At some point in our lives, many of us invariably end up having to pay a visit to the hospital to have our digestive system examined. The reasons for this may be various, ranging from a simple check-up to make sure everything is working fine, to perhaps a more targeted visit to suss out the underlying condition behind that persistent heartburn or irritable bowel.

Either way, chances are that your local gastroenterologist may recommend a procedure called a gastroscopy to examine your stomach, or perhaps a colonoscopy to examine your large intestine. Both these procedures involve the use of a somewhat invasive instrument called a flexible endoscope – this is a long flexible tube with a camera at its tip which is inserted through your mouth or through your colon to permit the specialist to have a good look at your insides.

In most patients, these procedures are generally enough to establish a diagnosis. However, bearing in mind that there are approximately 17 feet of small intestine that are beyond the reach of normal endoscopes, there are undoubtedly going to be times when further tests are needed to look into the small bowel for abnormalities or causes of bleeding.

Until recently, such studies required surgical exploration of the abdomen, which is very invasive, or radiologic tests, which are less accurate. Thankfully however, some years ago biomedical engineers devised a new procedure which mostly accomplishes the latter objective through a less invasive form. The procedure is called capsule endoscopy and is designed to wirelessly image the small bowel through the use of a simple ingestible capsule, which is swallowed by the patient.

Equipped with a miniature camera, a battery, and some LED light sources, the capsule takes approximately four to 35 images per second as it travels down the digestive system. The images are transmitted wirelessly to a receiver strapped around the patient’s waist, who may in the meantime conveniently return home to continue with his or her daily business. When the capsule is eventually passed out by the patient, the receiver collates all the images into a video. This is then downloaded onto a computer for viewing by the specialist during the next visit.

Surprisingly, this seemingly advanced and futuristic technology has already been commercially available for about 10 years, and the procedure is in fact carried out even locally in the private healthcare sector. Nowadays, even different types of capsules are available, designed to examine the oesophagus or colon for instance.

However, before you rush to cancel any pending endoscopies in favour of this type of procedure, it might be worth noting that the capsules have some limitations in their use.

Capsule endoscopy is a purely diagnostic procedure – in other words, it may only be used by the specialist to see pictures and determine what is wrong in your digestive tract. This is in contrast to regular gastroscopies and colonoscopies, whereby the gastroenterologist may insert special instruments through working channels available in the flexible endoscope, in order to take samples, or to cauterise any tissue where necessary. This therefore makes capsule endoscopy only indicated in certain conditions, and your specialist would be able to prescribe it where appropriate.

A custom software algorithm was developed, which is capable of recognising in which digestive organ the capsule is located

A second issue is that the capsule’s movements are at the mercy of your own bowel movements. Hence, as the capsule travels downwards, its camera might coincidentally be facing the other way when passing close to a malignancy in your digestive tract – thus failing to register the problem entirely.

Modern capsules mitigate this problem by being equipped with two cameras on either end, effectively doubling up the capsule’s field of vision. However, the slight possibility of missing a malignancy remains present.

A third issue with the procedure is the huge quantity of images it generates. Regardless of the type of capsule used, these will indiscriminately take pictures until the battery runs out. This makes the screening task at the end rather laborious for the specialist, who has to examine a lot of images to make sure no important details are missed.

Over the past year, the Centre for Biomedical Cybernetics at the University of Malta has carried out collaborative research with Cardiff University and Saint James Hospital to devise a partial solution to this latter issue.

After consulting on some clinical aspects of the problem with Prof. Thomas Attard, paediatric gastroenterologist, and with Joe Garzia, endoscopy nurse, a custom software algorithm was developed which is capable of recognising in which digestive organ the capsule is located. By employing image processing techniques, the software examines certain features from each image, such as colour and texture, and uses these to determine where the image was taken from.

Images taken from the stomach, for instance, tend to have a pinkish hue, whereas images taken from the small intestine and large intestine tend to have yellow and green tinges respectively.

Additionally, the anatomy of the small intestine gives it a velvet-like texture, arising from small projections (called villi) which help in improving nutrient absorption.

These particular features belonging to different digestive organs are exploited by the algorithm, which then classifies and labels each image accordingly.

Once the images are tagged with such labels, the algorithm clearly demarcates the boundaries between digestive organs, such as that between the stomach and small intestine for instance.

The specialist may then reduce the screening time by focusing only on the images of a single digestive organ, particularly when this happens to be the suspect region of pathology.

While the software solution is still in its research phases, its underlying techniques and associated results have already been successfully published at the Engineering in Medicine and Biology conference in Japan last July.

Further developments might one day see the solution being implemented for use in day-to-day hospital examinations. The algorithm could also possibly be implemented directly onto the capsule and receiver system, thus creating an intelligent form of recording which adapts frame-rate or image quality when the desired part of the digestive system is detected.

The team at the Centre for Biomedical Cybernetics, headed by Prof. Kenneth Camilleri, will next be researching further image processing techniques that permit direct and automatic detection of abnormal conditions and pathologies within the digestive tract.

It is hoped that such techniques would also help to dramatically reduce screening time for the specialist, as the software solution would automatically scan through the entire video and highlight just the images which are showing potential malignancies.

The research team

Ing. Carl Azzopardi works as a biomedical engineer with the Centre for Biomedical Cybernetics and is specialised in the field of clinical engineering, with a particular focus on medical imaging. He also heads the Biomedical Engineering Department at Saint James Hospital Group.

Prof. Kenneth P. Camilleri is the director of the Centre for Biomedical Cybernetics and works extensively on the application of signal and image processing to biomedical engineering, with a particular interest in brain signal analysis and brain-computer interfacing.

Dr Yulia A. Hicks is a senior lecturer at Cardiff School of Engineering, Cardiff University. Her main research interests are in the areas of computer vision, image processing and statistical modelling.

The Centre for Biomedical Cybernetics

Recently set up at the University of Malta, the Centre for Biomedical Cybernetics aims to spearhead research in the foremost areas of biomedical engineering and cybernetics. Its competence lies in the application of systems theory, signal and image processing, computational intelligence and automation, among others fields, to medical problems.

The centre seeks to foster an interdisciplinary environment which brings together engineers and clinicians, whereby ideas may be shared, discussed, and taken forward in terms of academic research and innovation.

The centre invites medical professionals with an interest in developing novel ideas through the aid of technology to contact the centre in order to explore these further.

Acknowledgements

The author thanks Saint James Hospital and its Endoscopy Department for providing anonymised data for the study and for providing a demo capsule for imaging purposes. The author also thanks Prof. Thomas Attard and Joe Garzia for valuable clinical insight.

The research work disclosed in this article is funded by the Strategic Educational Pathways Scholarship (Malta). The scholarship is part-financed by the European Union – European Social Fund (ESF) under Operational Programme II – Cohesion Policy 2007-2013, Empowering People for More Jobs and a Better Quality of Life.