In this paper, we propose an intelligent coordinate registration method. This method simplifies and automates the registration procedure between image space coordinates and patient space coordinates with a high accuracy. Compared with paired-point registration, three-dimensional registration can be performed with only one Navpass model instead of three or more points. The Navpass is a real-world model designed by our group, which can be easily detected in Computed Tomography (CT) images. First, the pose information such as position and orientation of the Navpass is detected automatically on the images by our Dynamic Region Growing (DRG) algorithm. Then, we calculate the registration transformation when the corresponding location in the patient coordinates obtained by electromagnetic tracking device in real time. In the navigation procedure, the method is capable of computing the registration error in real-time during the respiratory cycle and assisting the surgeon to insert needles within the minimal error. When the target registration error is beyond a certain threshold, the registration matrix would be updated automatically. Our method has been tested on a real-world navigation system with a specially designed phantom. Based on the practical and extensive experiments, we conclude that our method can provide fast and accurate registration error, and that it helps reduce the operation time dramatically for the surgeons as well as the patients.

In this paper, we propose a dynamic coordinate registration method. This method simplifies and automates the registration procedure between image space coordinates and patient space coordinates with a high accuracy. Compared with paired-point registration, three-dimensional registration can be performed with only one Navpass model instead of three or more points. The Navpass is a real-world model designed by our group, which can be easily detected in Computed Tomography (CT) images. First, the pose information such as position and orientation of the Navpass is detected automatically on the images by our Dynamic Region Growing (DRG) algorithm. Then, we calculate the registration transformation when the corresponding location in the patient coordinates obtained by electromagnetic tracking device in real time. In the navigation procedure, the method is capable of computing the registration error in real-time during the respiratory cycle and assisting the surgeon to insert needles within the minimal error. When the target registration error is beyond a certain threshold, the registration matrix would be updated automatically. Our method has been tested on a real-world navigation system with a specially designed phantom. Based on the practical and extensive experiments, we conclude that our method can provide fast and accurate registration with a 1.7 ± 0.03 mm target registration error, and that it helps reduce the operation time dramatically for the surgeons as well as the patients.

In this paper, we aim to build a 3D medical image processing based surgical navigation system, to assist the operation with smooth performance. We first focus on the parallel implementation of Graph Cuts algorithm under CUDA for organ segmentation. The algorithm is modified and optimized and satisfactory results are obtained. A GPU parallel accelerated Marching Cubes algorithm for visualization is designed and experimental results are given. A surgical planning method is also developed, which can effectively reduce the risk of the radiation that the patients receive and accidental injury of the puncture needles in the traditional surgery. Two different navigation equipment's connection and communication are implemented with our system. The above main functions are integrated to a surgical navigation system. Some experiments are made (animal experiments included) to test this platform for accuracy, reliability and ease of use.

In this paper, we propose a dynamic coordinate registration method. This method simplifies and automates the registration procedure between image space coordinates and patient space coordinates with a high accuracy. Compared with paired-point registration, three-dimensional registration can be performed with only one Navpass model instead of three or more points. The Navpass is a real-world model designed by our group, which can be easily detected in Computed Tomography (CT) images. First, the pose information such as position and orientation of the Navpass is detected automatically on the images by our Dynamic Region Growing (DRG) algorithm. Then, we calculate the registration transformation when the corresponding location in the patient coordinates obtained by electromagnetic tracking device in real time. In the navigation procedure, the method is capable of computing the registration error in real-time during the respiratory cycle and assisting the surgeon to insert needles within the minimal error. When the target registration error is beyond a certain threshold, the registration matrix would be updated automatically. Our method has been tested on a real-world navigation system with a specially designed phantom. Based on the practical and extensive experiments, we conclude that our method can provide fast and accurate registration with a 1.7±0.03 mm target registration error, and that it helps reduce the operation time dramatically for the surgeons as well as the patients.

A novel preoperative surgery planning method is proposed for microwave ablation. An iterative framework for necrosis field simulation and 3D necrosis zone reconstruction is introduced here. The necrosis field of the ablation is computed with an adaptable method based on surgery trajectories, and then the 3D model of the necrosis zone is reconstructed and superimposed to the patient anatomy structures using advanced visualization techniques GPU-accelerated algorithms. The full surgery planning with multiple antennae is performed by the operator in an interactively way, until the optimal surgery plan is achieved. Experiments have been performed and the actual necrosis field is measured by comparing to postoperative CT images. Results show that this method is relative accurate for preoperative trajectory plan and could be used as an assistant to the clinical practice.

