Every year, thousands of patients in UK hospitals have feeding tubes accidentally placed into their lungs instead of their stomachs, causing severe injury or death. The current gold standard for checking tube position is an X-ray, which exposes patients to radiation, delays feeding, and requires moving equipment or patients. Researchers at this institution have built a bedside optical device that uses pulses of light and ultra-fast photon detectors—similar to technology being developed for self-driving cars—to track the tube’s tip in real time without any X-rays. The system exploits the fact that a tiny fraction of light from a source inside the body travels in a straight line to the surface, arriving faster than scattered light. By timing these “line-of-sight” photons, the device can pinpoint the tube’s location. The team has already demonstrated the principle in the lab and is now building a clinical prototype. If successful, the device could eliminate the need for X-ray confirmation, speed up feeding, reduce harm, and cut costs—a practical improvement to a routine but risky hospital procedure.
View original technical description
Placement of enteral feeding tubes (nasogastric tubes/NGTs) is a standard medical procedure, yet misplacement is a serious issue (e.g. food entering the lungs with consequent death and disability from pulmonary complications). Current practice relies heavily on ionising X-ray radiation to localise medical devices during procedures causing significant disruption given the need to protect staff as well as waiting for the availability of, or transport to, X-ray equipment. There are therefore significant clinical drivers to develop alternative placement confirmation methods. We have developed an optical technology (no X-rays) to augment and guide NGT placement using a compact bed-side system. We have demonstrated a ground-breaking approach to track and show correct NGT stomach placement (or misplacement in the lung) using single photon imaging. Our approach exploits the fact that if a point source of light is placed inside the body, a tiny fraction of the light will emerge from the body effectively in a straight line. Crucially, these line-of-sight photons (particles of light) hold precise information about the spatial location of the point source inside the tissue. We can utilise this information as the line-of-sight photons exit the body faster than the more diffuse photons that have been scattered along a longer path to exit the body - we are able to use a technique known as time-correlated single-photon counting (TCSPC) to detect and specifically observe and use the fast photons. In contrast to "normal" cameras, which do not record the arrival time of the photons, TCSPC-based imaging relies on using a light source that produces short pulses at precisely known times, together with a single-photon sensitive detector that records arrival times. Therefore, TCSPC imaging allows us to design an imaging system that can selectively detect and image the location of the line-of-sight photons before the diffuse scattered photons start to emerge, allowing us to precisely locate the source. Although we have initially demonstrated the potential of this technique to locate NGTs, we are increasing the detection speed to provide real time tracking using the newest detectors - similar to those in development for self-driving cars. We have also combined our technique with new optical fibres optics placed in the NGTs, which have light sources spaced along their length allowing observation of the full NGT path during placement or reconfirmation. We now intend to finalise the clinical prototype device and complete preclinical validation, including determining diagnostic accuracy. We will then move to evaluating feasibility and safety of the devices' ability to guide NGT placement in patients. Our goal is to reduce significant adverse events associated with NGT placement (and reconfirmation), make placement faster, enhance clinical workflows (removing need for x-rays), improve patient outcomes through initiating feeding/medication earlier and reduce overall costs.
John Norrie (Co-Investigator)Kev Dhaliwal (Co-Investigator)Michael Tanner (Principal Investigator)Robert Thomson (Co-Investigator)Thomas Henry John Craven (Co-Investigator)
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know