Physicians need to see the inside surface of heart chambers in minimally invasive medical procedures (e.g. trans-apical aortic valve replacement, catheter-based ablation) but the lack of transparency of blood in the visible region prevents the use of conventional visible-light imaging. Furthermore, while anatomical structures are three dimensional, current infrared (IR) imaging systems are able to acquire only low-resolution 2D images that lack depth information. This fundamental restriction greatly limits clinicians’ ability to perceive and understand the complex intracardiac anatomy.
We will develop the design principles for real-time IR medical imaging systems that are able to penetrate the blood, visualise the region in front of the device, and reconstruct the 3D surface profile of tissue in the field of view. We will construct a series of prototype systems to validate these principles, and to demonstrate their benefits in guiding medical procedures.
This project will show that high-quality 3D information can be extracted from noisy 2D IR images taken through blood, and that this information can be conveyed to surgeons to enhance their ability to perform cardiac repair procedures. To do this, we will first exploit new IR imaging technology to improve 2D imaging quality. Second, we will extend optical metrology techniques to scattering media (e.g. blood) to pull out 3D information without being affected by artefacts. Third, we will develop new display techniques to convey the resulting 3D information to the physician for real-time use in procedure guidance.
With the availability of this device, the visualisation of soft tissue through blood that has held back the development of beating heart procedures can be solved. In the long-term, patients treated by an endoscope procedures, as compared to those who have undergone invasive open-heart surgeries, will be expected to have a shorter rehabilitation period, leading to savings in health-care resources and expenses.