The paper aims at providing an intelligent driver assistance system on mobile devices, which is mainly composed of the two parts - off road warning and distance warning of front vehicles. In this application, the paper proposes an efficient lane and front vehicles detection and tracking method which has been experiments on mobile phones. The method is proved to be efficient and practical even though the compute capacity is limited on mobile phones.

A novel preoperative surgery planning method is proposed for microwave ablation. An iterative framework for necrosis field simulation and 3D necrosis zone reconstruction is introduced here. The necrosis field of the ablation is computed with an adaptable method based on surgery trajectories, and then the 3D model of the necrosis zone is reconstructed and superimposed to the patient anatomy structures using advanced visualization techniques. The full surgery planning with multiple antennae is performed by the operator in an interactively way, until the optimal surgery plan is achieved. Experiments have been performed and the actual necrosis field is measured by comparing to postoperative CT images. Results show that this method is relative accurate for preoperative trajectory plan and could be used as an assistant to the clinical practice.

A novel preoperative surgery planning method is proposed for microwave ablation. An iterative framework for necrosis field simulation and 3D necrosis zone reconstruction is introduced here. The necrosis field of the ablation is computed with an adaptable method based on surgery trajectories, and then the 3D model of the necrosis zone is reconstructed and superimposed to the patient anatomy structures using advanced visualization techniques GPU-accelerated algorithms. The full surgery planning with multiple antennae is performed by the operator in an interactively way, until the optimal surgery plan is achieved. Experiments have been performed and the actual necrosis field is measured by comparing to postoperative CT images. Results show that this method is relative accurate for preoperative trajectory plan and could be used as an assistant to the clinical practice.

This paper proposes a circular road sign detection and analysis method based on monocular vision, which can be used for visual localization of mobile robots. To achieve a stable and reliable real-time detection, a simple but useful circular road sign has been designed. By color-based image segmentation, edge detection, contour selection, ellipse fitting and sign verification, road sign can be detected from monocular vision. Furthermore, perspective geometry is carried out to calculate the robots position based on sign detection result and sign position. The proposed method is experimented on a robot platform. The experimental results show that the method achieves a promising performance and meets the requirement of real-time and practical situations.

An optimization design method based on 3-D inverse design method and CFD was developed by combining design of experiment method, response surface method and simulated annealing optimization calculation. The method was applied to a centrifugal fan blade design by setting the impeller efficiency and the circulation distribution parameters as the optimization object and the optimization design variables respectively. The efficiency was improved by 2.2% with the optimization method. The effects of the circulation parameters and their interaction effects on the impeller efficiency were analyzed according to the constructed response surface between the design parameters and the impeller efficiency. It is shown that the inlet circulation distribution on the hub plays an important role on the efficiency. Comparing with the hub, the circulation distribution shape on the shroud has more influences on the efficiency.

Radioactive particle implantation, as a means of interventional therapy in the treatment of malignant tumors, has achieved good results and rapid development in recent years. This technique uses computers to process CT images and provide doctors with real-time navigation. We aim to design and build a 3D image based operation navigation platform to assist the implementation of interventional operations. Features of our operation navigation platform include organ segmentation, operation planning and operation navigation. We used the improved graph cuts algorithm to ensure the speed and accuracy of segmentation; we designed an algorithm to find the range of target, and automatically compute the ideal positions of radioactive particles; we connected our operation platform with a navigation equipment to implement the real-time navigation of the position and orientation of the probe. Results on real CT images of patients demonstrate the effectiveness of the platform.

为了方便临床医生使用医学影像进行手术的规划和引导,提出了一种新颖的计算机辅助肝癌微波消融手术方法.采用基于GPU加速的三维可视化技术实时重构影像的三维模型,显示出患者器官的解剖关系;采用基于生物传热学的三维热场计算技术,规划手术路径及消融设备的作用时间和功率.经过大量的临床应用以及手术影像对比,表明该方法能够有效计算消融手术的作用效果,并在手术的引导过程中发挥了重要的作用,提高了手术的临床精度和效果.

