| Part I A Comparative Study of Digital and Traditional Skull Base surgeryBackgroundThe structures of the skull base are dense and complex; the accurate intra-operative orientation is, therefore, of significant importance in surgery involving the skull base and requires clinical experience and expert skills as well as extensive knowledge of the related anatomy. Skull base or cranial base surgery became a multidisciplinary subspecialty in the 1980s.Traditionally, the orientation of skull base surgery depended mainly on data from the anatomical parameters of some important skull base structures. These data were obtained by statistically analysing measurements made with scientific callipers from groups of anatomy specimens, and so are of limited use in determining the individual anatomical features of a particular patient. In addition, because of the sampling error and ununiformity of operator and research condition, there were significant differences about every report.The digital imaging technique can provide individual anatomical information, but its accuracy was affected by the quality of image, especially MRI. Because of unstability of magnetic field and the shadow of air-water and air-fat, the image was apt to deform and influence the accuracy. CT takes information via X-ray and has high accuracy so that the image will not deform, but the thick slice of CT may also affect the accuracy of image. On account of the advanced image technique, the difference between the anatomical information by CT image and the actual anatomical information was investigated.The neuronavigation system gathers the individual information of neuro-imaging, individualizes patients' anatomical structure and pathological lesions, labels them and performs 3D reconstruction, so that offer more accurate anatomical features of patients. Moreover, according to the principle of virtual image, the neuronavigation system can provide the surgeon of the information of unvisible and inner intracranial vital structures, locate the vital structures accurately in operation under real-time guidance, and make up the deficiency of the surgeon's experience and local anatomy utmostly. The virtual image function of the neuronavigation system can even feed-through with results of operating microscope system, form a virtual field called "perspective mode", and help surgeons know the deep vital structures. This neuronavigation system can reduce the operation difficulty of surgery and increase safety of surgeryIntra-operative brain shift is the most serious problem that can compromise the accuracy of orientation and depreciate the value of an image-guided surgery system.As fixed bony structures in skull base, which do not shift at all during the operation, are used as landmarks. Thus, the image-guided surgery system may be more suitable for skull base surgery than other neurosurgical operations.ObjectiveTo compare the respective accuracy of individual based CT and Image-Guided Surgery system with group-based conventional scientific calipers for the methodological study of skull base surgery and quantitatively evaluate the practicability of the replacement of the stereotype of direct anatomical measurements for the conventional studies of the skull base surgery with digitalized techniques.Methods1Manufacture of specimen the brain and cranium were removed from 25 formalin-fixed cadaver skulls. Bony structures including the lateral semicircular canal (LSC), anterior semicircular canal (ASC), posterior semicircular canal (PSC), foramen ovale (FOV), foramen spinosum (FSP), hypoglossal canal (HYC) and internal acoustic pore (IAP) were exposed for measurement.2 Acquisition of image information the skulls were scanned using a 16-slice CT scanner and the data was input into a BrainLab image-guided surgery system (BrainLab).3 Measurement of image-guide surgery Image information was imported to image-guide surgery and reconstruction. The target location error and the distance between selective structures were measured after completion of Z-touch laser registration.4 Coefficient of variation Two independent surgeons selected one cadaver skull respectively and measured the distance between two sides FOV with three measurement modes, the measurement of each mode was performed by each surgeon in 30 times, to compare coefficient of variation of three measurement modes.Results1 The measurement results of anatomical, image and image-guide surgery LSC—FOV:37.74±2.42mm,37.46±2.54mm,37.95±3.50mm; ASC-FSP:28.00±2.24mm,27.81±2.13mm,27.68±2.85mm; PSC-HYC:34.28±3.50mm,34.50±3.39mm,34.75±3.46mm; ASC-IAP:13.87±1.61mm,13.75±1.37mm,14.05±1.82mm. statistical analysis shows that all P-value exceed 0.05, there are no statistical difference in three measurements.2 Mean target localization error is 2.09±0.65mm for Z-touch laser registration. 3 Coefficient of variation Anatomical measurement and image measurement and image-guide surgery measurement experimenter 1,0.40%,0.39%,0.40%; experimenter 2,0.69%,0.47%,0.68%.Conclusions1 Compared to skull base surgery based on regional anatomy,digital skull base surgery based on CT and computer can supply more accurate individual anatomical information.2 Compared to manual operation, automated and digital operation more flexible and less variable. Digital methods are more flexible and less variable。 