scispace - formally typeset
Search or ask a question

Showing papers presented at "Computer Assisted Radiology and Surgery in 2006"


Journal ArticleDOI
09 Nov 2006
TL;DR: It is necessary to develop strategies for improving surgical/interventional workflows assisted by surgical systems and technologies and to require an appropriate IT infrastructure as well as communication and interface standards to allow data change between surgical system components in the OR.
Abstract: A recent report predicted an increase in demand for 2 surgical services to be as high as 14 to 47% in the 3 workload of all surgical fields by 2020 [1]. Difficulties 4 which are already now apparent in the operating room 5 (OR), such as the lack of seamless integration of Surgi6 cal Assist Systems (SAS) into the surgical workflow, will 7 be amplified in the near future. There are many SASs 8 in development or are employed in the OR, mostly in 9 an isolated fashion. Their routine use in the OR, how10 ever, is impeded by the absence of appropriate integra11 tion technology and standards. It is, therefore, necessary 12 to address this situation and to develop strategies for 13 improving surgical/interventional workflows assisted by 14 surgical systems and technologies. 15 Appropriate use of Information and Communication 16 Technology (ICT) and Mechatronic (MT) systems as 17 part of a re-engineered workflow is considered by many 18 experts as a significant contribution to solve the prob19 lem. This will require an appropriate IT infrastructure 20 as well as communication and interface standards, such 21 as DICOM and suitable extensions, to allow data inter22 change between surgical system components in the OR. 23

74 citations


Journal ArticleDOI
18 Jul 2006
TL;DR: A model for defining and reporting reference-based validation protocols in medical image processing, which facilitates the standardization of validation terminology and methodology, improves the comparison of validation studies and results, provides insight into the validation process, and may lead to better quality image management and decision making is proposed.
Abstract: Objectives. Image processing tools are often embedded in larger systems. Validation of image processing methods is important because the performance of such methods can have an impact on the performance of the larger systems and consequently on decisions and actions based on the use of these systems. Most validation studies compare the direct or indirect results of a method, with a reference that is assumed to be very close or equal to the correct solution. In this paper, we propose a model for defining and reporting reference-based validation protocols in medical image processing. Materials and Methods. The model was built using an ontological approach. Its components were identified from the analysis of initial publications (mainly reviews) on medical image processing, especially registration and segmentation, and from discussions with experts from the medical imaging community during international conferences and workshops. The model was validated by its instantiation for 38 selected papers that include a validation study, mainly for medical image registration and segmentation. Results. The model includes the main components of a validation procedure and their interrelationships. A checklist for reporting reference-based validation studies for medical image processing was also developed. Conclusion. The proposed model and associated checklist may be used in formal reference-based validation studies of registration and segmentation and for the complete and accurate reporting of such studies. The model facilitates the standardization of validation terminology and methodology, improves the comparison of validation studies and results, provides insight into the validation process, and, finally, may lead to better a quality management and decision making.

73 citations


Journal ArticleDOI
15 Nov 2006
TL;DR: OsiriX, a fast DICOM viewer program for the Apple Macintosh, illustrates how open software development for medical imaging tools can be successfully designed, implemented and disseminated.
Abstract: Purpose Open source software (oss) development for medical imaging enables collaboration of individuals and groups to produce high-quality tools that meet user needs. This process is reviewed and illustrated with OsiriX, a fast DICOM viewer program for the Apple Macintosh.

53 citations


Journal ArticleDOI
11 Oct 2006
TL;DR: It is shown that the daVinci system can successfully localize a point with an approximate fiducial localization error (FLE) of 1 mm, demonstrating that the Daisy is within the accuracy limits of the other commonly used surgical localizers.
Abstract: The use of robotics in medicine has been growing in recent years. The introduction of specialized robots for surgery such as the daVinci (Intuitive Surgical, Sunnyvale, CA, USA) has largely motivated this growth. The currently available robotic-assisted surgery systems rely only on direct vision through the use of a laparoscopic camera for surgical guidance. While these cameras provide a stereoscopic view of the surface, they are incapable of showing beneath the surgical surface. In other surgical developments, image guidance has become an integral part of navigation and planning, allowing understanding of the locations beneath the surgical plane. As an early step in the development of a robotic image-guided surgical (RIGS) system, the physical limitations of a surgical robot must be assessed. In this paper the accuracy and the precision of the daVinci surgical system are quantified. It is shown that the daVinci system can successfully localize a point with an approximate fiducial localization error (FLE) of 1 mm. This FLE value demonstrates that the daVinci is within the accuracy limits of the other commonly used surgical localizers.

