SimVis Medical - Image Guided Surgery

The aim of this research is to pioneer image analysis, registration and calibration techniques and their innovative application to surgery and diagnosis so as to provide benefit to both healthcare providers and patients.

A clinical Computer Assisted Orthopaedic Surgical System

This project concerns the development of a prototype clinical system for performing fluoroscopic image guided orthopaedic surgical procedures namely: distal locking of intramedullary nails; dynamic hip screw guide wire placement; and cannulated hip screw placement.  All of these procedures require the accurate placement of a guide wire within the patient's anatomy.   The system is based around an x-ray vision system and a passive manipulator arm.  It employs a novel phantom-based image/anatomy registration technique that is far superior to those used by recent virtual fluoroscopy products (which use techniques we pioneered in 1993-96).  The project also involves the evaluation of the system through a series of clinical trials, and involves the use of a new framework approach for quantitative simulation for the accuracy performance of image guided surgery systems.

The system has been approved by the Medical Devices Agency and has both UK and EU patents filed.  The project also received the 1995 BCS UK IT award.

Details: CAOSS for dynamic or compression hip screw placement

Computer assistance for osteotomy and joint replacement operations

This project concerns the development of a prototype image guided robotic system for performing osteotomy and joint replacement operations such as the knee (i.e. bone cutting).  This is a collaborative multidisciplinary project with Dundee and Loughborough Universities who have developed a saw and robot specifically for bone cutting.  The project addresses the cutting and realignment of bone.  Sensors attached to the saw enable measurement of forces and vibrations, enabling precise positioning by the controlling robot, throughout the cutting phase of the procedure.  The role of the Hull Group is to provide the software for 2D x-ray and 3D CT osteotomy planning and control of the robot, a new algorithm for planning the cutting of a wedge from a bone and new surface-surface registration algorithms.  A complete laboratory prototype system exists that can perform osteotomies.

Virtual Fluoroscopy

The C-arm fluoroscope is an intra-operative 2D imaging device for orthopaedic surgery but it poses a significant radiation hazard to staff and patients.  A technique known as virtual fluoroscopy (VF) is being developed that enhances the fluoroscope’s capability so that it produces virtual images of patient anatomy and surgical instruments during surgery. In VF a number of fluoroscopic images the patient are captured just prior to surgery.  The position of image space for each image is also recorded. During surgery the position of the surgeon’s instruments are tracked optically and this allows the computer to overlay a projection of the instrument onto the previously captured images to produce virtual images. By tracking the movement of bones during surgery, spatial synchronisation is maintained in the virtual images between bones and instruments.  This project uses the new technique (developed as part of the CAOSS project) for registration and calibration of the fluoroscope’s image space. This technique uses a small registration phantom placed close to the patient whilst imaging.

Image Guided Uni-compartmental Knee Replacement

This project is investigating the effectiveness of using image guided surgery techniques for positioning the femoral and tibial components of a uni-compartmental knee prosthesis.  This prosthesis replaces the joint surfaces of the medial compartment of the knee joint and is a treatment for osteoarthritis. The prosthesis comprises a tibial and a femoral component and a polyethylene bearing.  The image guidance goal is to create a mechanical axis of the prosthetic joint that is neutral. The joint line should be parallel to the floor and perpendicular to the mechanical axis of the femur and tibia. Currently, complex instrumentation greatly assists accurate component placement but the operation still requires a lot of surgical skill.  The image guidance objectives are thus to simplify the instrumentation, reduce the level of surgical skill and to improve the outcome of surgery. Image guidance is being provided by the Virtual Fluoroscopy technique.

Computer assisted post-operative assessment of implant position

This project aims to develop a tool to provide computer assisted post-operative assessment of implant position from either plate or fluoroscopic x-ray images.

Details: Computer-assisted post-operative assessment for DHS placement

Registration of Medical Images with Patients

Providing accurate registration of an image-based trajectory plan with a patient is crucial to computer-assisted surgery.  The problem is being approached from two different directions, registration by discrete data matching and registration by image matching.  A number of discrete data matching algorithms are being investigated including 3D points to 3D points, 3D surface to 3D volume and 2D projections to 3D volume.  Imaging matching requires the iterative matching of simulated x-ray images with real intra-operative x-ray images to provide the registration.

Details: 3D registration through pseudo x-ray image generation

Computer assistance for minimally invasive knee replacement

This project aims to develop a laboratory-based prototype for assisting the surgeon in the planning and insertion of an uni-compartmental knee replacement prosthesis.

Analysis of X-ray Images and Anatomical Feature Detection

Automated image analysis for computer-assisted surgery offers the potential of improved accuracy and more consistent surgery.  Analysis techniques being pursued include a) improving image quality to make subsequent automated analysis more effective, b) automatic extraction from various forms of images of anatomic and implant features for various orthopaedic procedures.  Knowledge-based approaches are being incorporated within these techniques.

Modelling of x-ray image generation

The intra-operative mobile x-ray system, known as an image intensifier, produces a real-time but distorted x-ray image. It is not suitable as a vision component of a computer assisted surgical system without extensive calibration.  This work aims to model the image generation process within the image intensifier, in order to transform the qualitative device into a very accurate quantitative vision system.  The calibration technique developed permits high precision 2.5D anatomy reconstruction from two fluoroscopic images (its accuracy outperforms any published techniques).

Surgical manipulators

Surgical manipulators provide the physical link between the computer based surgical plan and the patient's anatomy.  This project aims to develop a generic manipulator that is suited to a wide range of computer-assisted orthopaedic procedures.

Surgical planning system

Surgical planning for computer-assisted surgery can be based on either pre-operative CT/MR datasets or intra-operative x-ray images.  This project aims to develop a planning scheme based on CT/MR datasets for a range of spinal procedures.  Incorporated within the planning scheme is the provision for landmark selection (both manual and automatic) to be used for registration of the plan with the patient's anatomy.

Future Activity

The group aims to develop new techniques for image/anatomy registration, fusion of images, image analysis and take advantage of new imagining and display advances such as 3D ultrasound, digital fluoroscopy, retinal displays and develop innovative image guided systems for microscope surgery, Osteotomy operations, spinal surgery, etc.