3 Digital skull base surgery will replace traditional skull base surgery and be used extensively. PartⅡComparative Study of Z-touch and Facial fiducial Marker RegistrationBackgroundImage-guided surgery has been applied increasingly in intracranial surgery. It pairs the patient's image and his anatomical structures and establishes a correlation beween operative coordinate system and image coordinate system, and establishment of reforming relation between two coordinate systems. After the registration were completed, the technique allows the location of surgical instruments to be monitored using ultrasound or infrared light and assists the surgeon in selecting the appropriate operative approach, locating the lesion exactly, and avoiding injury of normal tissue. As a result, it lessens the risk of surgery.The registration of image space to physical space is vital during image-guided surgery. Markers are now commonly used for registration and give higher precision but at an increased cost. Implanted marker registration required implantion of titanium screws befor image examination, it crease the distress of patient. Facial fiducial marker registration requires the attachment of radiolucent or MR-compatible markers to the skin over the entire area of interest prior to imaging, however, movement or swelling of the skin and the actual location of marker can all affect the accuracy. Occasionally, the marker is lost and must be replaced, which prolongs the surgery. Although anatomic landmark registration is convenient to perform, it is difficult to locate anatomic landmarks accurately in the image; hence it has a low level of precision.The BrainLAB navigation system utilizes a wireless non-contact laser registration instrument, named the Z-touch, which emits detectable infrared light. The frontal region, per-orbital cavity, and nasal root of the patient are scanned with the Z-touch laser pointer, the infrared light that is reflected is received, recognized, and transformed into digital information by digital cameras. These data are then analyzed by computer to establish the relationship between the image space and physical space coordinates. As a non-contact registration, Z-touch can perform re-registraton in intraoperative time without any especial instruments. Our study is designed to evaluate the accuracy and clinical value of Z-touch registration. However, the research on Z-touch laser registration focused on registration accuracy and target accuracy, and few people pay more attention on the research and comprehensive evalution.ObjectTo evaluate comprehensively the Z-touch laser registration by comparing the requiring time, registration accuracy, target localization accuracy between Z-touch laser registration and facial adhesion marker registration, meanwhile, to preliminarily explore the factors effecting target localization accuracy.Methods1 Acquisition of image information Sixteen cadaver heads were selected as research objects. Six registration markers were attached asymmetrically and dispersedly to the external surface of the cadaver head. The foramen spinosum, internal acoustic pore, and six titanium screws implanted into the skull base were used as targets to measure the actual deviation of localization after registration. The skull was scanned using a high-resolution 16-slice CT-scanner. Image information was inputted into compact disc.2 Preoperation management of image information Image information was imported to preoperation planning workstation, Reconstructed three-dimensional images of corpse head and locate registration markers:confirm markers in a certain sequence on 3-D images, click the center of markers by the mouse in order as precise as possible to guarantee the higher accuracy at the time of registration, then conduct the data into the ZIP disk.3 Skin fiducial marker registration Image information was inputted into neuronavigation. The cadaver head was fixed on a Mayfield headrest, together with a reference star, skin fiducial marker registration was completed and required time for registration and registration accuracy and target location error.4 Z-touch laser registration The hand-held Z-touch laser pointer was used to acquire up to 100 surface points by selectively scanning characteristic facial features. These acquired surface points were matched accurately to the image of subject by the software to complete registration. The required time for registration and registration accuracy and target location error was recorded.5 Target localization error The target localization error is the best index for assessing the quality of registration and the true application accuracy. To determine this value, the tip of the pointer was placed on the landmark or target and the application accuracy was calculated by comparing the position of the tip as shown in the CT image with the position in reality.6 Analysis of results All data were analyzed with SPSS 13.0. The average values of time consuming, registration precision, and target localization error were analyzed using a non-parametric test. Correlation of target localization error with registration precision was assessed by Pearson's correlation analysis. The target localization error of different target was analyzed in Kruskal-Wallis test, a P-value<0.05 was considered to be statistically significant.Results1 Required time of Z-touch laser is 121.69±16.16s; Skin fiducial marker is 368.75±66.76s. There are not difference between two registration modes (P=0<0.05).2 Registration accuracy of Z-touch is 1.16±0.24 mm; Registration accuracy of skin fiducial marker is 0.43±0.13 mm. Statistical difference was found between two registration modes (p<0.005).3 Target localization error of Z-touch is 2.19±0.24 mm; Target localization error of skin fiducail marker is 1.27±0.15mm.The difference of target localization error between two registrations is statistical significant (P=0<0.05).4 The target localization error of target 1,foramen spinosum(left),internal acoustic pore (left), target 2, target 3, foramen spinosum (right),internal acoustic pore (right), target 5, target 6, for Z-touch laser, are 1.01±0.15mm,1.81±0.31mm,2.17±0.18mm,2.54±0.25mm,2.65±0.37mm,3.17±0.46mm,1.88±0.29mm,2.21±0.23mm,2.61±0.19mm,2.93±0.39mm; for skin fiducial marker registration is0.92±0.16mm,1.06±0.13mm,1.25±0.18mm,1.48±0.29mm,1.26±0.12mm,1.73±0.35mm,1.10±0.16mm,1.34±0.16mm,1.36±0.36mm,1.21±0.15mm.A strong correlation was found between registration accuracy and target localization error in two registration modes.Conclusion1 To compare with marker registration, Z-touch laser scanning has a lower level of precision but higher acceptability and less time requiring and medical cost.2 Z-touch laser scanning registraion can replace marker registration and was used extensively.
|
PDF Full Text Request