51 citations


Journal ArticleDOI
11 Oct 2006
TL;DR: Automatic segmentation of the airway imaged using CBCT is feasible and this method can be used to evaluate airway cross-section and volume comparable to measurements extracted using manual segmentation.
Abstract: Objective To segment and measure the upper airway using cone-beam computed tomography (CBCT). This information may be useful as an imaging biomarker in the diagnostic assessment of patients with obstructive sleep apnea and in the planning of any necessary therapy. Methods With Institutional Review Board Approval, anonymous CBCT datasets from subjects who had been imaged for a variety of conditions unrelated to the airway were evaluated. DICOM images were available. A segmentation algorithm was developed to separate the bounded upper airway and measurements were performed manually to determine the smallest cross-sectional area and the anteriorposterior distance of the retropalatal space (RP-SCA and RP-AP, respectively) and retroglossal space (RG-SCA and RG-AP, respectively). A segmentation algorithm was developed to separate the bounded upper airway and it was applied to determine RP-AP, RG-AP, the smallest transaxial-sectional area (TSCA) and largest sagittal view airway area (LCSA). A second algorithm was created to evaluate the airway volume within this bounded upper airway. Results Measurements of the airway segmented automatically by the developed algorithm agreed with those obtained using manual segmentation. The corresponding volumes showed only very small differences considered clinically insignificant. Conclusion Automatic segmentation of the airway imaged using CBCT is feasible and this method can be used to evaluate airway cross-section and volume comparable to measurements extracted using manual segmentation.

50 citations


Journal ArticleDOI
17 Mar 2006
TL;DR: The focus of the International Journal of CARS is on providing a balanced and in-depth information on new diagnostic and therapeutic processes, including results from multidisciplinary R&D efforts, providers′ experiences, patient outcomes, economic considerations, and scientific/medical validation results.
Abstract: The international journal for computer assisted radiology and surgery (CARS) is a peer-reviewed journal that aims to provide a forum to close the gap between the medical and technical disciplines and to encourage interdisciplinary research and development activities in an international environment CARS has been defined by L Berliner as: “CARS: an international, academic society, guided by a sense of ethics, dedicated to bridging the gap between the disciplines of biomedical engineering and clinical medicine (including surgery, informatics, and diagnostic and interventional imaging), through the organization of scientific meetings and symposia to promote education and the exchange of new ideas, and to provide for the presentation and publication of original research” To increase the value of health care for the citizens, the focus of the International Journal of CARS is on providing a balanced and in-depth information on new diagnostic and therapeutic processes This includes results from multidisciplinary R&D efforts, providers′ experiences, patient outcomes, economic considerations, and scientific/medical validation results It can be expected that the resulting awareness by users and providers will speed up the acceptance of CARS into clinical practice The main emphasis in soliciting papers for the International Journal of CARS, however, will be on IT and mechatronic system design topics and clinical application fields, such as:

43 citations


Journal ArticleDOI
08 Mar 2006
TL;DR: A chain of algorithms are developed that lead to a closely related implementation of image segmentation with deformable models and 3D mesh generation that outperforms the existing methods in the majority of cases.
Abstract: Computational blood flow and vessel wall mechanics simulations for vascular structures are becoming an important research tool for patient-specific surgical planning and intervention. An important step in the modelling process for patient-specific simulations is the creation of the computational mesh based on the segmented geometry. Most known solutions either require a large amount of manual processing or lead to a substantial difference between the segmented object and the actual computational domain. We have developed a chain of algorithms that lead to a closely related implementation of image segmentation with deformable models and 3D mesh generation. The resulting processing chain is very robust and leads both to an accurate geometrical representation of the vascular structure as well as high quality computational meshes. The chain of algorithms has been tested on a wide variety of shapes. A benchmark comparison of our mesh generation application with five other available meshing applications clearly indicates that the new approach outperforms the existing methods in the majority of cases.

