In the UK, about 25% of children with repaired clefts need further surgery as they grow up to improve their communication. To help them decide if a person needs surgery, surgeons and Speech & Language Therapists look at how a person's soft palate moves when he or she talks using x-rays or endoscopy (looking with a telescope at how the palate is functioning). Neither of these investigations gives a three-dimensional view and neither allows the muscles of the palate to be identified. Endoscopy is invasive, uncomfortable and is often not possible in very young children.

In 2008, CLEFT provided £7,500 towards the purchase of an MRI compatible microphone to record people speaking inside the scanner (as it is a very strong magnet a standard one would not work). The microphone has since been used by the MRI physics team at Barts Health NHS Trust as part of their work on developing techniques to image the palate during speech and in three-dimension at rest (

In 2013, CLEFT funded a further £65,000 towards a PhD studentship to find better ways to take Magnetic Resonance Imaging (MRI) videos of the inside of a person’s mouth while they are talking. The videos are usually two-dimensional videos of the central slice of the head (similar to the view of the Queen’s head on a coin) and are taken using an MRI scanner.

The image above shows nine consecutive MR images of a volunteer saying "one".

MRI techniques have improved over recent years and offer the potential for a much better understanding of the function of the palate and its individual muscles. It is hoped that the three-dimensional anatomical information which MRI can offer will enable clinicians to see the movement of the soft palate more clearly, and also to take detailed pictures of the muscles that move the soft palate.

The video clip shows a 13 year old cleft patient counting on an MRI. It is this kind of information that will help surgeons and Speech and Language Therapists to make more comprehensive pre- and post-operative assessments of cleft patients and improve our understanding of how the palate functions.

The researcher, Andreia Freitas, looked at ways to improve how images are taken by the scanner (image acquisition) and how the computer puts the image data together (image reconstruction). The research found ways to take videos with a better image quality more quickly on the type of scanner usually found in a hospital. Two scientific journal publications describe this research (Freitas 2016, Freitas 2018). The PhD research was then built on to look at ways to take videos from several different angles at the same time (Ruthven 2019).

 There are problems with carrying out MRIs in young children because it is a challenging environment for them – they have to lie still as they go through the ‘tunnel’ to be scanned and the machine emits a loud noise. The study was carried out on older children and adults for this reason.

This is ground-breaking in the investigation of cleft palate as it is a dynamic, moving MRI rather than a static MRI picture. It could also be used in other areas of medicine. The possibility of a non-invasive, non-irradiating investigation of inadequate palate function which is acceptable to young patients makes this an important piece of research.

Following a successful application for funding to the National Institute for Health Research (NIHR), we are now looking at how these videos could be joined together to create three-dimensional videos for each patient.



Miquel ME, Birch MJ, 2013, Magnetic Resonance Imaging of the Cleft Palate: from technical implementation to patient management, CLEFT, £72,246

Miquel ME, Birch MJ, 2008, Magnetic Resonance Imaging of speech in patients with repaired cleft palate, CLEFT, £7,500


Freitas AC, Wylezinska M, Birch MJ, Petersen SE, Miquel ME (2016) Comparison of Cartesian and non-Cartesian real-time MRI sequences at 1.5T to assess velar motion and velopharyngeal closure during speech. PLOS ONE, DOI:

Freitas AC, Ruthven M, Boubertakh B, Miquel ME (2018) Real-time speech MRI: commercial Cartesian and non-Cartesian sequences at 3T and feasibility of offline TGV reconstruction to visualise velopharyngeal motion. European Journal of Medical Physics – Medica Physica 46:96-103.

Ruthven M, Freitas AC, Boubertakh B, Miquel ME (2019) Application of radial GRAPPA techniques to single- and multi-slice dynamic speech MRI. Magnetic Resonance in Medicine 82:948-958.

PhD thesis:

de Freitas AC (2018) Improving real-time MRI for the clinical assessment of velar closure and velopharyngeal motion during speech. PhD thesis, Queen Mary University of London.