Use of facial asymmetry in better diagnosis and treatment of Plagiocephaly

Lead Research Organisation: University of the West of England
Department Name: UNLISTED


There is an increasing tendency for young children to develop a medical condition (known a Plagiocephaly), which can result in deformity of their heads and distortions of the faces. It is possible to help remove the condition by having the child wear a carefully designed helmet that encourages the skull to grow in a direction that will tend to minimize the deformity. In order to manufacture such a helmet, the doctors involved need to have an accurate model of the shape of the child?s head. At the moment this model is made by making a plaster cast of the head; which is a procedure that that can be quite upsetting for the child, or which necessitates the use of a general anesthetic, with all the associated risks which that entails. The proposed project aims to overcome this problem by using cameras and lights in order to quickly and accurately measure the child?s head and face. As well as enabling manufacture of the helmet, this data will help with finding a link between the deformity of the head and the distortions of the face. This will help to show whether looking closely at the face could be way to diagnose Plagiocephaly. It will also provide a method for monitoring and evaluating the success of the helmet treatment.

Technical Summary

Plagiocephaly is a cranial disorder where the two sides of the skull develop inconsistently, so that the shape of head exhibits a pronounced asymmetry, flattening or other abnormal shape. This research project will combine novel and advanced machine vision techniques, 3D metrology system and shape modelling in order to determine:

Whether Plagiocephaly is always associated with a measurable asymmetry of the face.

How close a correlation there is between the nature of 3D head deformation and 3D face shape. If there is a close correlation, can face shape / asymmetry be used as a diagnostic aid or in assessing treatment? (i. e. do we always need to look at the whole head?)

If it possible to differentiate between differing conditions, e.g. position Plagiocephaly, ocular Plagiocephaly, Torticollis, etc using 3D face measurements. i. e. do they have particular, distinct 3D face signatures?

We propose to employ a photometric stereo (PS) method for the capture of 2D and 3D surface details of the human face, and subsequently to analyze the symmetry of the whole face based on its 2D and 3D information, and to use this information to assess the severity of Plagiocephaly.


10 25 50