Ballistic and Snake Photon Imaging for Locating Optical Endomicroscopy Fibres
M.G. Tanner, T.R. Choudhary, T.H. Craven, B. Mills, M. Bradley, R.K. Henderson, K. Dhaliwal, and R. R. Thomson
Heriot-Watt University / University of Edinburgh
The principle is one that every child knows: that red light can pass through the body, as seen when you put a torch behind your hand. In fact, near-infrared light, which lies just outside of the visible spectrum, passes more easily through tissue. However, all of the complicated structures found within the body (e.g. internal organs and muscle) cause the light to bounce around excessively as it passes through, and the tiny amount that does come out the other side would simply appear as a vague glow – this is not useful for working out where you left your endoscope!
Our project is enabled by the latest developments in ‘single photon detection’ technologies and novel ways to apply them to medical imaging. Photons, the particles of light, have tiny amounts of energy (a 100 Watt light bulb consumes ~300 million million million times this in a second), and exquisitely sensitive detectors are required to detect them individually. In the physics lab, numerous varieties of detector have been developed with increasing sensitivity, but these tend to be accompanied by large racks of equipment to run them.
The step-change in technology, now available via Proteus, is that thousands of single photon detectors, integrated with all the electronics required to operate them, can now be found on a single silicon chip. The detectors developed within the University of Edinburgh are akin to a digital camera chip, with the important distinction that they can observe the arrival of single photons, and the time of that arrival, on each pixel to form a camera image. This enables new imaging capabilities, specifically the observation of the minute amounts of light that can pass through the human body.
Our aim was to develop a camera that can actually see near-infrared light as it travels through the body. When applied in a clinical setting, the clinician can use this camera to locate a medical device inside a patient using a special light source in a normally-lit room. With medical device insertion and endoscopic investigations occurring as everyday procedures, knowing where the device is without an X-ray could change clinical practice and speed up procedures for large numbers of patients. If we can install it as a heads-up display within a pair of glasses, we get pretty close to acquiring X-ray vision! The camera had to be able to measure information with a timing accuracy faster than the speed of light (or its transit time) in order to detect the first photons.
Usually, light scatters or bounces off tissue rather than travelling straight through, which makes conventional through-tissue imaging practically impossible as the scattering blurs the image and causes the loss of all information about the tissue structure when passing through thick organs. Taking advantage of single photon detection solves this problem; not just for the sensitivity of observing the small number of photons passing through tissue, but also for recording the time they take to do so. Light which is highly scattered in the tissue travels a longer distance, and arrives later at the camera. A small fraction of the light scatters relatively little, and travels in a nearly direct (or ballistic) path to the camera, arriving much sooner. Operating the camera in a mode similar to a video camera, the early arrival of this direct path light can be separated from the later more scattered light, in a concept known as “ballistic imaging”.
Our prototype demonstrations have already shown that a point light source can be located through ~20 cm thick tissue under normal lighting conditions with this technique. This project intends to develop this for medical application in a number of ways. A system will be refined to enable clinicians to locate inserted medical devices at the bedside, visualising both the tip and length of the device.