34 citations


Journal ArticleDOI
03 Mar 2006
TL;DR: Semi-automatic centerline extraction simplifies and automates determination of the plane orthogonal to one vessel, thereby permitting automatic simultaneous minimization of the misalignment with several vessels in PC MRA.
Abstract: The accuracy of 2D phase contrast (PC) magnetic resonance angiography (MRA) depends on the alignment between the vessels and the imaging plane. PC MRA imaging of blood flow is challenging when the flow in several vessels is to be evaluated with one acquisition. For this purpose, semi-automatic determination of the plane most perpendicular to several vessels is proposed based on centerlines extracted from 3D MRA. Arterial centerlines are extracted from 3D MRA based on iterative estimation-prediction, multi-scale analysis of image moments, and a second-order shape model. The optimal plane is determined by minimizing misalignment between its normal vector and the centerlines’ tangent vectors. The method was evaluated on a phantom and on 35 patients, by seeking the optimal plane for cerebral blood flow quantification simultaneously in internal carotids and vertebral arteries. In the phantom, difference of orientation and of height between known and calculated planes was 1.2° and 2.5 mm, respectively. In the patients, all but one centerline were correctly extracted and the misalignment of the plane was within 12° per artery. Semi-automatic centerline extraction simplifies and automates determination of the plane orthogonal to one vessel, thereby permitting automatic simultaneous minimization of the misalignment with several vessels in PC MRA.

33 citations


Journal ArticleDOI
20 Jun 2006
TL;DR: The study suggests an artificial neural network can be trained to make the correct interpretations of dental caries, and could be a prototype for caries detection but should be improved for classifying caries depth.
Abstract: Objects A neural network was developed to diagnose artificial dental caries using images from a charged-coupled device (CCD)camera and intra-oral digital radiography. The diagnostic performance of this neural network was evaluated against a gold standard.

31 citations


Journal ArticleDOI
03 Mar 2006
TL;DR: It is demonstrated that automatic STN localization is possible and accurate with non-rigid registration algorithms, and an affine registration with only 12 degrees of freedom is not enough for this application.
Abstract: Objects In functional stereotactic neurosurgery, one of the cornerstones upon which the success and the operating time depends is an accurate targeting. The subthalamic nucleus (STN) is the usual target involved when applying deep brain stimulation for Parkinson’s disease (PD). Unfortunately, STN is usually not clearly visible in common medical imaging modalities, which justifies the use of atlas-based segmentation techniques to infer the STN location. Materials and methods Eight bilaterally implanted PD patients were included in this study. A three-dimensional T1-weighted sequence and inversion recovery T2-weighted coronal slices were acquired pre-operatively. We propose a methodology for the construction of a ground truth of the STN location and a scheme that allows both, to perform a comparison between different non-rigid registration algorithms and to evaluate their usability to locate the STN automatically. Results The intra-expert variability in identifying the STN location is 1.06±0.61 mm while the best non-rigid registration method gives an error of 1.80±0.62 mm. On the other hand, statistical tests show that an affine registration with only 12 degrees of freedom is not enough for this application. Conclusions Using our validation–evaluation scheme, we demonstrate that automatic STN localization is possible and accurate with non-rigid registration algorithms.

26 citations


Journal ArticleDOI
13 Jun 2006
TL;DR: Automated segmentation of maxillary sinuses from CBCT data sets is feasible using the proposed method and might be useful for visualization, pathological diagnosis, and treatment planning ofmaxillary sinus disorders.
Abstract: Purpose Segmentation of the maxillary sinuses for three-dimensional (3D) reconstruction, visualization and volumetry is sought using an automated algorithm applied to cone beam computed tomographic (CBCT) data sets Materials and Methods Cone beam computed tomography (CBCT) data sets of three subjects aged 9, 17, and 27 were used in 3D segmentation and reconstruction The maxillary sinuses were obtained by propagation from one start point in the right sinus and one start point in the left sinus to the whole regions of both sinuses The procedure was based on voxel intensity distributions and common anatomic structures, specifically each middle meatus of the nasal cavity A program was written in C++ and VTK languages to demonstrate the surface topological shapes of the maxillary sinuses Results The developed segmentation algorithm separated maxillary sinuses successfully permitting accurate comparisons It was robust and efficient 3D morphological features of the maxillary sinuses were observed from three human subjects Conclusions Automated segmentation of maxillary sinuses from CBCT data sets is feasible using the proposed method This tool might be useful for visualization, pathological diagnosis, and treatment planning of maxillary sinus disorders

Journal ArticleDOI
15 Nov 2006
TL;DR: 3-T fMRI was accurate for locating neurologic function before tumor resection near eloquent cortex and returned to normal in all patients within 1 month.
Abstract: Objective We sought to determine the safety and efficacy of using 3-tesla (T) functional magnetic resonance imaging (fMRI) to guide brain tumor resection.

