular diseases. However, most of them have been designed without considering techniques and skills of natural surgical manipulations, which are key factors to clinical success of percutaneous coronary intervention. This paper proposes an HMM-based framework to recognize six typical endovascular manipulations for surgical skill analysis. A simulative surgical platform is built for endovascular manipulations assessed by five subjects (1 expert and 4 novices). The performances of the proposed framework are evaluated by three experimental schemes with the optimal model parameters. The results show that endovascular manipulations are recognized with high accuracy and reliable performance. Furthermore, the acceptable results can also be applied to the design of next generation vascular interventional robots.Robotic surgical systems are becoming increasingly popular for the treatment of cardiovascular diseases. However, most of them have been designed without considering techniques and skills of natural surgical manipulations, which are key factors to clinical success of percutaneous coronary intervention. This paper proposes an HMM-based framework to recognize six typical endovascular manipulations for surgical skill analysis. A simulative surgical platform is built for endovascular manipulations assessed by five subjects (1 expert and 4 novices). The performances of the proposed framework are evaluated by three experimental schemes with the optimal model parameters. The results show that endovascular manipulations are recognized with high accuracy and reliable performance. Furthermore, the acceptable results can also be applied to the design of next generation vascular interventional robots.

Development of a multi-modal interactive system for Endoscopic Endonasal Approach surgery simulation

Endoscopic Endonasal Approach (EEA) surgery, as a minimally invasive technique, is now the preferred treatment for most pituitary and skull base tumors. This procedure uses the nostrils as natural corridors to remove relevant lesions, which requires a high level hands-on skills and rich clinical experience. With the aid of virtual reality technology, virtual surgical environments can provide a new solution for surgical training, planning, and rehearsal. In this paper, a multi-modal interactive system for EEA surgery simulation is presented. Due to the complex surgical procedures, the bone drilling task is identified as the core part for neurosurgeons to train and build their skills. A multi-modal interactive system providing visual, haptic and audible feedback is developed for bone drilling task. Besides, a customized 3-DOF haptic interface is utilized to control a virtual drill in the virtual environment rendered with patient-specific preoperative image data set. The preliminary experiment shows that this multi-modal interactive system can provide an immersive experience for neurosurgeons to perform the bone drilling task in a cost effective way.

Lumen segmentation and visualization of abdominal aorta using geodesic active contours for intravascular surgical simulation

Percutaneous transluminal coronary angioplasty (PTCA) has been proved to be a standard solution to most cardiovascular diseases (CVDs). The surgical simulator provides the trainees a new vehicle to learn this skill much more conveniently and effectively. The blood vessel model is at the core of the virtual environment. In this paper, a robust and semi-automatic approach to segment the abdominal aorta from the computed tomography angiography (CTA) is developed. The proposed approach employs the geodesic active contours method as the main component. The edge potential map is generated by applying nonlinear mapping function. The initial contours are evolved by applyging the fast marching method. The surface information representing the vessel is extracted by the marching cubes method. This approach has been proved successful for the construction of 3-D surface model of the aorta based on the CTA series.