Characterisation and modelling of inter-core coupling in coherent fibre bundles

A. Perperidis, H.E. Parker, A. Karam-Eldaly, Y. Altmann, K. Dhaliwal, R.R. Thomson, M.G. Tanner, and S. McLaughlin
Heriot-Watt University / University of Edinburgh


Coherent fibre bundles are optic fibres that transmit images from one end to the other and can be used to see inside the body. Our paper looks at an effect called inter-core coupling, which can be seen in all fibre bundles, and reduces overall imaging quality. We discuss a method that can quantify this inter-core coupling.


Doctors use imaging fibres to see inside the body in order to investigate disease. These imaging fibres are made up of thousands of ‘cores’ that each transmit light from one end to the other. By using thousands of these cores, all closely packed together, we can build up an image. This is similar to the idea of pixels on a screen. Sometimes, light that is transmitted down one core can ‘leak’ into its neighbours in a process called ‘inter-core coupling’. Inter-core coupling results in a blurring of the image which reduces its overall quality, and can therefore negatively affect the detection of disease. This blurring effect can be reduced by cleverly designing fibres with low inter-core coupling efficiency. However, even with the best fibre designs, it still remains a problem. We have developed a method of assessing how the light spreads through inter-core coupling of any particular fibre.


Inter-core coupling can be reduced, but not eliminated, through clever fibre design. It became apparent to us that developing an accurate description of a fibre’s inter-core coupling would lead to an understanding of the light spread and we could correct for the negative effects through the use of mathematics (image processing algorithms).


We performed our method on a commercially-available imaging fibre called FIGH-30-650S. In clinical practice, the whole end face of a fibre with its thousands of cores is illuminated by the sample of interest. In order to quantify inter-core coupling, we illuminated only one core at a time with a laser and measured the light spread that we saw coming out the other end of the fibre. By repeating this over hundreds of cores within the fibre, we quickly built up a distribution of inter-core coupling which we could then statistically analyse.


We were able to determine the typical inter-core coupling of this imaging fibre and can use the results to create image-processing algorithms that can enable better imaging for doctors. This method will be used on bespoke fibres that we design and produce at University of Bath for imaging the lung.

The image on the left shows the face of the fibre with many cores. By illuminating just one core with light (green), we can then measure the amount the light has spread once it has reached the other end of the fibre.
The image on the right is an example of the spread of light we observed. The central white core (marked with a red dot) is the core that we are intentionally illuminating. The other white cores are due only to inter-core coupling. It is clear that inter-core coupling can be quite extensive.