Journal ArticleDOI
16 Nov 2006
TL;DR: This paper proposes and validates a steering model to be capable of predicting the system dynamics during needle-tissue contact procedure and shows that the model can track the needle tip trajectory with satisfactory accuracy even in the presence of large disturbances and noises.
Abstract: Purpose: Needle steering can improve targeting accuracy and guide the tip to areas that are currently not accessible. This paper proposes and validates a steering model to be capable of predicting the system dynamics during needle-tissue contact procedure.

Journal ArticleDOI
17 Feb 2006
TL;DR: A fast multistage hybrid algorithm for 3D segmentation of medical images that first employs a morphological recursive erosion operation to reduce the connectivity between the object to be segmented and its neighborhood, then the fast marching method is used to greatly accelerate the initial propagation of a surface front.
Abstract: In this paper, we propose a fast multistage hybrid algorithm for 3D segmentation of medical images. We first employ a morphological recursive erosion operation to reduce the connectivity between the object to be segmented and its neighborhood; then the fast marching method is used to greatly accelerate the initial propagation of a surface front from the user defined seed structure to a surface close to the desired boundary; a morphological reconstruction method then operates on this surface to achieve an initial segmentation result; and finally morphological recursive dilation is employed to recover any structure lost in the first stage of the algorithm. This approach is tested on 60 CT or MRI images of the brain, heart and urinary system, to demonstrate the robustness of this technique across a variety of imaging modalities and organ systems. The algorithm is also validated against datasets for which “truth” is known. These measurements revealed that the algorithm achieved a mean “similarity index” of 0.966 across the three organ systems. The execution time for this algorithm, when run on a 550 MHz Dual PIII-based PC runningWindows NT, and extracting the cortex from brain MRIs, the cardiac surface from dynamic CT, and the kidneys from 3D CT, was 38, 46 and 23 s, respectively.

Journal ArticleDOI
14 Nov 2006
TL;DR: High precision and easy handling as well as a low complication rate and the high contentedness of the patients make the individual titanium skull implants valuable for cranioplasty, especially in complicated applications with very large defects, multiple previous operations and additional irradiations.
Abstract: Objective The TICC (Tomography, Image processing, CAD, CAM) processing chain developed at the Ruhr-University Bochum in Germany has already been established since several years for the reconstruction of large pre-existing posttraumatic skull defects with individual prefabricated implants made of pure titanium. So far, more than 500 titanium implants have been inserted with great success at more than 60 clinical centres worldwide. The aim of our study was to evaluate all implants inserted between 1994 and 2000.

Journal ArticleDOI
28 Oct 2006
TL;DR: The proposed graphical environment was able to fit the elements of the ideal dentition curve to patient computed tomography under predefined centric occlusion and rotation and scaling transformations of teeth were possible in the reformatted volumetric views about any of the axes of the teeth’s own reference space.
Abstract: Purpose To develop an easy-to-use, dentition planning method which is based on three-dimensional (3D) computer planning technology to replace conventional plaster-cast occlusion planning techniques.


Journal ArticleDOI
11 Oct 2006
TL;DR: Maracas software generates accurate 3D centerlines of vascular segments with minimum user intervention, and semi-automatic quantification of CA stenosis is also accurate, except in very severe stenoses that cannot be segmented.
Abstract: Object: To assess the accuracy and reproducibility of semi-automatic vessel axis extraction and stenosis quantification in 3D contrast-enhanced Magnetic Resonance Angiography (CE-MRA) of the carotid arteries (CA). Materials and methods: A total of 25 MRA datasets was used: 5 phantoms with known stenoses, and 20 patients (40 CAs) drawn from a multicenter trial database. Maracas software extracted vessel centerlines and quantified the stenoses, based on boundary detection in planes perpendicular to the centerline. Centerline accuracy was visually scored. Semi-automatic measurements were compared with: (1) theoretical phantom morphometric values, and (2) stenosis degrees evaluated by two independent radiologists. Results: Exploitable centerlines were obtained in 97% of CA and in all phantoms. In phantoms, the software achieved a better agreement with theoretic stenosis degrees (weighted kappa κ w = 0.91) than the radiologists (κ w = 0.69). In patients, agreement between software and radiologists varied from κ w =0.67 to 0.90. In both, Maracas was substantially more reproducible than the readers. Mean operating time was within 1 min/ CA. Conclusion: Maracas software generates accurate 3D centerlines of vascular segments with minimum user intervention. Semi-automatic quantification of CA stenosis is also accurate, except in very severe stenoses that cannot be segmented. It substantially reduces the inter-observer variability.

