The Phantom based Computer assisted orthopaedic surgical system (CAOSS) has been developed collaboratively by the University of Hull and the Hull Royal Infirmary, to assist in operations like dynamic hip screw fixation. Here we present summary of our system.

CAOSS comprises a personal computer based computer system, a frame grabber with video feed from a C-arm image intensifier, an optical tracking system and a radiolucent registration phantom which consists of an H arrangement of 21 metal balls. The phantom is held in position by the optically tracked end-effector. Knowing the optical position of the phantom, a registration algorithm calculates the position of C-arm in coordinate space of the optical tracking system.

Computer based planning uses an anteroposterior (AP) and lateral image of the fracture. Marks are placed on the 2D projections of femoral shaft, neck and head on the computer screen, which are then used to create 3D surgical plan. The computer then plans a trajectory for the guide wire of DHS. The depth of the drill hole is also calculated. The trajectory is then shown on both AP and lateral images on the screen.

CAOSS meets all the requisite of electrical and electromagnetic radiation standards for medical equipment. There has been extensive validation using software simulation, performance evaluation of system components, extensive laboratory trials on plastic bones. The positional accuracy was shown to be within 0.7mm and angular accuracy to be within 0.2°. The system was also validated using Coordinate Measurement Machine.

Our system has the unique feature of the registration phantom which provides accurate registration of the fluoroscopic image.

Aim: The Computer Assisted Orthopaedic Surgical System [CAOSS] is designed to assist the surgeon in performing the task of accurate placement of the distal locking screws via a trajectory that is planned by one AP and Lateral image from the conventional C-Arm. Methodology: Two near orthogonal x-ray images containing the distal femur with the registration phantom and including the distal end of the nail with the two locking holes are obtained using a standard C Arm and then processed after distortion correction. The phantom is supported by an end effector, which is continuously tracked in 3D space by an overhead camera. Features of interest are extracted and the image registered in 3D space through the evaluation of the phantomñs projection. A computer-based model of the anatomical region is developed and the position of the screws planned. Even if the distal locking hole image is not a true circle, the software is robust enough to detect the difference in curvature of the upper and lower part of the ellipse and thus calculate the necessary angle at the time of insertion. Once the trajectory is accepted, the surgeon implements the plan by moving a passive manipulator arm, while receiving visual positional cues from the computer in the form of a targeting screen. When the targeting is complete; the arm is locked in position and the trajectory implemented. Two individuals used the device for distal locking of Richards intra medullary femoral nail in several saw bone models. Results and Conclusions: Successful locking was accomplished in all cases by using the trajectory planned using one AP and Lateral image. This was the case even when the image was not a true lateral of the locking hole. The results of this study using this new versatile system, including the number of x-rays required, duration of x-ray exposure and time for distal targeting and locking are presented.

Aims: Dynamic hip screw for intertrochanteric fractures is one of the most common procedures performed by orthopaedic surgeons. The prerequisite for proper placement of the implant is accurate insertion of the guide wire. The Computer Assisted Orthopaedic Surgical System [CAOSS] is designed to assist the surgeon by planning the trajectory based on one intra-operative AP and Lateral image from a C-Arm. Methodology: After closed reduction on the fracture table, two near orthogonal x-ray images containing the proximal femur with the registration phantom are obtained using a standard C-Arm and then processed after distortion correction. The phantom is supported by an end effector, which is continuously tracked in 3D space. Features of interest are extracted and the image registered in space through the evaluation of the phantom’s projection in the x-ray image. The versatility of the CAOSS is increased by the provision allowing the adjustment of the planned trajectory to the surgeon’s satisfaction. Once the trajectory is accepted, the surgeon implements the plan by moving a passive manipulator arm, while receiving visual positional cues from the computer in the form of a targeting screen. When the targeting is complete; the arm is locked in position and the trajectory implemented. Results: We present the results of the pilot clinical study involving 10 patients using this device. The results obtained were compared with an equal number of patients randomly selected from the complete neck of femur database, who had undergone a conventional DHS placement, during the last one-year. Accuracy of placement of the implant was assessed by an independent observer and by a previously validated computer program that assesses the accuracy from scanned post operative X-rays. The average targeting time was 6 minutes and overall there was no significant difference between the two groups.