Manual segmentation of computed tomography (CT) datasets was performed for robot-assisted endoscope movement during functional endoscopic sinus surgery (FESS). Segmented 3D models are needed for the robots' workspace definition. A total of 50 preselected CT datasets were each segmented in 150-200 coronal slices with 24 landmarks being set. Three different colors for segmentation represent diverse risk areas. Extension and volumetric measurements were performed. Three-dimensional reconstruction was generated after segmentation. Manual segmentation took 8-10 h for each CT dataset. The mean volumes were: right maxillary sinus 17.4 cm(3), left side 17.9 cm(3), right frontal sinus 4.2 cm(3), left side 4.0 cm(3), total frontal sinuses 7.9 cm(3), sphenoid sinus right side 5.3 cm(3), left side 5.5 cm(3), total sphenoid sinus volume 11.2 cm(3). Our manually segmented 3D-models present the patient's individual anatomy with a special focus on structures in danger according to the diverse colored risk areas. For safe robot assistance, the high-accuracy models represent an average of the population for anatomical variations, extension and volumetric measurements. They can be used as a database for automatic model-based segmentation. None of the segmentation methods so far described provide risk segmentation. The robot's maximum distance to the segmented border can be adjusted according to the differently colored areas.

The preferred treatment of femoral (thigh bone) shaft fractures nowadays is the minimally invasive technique of intramedullary nailing.However, in addition to its advantages, this technique also has a number of disadvantages, such as the frequent occurrence of malaligned fracture reductions and high X-ray exposure, especially to the operating team. The aim of our research is to overcome these shortcomings by utilizing modern techniques such as three-dimensional (3D) imaging, navigation, and robotics. In this paper we present the current state of our interdisciplinary research project. We first introduce a telemanipulated fracture reduction procedure, which is based on 3D imaging data. This set-up is improved one step further towards an automated fracture reduction procedure. Finally, two drilling tasks, namely the opening of the medullar cavity and the distal locking of the intramedullary nail, are presented, which are supported by automated X-ray-based image analysis and robot-assisted drill guidance. We show that high reduction accuracies can be achieved with our robotic system. Furthermore, the robot-assisted drill guidance achieves superior results with respect to increased precision and decreased Xray exposure compared with the conventional procedure. We conclude that this surgical procedure benefits conspicuously from the support of robotic assistance systems and that further research and development in this field is worthwhile.

Reduction is a crucial step in fracture treatment. We determined intraoperative peak forces and torques during fracture reduction in seven patients with eight fractures of the femoral shaft. All fractures were temporarily stabilized by external fixation. Force and torque measurements were performed during the subsequent intramedullary nailing procedure. A three-dimensional load cell was attached to the distal femur fragment using two Schanz screws. All forces and torques were registered on-line during the reduction process. The maximum resulting force was 411 N, the maximum resulting torque 74 N Á m. The highest force was observed along the shaft axis with 396 N for distraction. The maximum torque value was measured around the frontal axis, being 74 N Á m for antecurvature. These results may assist the development of new reduction techniques and devices.

High tibial osteotomy (HTO) is a commonly used surgical technique for treating moderate osteoarthritis (OA) of the medial compartment of the knee by shifting the center of force towards the lateral compartment. Previous studies have documented the effects of HTO on the biomechanics of the knee. However, the effects of the procedure on the contact pressures within the ankle joint have not been as well described. Seven cadavers underwent an HTO procedure with sequential 5˚valgus realignment of the leg up to 15˚of correction. An axial force of up to 550 N was applied and the intraarticular pressure was recorded. Minor valgus realignment of the proximal tibia does not significantly alter the biomechanics of the ankle. However, moderate-to-large changes in proximal tibial alignment result in significantly decreased tibiotalar contact surface area and in changes in intraarticular ankle pressures. These findings are clinically relevant, as they provide a biomechanical rationale for the diagnosis and treatment of ankle symptoms in the setting of lower limb malalignment or after alignment correction procedures. ß

Reduction in femoral shaft fractures can be difficult to achieve with minimally invasive techniques. Malalignment and high intra-operative radiation exposure can result. The hypothesis was that robot-assisted fracture reduction could improve the quality of reduction while reducing the amount of radiation exposure. A robot system was developed that allows fracture manipulation with a joystick as input device. The system provides the surgeon with haptic and metric feedback. Fifteen synthetic femurs were broken and reduced by simulated open (group A) and closed techniques (group B). These techniques were compared with the robot-assisted reduction with (group C) and without (group D) haptic and metric information. An image intensifier was simulated with two orthogonal cameras. All reduction techniques showed minor malalignment. In group C, the alignment was: procurvatum/recurvatum 0.6 degrees (0-2.0 degrees); varus/valgus 0.8 degrees (0-3.0 degrees); and axial rotation 0.8 degrees (0-3.1 degrees). A significant difference was seen between the groups (two-way ANOVA, p < 0.001). Axial rotation was significantly lower in group C than in group B (1.9 degrees; p < 0.001). The residual varus and valgus deviation was higher in group C compared with group A (0.4 degrees, p = 0.03). The median number of simulated radiographs was significantly less in group C (35) compared with group D (72; p < 0.001) and group B (49; p = 0.01). Robot-assisted fracture reduction of the femur provides high precision in alignment while reducing the amount of intraoperative imaging. Further research in this field is worthwhile.