Journal ArticleDOI
21 Feb 2006
TL;DR: Streamline representation of major fiber tract systems along with high-resolution anatomical data provides a reliable orientation for the neurosurgeon and integration of fiber tract data into a neuronavigation setup allows removal of tumors adjacent to eloquent brain areas with low morbidity.
Abstract: Streamline representation of major fiber tract systems along with high-resolution anatomical data provides a reliable orientation for the neurosurgeon. For intraoperative visualization of these data either on navigation screens near the surgical field or directly in the surgical field applying heads-up displays of operating microscopes, wrapping of all streamlines of interest to render an individual object representing the whole fiber bundle is the most suitable representation. Integration of fiber tract data into a neuronavigation setup allows removal of tumors adjacent to eloquent brain areas with low morbidity.

Journal ArticleDOI
23 Nov 2006
TL;DR: Semiautomated mastoid segmentation provides a feasible means for preoperative planning, based on postprocessing analysis and visualization of CT datasets, according to a semiautomatic pipeline for segmentation of the mastoid.
Abstract: Objective Preoperative planning of complete mastoidectomy based on postprocessing, interactive processing, and visualization of CT datasets was implemented and tested. The synthesized images are used to define an access path for the operative treatment of diseases at the lateral skull base. Alternatively, the mastoid may be resected to allocate space for implants such as hearing aids.

Journal ArticleDOI
11 Oct 2006
TL;DR: A model-based method for segmenting the muscles of mastication from MR images was developed and tested and the results show good agreement between manual and automatic segmentations.
Abstract: Objective The muscles of mastication play a major role in the orodigestive system as the principal motive force for the mandible. An algorithm for segmenting these muscles from magnetic resonance (MR) images was developed and tested. Materials and methods Anatomical information about the muscles of mastication in MR images is used to obtain the spatial relationships relating the muscle region of interest (ROI) and head ROI. A model-based technique that involves the spatial relationships between head and muscle ROIs as well as muscle templates is developed. In the segmentation stage, the muscle ROI is derived from the model. Within the muscle ROI, anisotropic diffusion is applied to smooth the texture, followed by thresholding to exclude bone and fat. The muscle template and morphological operators are employed to obtain an initial estimate of the muscle boundary, which then serves as the input contour to the gradient vector flow snake that iterates to the final segmentation. Results The method was applied to segmentation of the masseter, lateral pterygoid and medial pterygoid in 75 images. The overlap indices (κ) achieved are 91.4, 92.1 and 91.2%, respectively. Conclusion A model-based method for segmenting the muscles of mastication from MR images was developed and tested. The results show good agreement between manual and automatic segmentations.

Journal ArticleDOI
07 Nov 2006
TL;DR: Initial results demonstrate the approach is feasible and may be advantageous over alternative segmentation methods, and extensions of the model, which also incorporate prior knowledge about the spatial distribution of grey values, are currently under development.
Abstract: Purpose Improved segmentation of soft objects was sought using a new method that combines level set segmentation with statistical deformation models, using prior knowledge of the shape of an object as well as information derived from the input image.

Journal ArticleDOI
13 Jul 2006
TL;DR: Post-processing enhancement of PSP cephalograms is perceived to improve clarity of selected anatomic landmarks in PSP cephalometric landmark evaluation and image bit depth.
Abstract: Objective The conspicuity of cephalometric landmarks may be improved by pseudo color and emboss enhancements of 8 and 16 bit photostimulable phosphor (PSP) cephalograms reviewed by orthodontists.