Microwave ablation is a promising option in lung cancer therapy. However, it's rarely used in percutaneous lung cancer therapy compared to liver cancer, because the presence of a large amount of air within the lung creates significant back shadowing artifacts that preclude adequate delineation of anatomic details on sonography. To utilize microwave ablation in malignant lung tumor therapy, we developed a novel percutaneous intervention surgery navigation system (CAINS-I), which capitalizes on using computer assisted technology to help lung cancer patients whose condition are not amenable to surgical resection, sonographic guidance and intraoperative CT surgery. In these surgeries, preoperative CT images with patient respiration state are first acquired, which are then visualized using GPU-accelerated volume rendering. The optimal surgery trajectories are then planned based on 3D thermal field computation and surgery simulation in the surgery planning software. During the surgery, the patient breath is control by a portable volume ventilator system which could limit the movement and displacement of the tumor. Then the microwave probe is punctured into the tumor according to the dynamic respiratory state and the tumor is ablated by microwave energy. After the surgery, postoperative CT are acquired and compared to the preoperative CT, and the surgery is evaluated by compare preoperative and postoperative CT images. The development of this technique represented an advance from the traditional ways for lung cancer therapy and significantly extends the indications of microwave ablation.

Cognitive Science is widely studied as a high degree of cross-disciplinary. Various models are presented by several popular architectures. ACT-R (Adaptive Control of Thought-Rational) architecture is one of most popular cognitive architectures. A new combination of cognitive science and artificial intelligence is studied in this paper. Signal serials in visual modules are usually artificial defined in many cognitive models, including driver cognitive models. Vision based automatic navigation is explored for visual module of driver cognitive model. Signal serials in visual modules are provided by automatic vision based navigation systems from real driving video, and used for driver model in ACT-R architecture. Results are compared. Experimental result shows that the applying of vision navigation on cognitive model in ACT-R architecture is reliable.

In this paper,Laplacian of Gaussian(LoG) of contour curves is applied and its norm square shows a robust curvature feature.So,norm square of contour LoG is defined as the response function of corners.Further,a novel covariant region detector with rotation and scale invariants is designed based on the contour orientation of each point at neighbor of each corner detected by LoG and the corner itself.Finally,matching among image feature covariant regions is performed using cross correlation criterion and the results illustrate that the proposed algorithm has the properties such as simple calculation,easy realization,and better robustness.

Liver segmentation remains a difficult problem in medical images processing, especially when accuracy and speed are both seriously considered. Graph Cuts is a powerful segmentation tool through which the optimal results are got by considering both region and boundary information in images. However, the traditional Graph Cuts algorithms are always computationally expensive and inappropriate to be applied to real clinical circumstance. Recently, the GPU (Graphics Processor Unit) had evolved to be a cheap and superpower general purpose computing instrument, especially when NVIDIA released its revolutionary CUDA (Compute Unified Device Architecture). In this paper, we introduce a novel method to segment 3D liver images with GPU, using the Push-Relable style 3D Graph Cuts implementation. Some modifications such as 3D storage structures are also introduced which make our implement well fit to the GPU parallel computing capabilities. Experiments have been executed on human liver CT data and these experiments show that our method can obtains results in much less time compared to the implement with CPU.

A novel CUDA based high performance parallel voxel growing algorithm to segment 3D CT pulmonary volumes with GPU Acceleration is introduced in this paper. The optimal parameters for segmentation is dynamically iterative adjusted based on the statistical information about previous segmented regions. To avoid the disadvantage of leaking during segmentation with the conventional voxel-growing based methods, it adopts a process to mutually utilize segment results between both of lateral lung leaves, which in turn benefits the discriminative segmentation on left and right lung leaves. Experiments show that the algorithms obtain accurate results with a speed about 10-20 times faster than the traditional methods on CPU, which imply that this algorithm is potentially valid for future clinical diagnosis applications.

Volumetric medical image visualization techniques are widely used in clinical environment, such as image based diagnosis, image guided surgery planning and navigation systems. Many traditional visualization algorithms had achieved great visualization result in research environment. However, image size in medical applications is often very large, and it leads to tremendous computation time which is not appropriate for clinical practice. Recently, the GPU has become an economical and fast parallel co-processor. A contemporary GPU can have up to hundreds of stream processors so it has the potential for solving massive data-parallel visualization problems. This paper aims to introduce some GPU-accelerated visualization techniques in medical image visualization with GPU, including slice based rendering, 3D image surface reconstruction, and direct volume rendering.

In this paper, we introduce a parallel algorithm to implement the Region Growing algorithms in GPU, with the purpose of 3D organ segmentation. Extensive Experiments have been executed on human CT Data, and these experiments show that the algorithms obtain accurate results with a speed about 10-20 times faster than the traditional methods on CPU. Several improvements to the traditional region growing algorithms are also introduced in this paper. This method is integrated in several surgery planning and surgery navigation systems and has achieved good clinical results.