Journal ArticleDOI
18 Jul 2006
TL;DR: The HSP and BSP were calculated through maximizing local symmetry within the head or cerebrum followed by outlier removal and are preferred for a high accuracy Talairach transformation and localization of the anterior and posterior commissures.
Abstract: Most existing image-based approaches neglect the difference in radiological symmetry between the human brain and head. Thus, it is important to analyze and quantify the spatial relationship between the brain symmetry plane (BSP) and the head symmetry plane (HSP) on radiological images. The HSP and BSP were calculated through maximizing local symmetry within the head or cerebrum followed by outlier removal. The HSPs and BSPs for 145 diversified MRI datasets (80 normal, 23 pathological, and 42 synthesized) were extracted and compared. The average angular and distance deviations between the HSP and BSP were 0.49° and 1.65 mm, respectively. These deviations are dependent upon ethnicity and gender, being: (1) (0.56°, 1.85 mm) and (0.42°, 0.91 mm) for Caucasians and Asians, respectively; and (2) (0.33°, 1.17 mm) and (0.51°, 1.58 mm) for males and females, respectively. The HSP is generally different from the BSP on MR images. Statistically, they can be used interchangeably if accuracy of (0.49°, 1.65 mm) is acceptable. The BSP is preferred for a high accuracy Talairach transformation and localization of the anterior and posterior commissures. Either BSP or HSP can be used for medium accuracy Talairach transform. The HSP is preferred for detecting intracranial pathology.



Proceedings Article
01 Jan 2006
TL;DR: By registering a binary mask of the third and lateral ventricles of the patient with its corresponding binary masks of the atlas the authors obtain a fast, automatic and accurate pre-operative targeting comparable to the expert’s variability.
Abstract: Objects Deep brain stimulation (DBS) has turned out to be the surgical technique of choice for the treatment of movement disorders, e.g. Parkinson’s disease (PD), the usual target being the subthalamic nucleus (STN). The targeting of such a small structure is crucial for the outcome of the surgery. Unfortunately the STN is in general not easily distinguishable in common medical images. Material and Methods Eight bilaterally implanted PD patients were considered (16 STNs). A three-dimensional MR T1-weighted sequence and inversion recovery T2-weighted coronal slices were acquired pre-operatively. We study the influence on the STN location of several surrounding structures through a proposed methodology for the construction of a ground truth and an original validation scheme that allows evaluating performances of different targeting methods. Results The inter-expert variability in identifying the STN location is 1.61 ± 0.29 mm and 1.40 ± 0.38 mm for expert 1 and 2 respectively while the best choice of features using segmentation-based registration gives an error of 1.55 ± 0.73 mm. Conclusions By registering a binary mask of the third and lateral ventricles of the patient with its corresponding binary mask of the atlas we obtain a fast, automatic and accurate pre-operative targeting comparable to the expert’s variability.

Journal ArticleDOI
24 Feb 2006
TL;DR: A preliminary result using the medial based approach promises to discriminate at least globally between healthy and diseased hearts.
Abstract: Our work aims at investigating the suitability of the medial representation method to model and analyze shape and shape differences between healthy and diseased hearts. For this experimental study, we use MRI short axis scans of 11 healthy volunteers (age: 50±10) and 5 patients (age: 57±11) with dilativ cardiomyopathy. Controlled semi- automated segmentation provides labels, which are used for the modeling process. To evaluate the model to image accuracy the similarity index (SI), the mean Euclidean distance (ED), and the Hausdorff distance (HD) are calculated. A very high SI (SI > 0.9) for the ventricles is achieved. The mean ED is less than two times the voxel size (1.56 mm) and the HD values for both chambers are in the range of 4.8±3 mm. Applying extended principal component analysis (PCA) on all 16 subjects reveals the distribution of the individual shapes, where the first two PC cover more than 40%, and the first ten PC cover 95% of the shape space. The components show meaningful modes of variation, whereas the healthy and diseased hearts are clustered in the first two components. This preliminary result using the medial based approach promises to discriminate at least globally between healthy and diseased hearts.

Journal ArticleDOI
21 Jul 2006
TL;DR: An interactive virtual model of a proposed third molar surgical site including the third molars and the inferior dental canal was developed using 3-D volumetric imaging systems.
Abstract: Precise 3-dimensional localization of impacted mandibular third molars relative to the inferior dental canal (IDC) is critical to clinical management and surgical outcomes. Recently introduced dental 3-D volumetric imaging systems coupled with semi-automatic modeling techniques allows 3-D visualization of the IDC and the third-molar. Six impacted third molar sites were imaged with various 3-D volumetric imaging systems (NewTom 9000, Morita Accuitomo and Hitachi Mercuray). The spatial relationship of six impacted third-molars were visualized using imaging data obtained from these units. An interactive virtual model of a proposed third molar surgical site including the third molar and the inferior dental canal was developed.