An innovative assessment of the dynamics of facial movements in surgically managed Unilateral Cleft Lip using 4D imaging

Gattani, Shyam (2020) An innovative assessment of the dynamics of facial movements in surgically managed Unilateral Cleft Lip using 4D imaging. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2020GattaniPhD.pdf] PDF
Download (12MB)



Verbal and non-verbal facial expressions form an integral part of everyday social interactions among human beings. Facial expressions may be distorted and asymmetric in conditions like craniofacial anomalies, trauma and facial palsy. Cleft lip and palate is one such craniofacial anomaly, that is managed surgically early in life by means of cleft lip repair surgery, nasal reconstruction and lip revision procedures. The goal of surgical repair is to ensure the renewal of optimum form and function in these patients. Despite these surgical procedures, a certain degree of asymmetry persists in patients, which result in decreased aesthetics and function in areas associated with the poorly approximated muscle bundle or due to scar tissue around the surgically corrected muscle group. This asymmetry that ensues after surgery, affects the performance of facial expressions. It is important to quantify facial movement and facial expressions in an objective fashion, in order to be able to assess the effectiveness of surgery and decide on the need for further revision surgery. A lot of work on quantification of facial movements in cleft lip and palate patients has been carried out previously using 2D and 3D imaging modalities. 2D imaging systems do not analyse the depth of images and therefore these methods lose out on an important dimension of facial morphology that needs to be assessed. 3D systems analyse the depth of images but are unable to assess the moving face (the fourth dimension) and only quantify facial movement on still or static images (3D). The human face is rarely static in day to day life and therefore measurement of facial expressions requires that facial expressions be recorded in a dynamic way in such a way that the speed, magnitude and pattern of facial movement be recorded. Each facial expression has specific muscle groups that undergo contraction and relaxation in different phases of the movement starting from rest to peak expression and final resting position. Some of these muscle groups are surgically corrected or re-approximated during cleft lip repair whereas other muscles are relatively or completely untouched. It is therefore necessary to analyse the entire movement sequence for asymmetry as this will highlight which specific muscle groups are associated with high asymmetry scores during the movement, thereby enabling the surgeon to plan on revision procedures accordingly.

Furthermore, it is also essential that the entire facial topography be assessed. A lot of the previous work done on quantifying facial movement and assessing facial asymmetry used individual landmarks on the face and measured the displacement of these landmarks as asymmetry scores. The disadvantage of using individual landmarks is that the remaining facial surface is not taken into consideration and therefore results obtained may not be representative of the entire facial structure. The novelty in this study is that it used a generic face mesh which undergoes elastic deformation in a process called ‘conformation’ in order to resemble the patient’s facial morphology. This conformed face mesh is then used to assess and quantify facial movement and measure facial asymmetry. This study therefore assesses the facial surface in its entirety along with analyzing asymmetry on a moving or dynamically changing facial surface. Colour maps were used to understand the directionality of facial asymmetry and asymmetry was seen in all three spatial planes-the x, y and z directions.


To characterise and assess facial asymmetry during four facial expressions- maximum smile, cheek puff, lip purse and grimace, in patients with surgically managed Unilateral Cleft Lip (UCL), using a real time 3D imaging (Di4D) system, and compare these with asymmetry that is seen in an age and sex matched control group.


Prospective two cohort comparative study. 25 surgically managed UCL cases and 75 controls at 8-10 years of age were recruited.


Facial movements during each of the four expressions were recorded using stereo-photogrammetry at a rate of 60 3D facial images per second. Each expression took approximately 3 seconds and generated 180 3D facial images for the analysis. A generic facial mesh which consists of more than 7000 quasi landmarks, was used for the assessment of facial asymmetry. This was wrapped (conformed) on the 3D image, the conformed mesh was mirror imaged, the original and mirror images were superimposed and the distances between the corresponding landmarks of the original and mirrored images of the five selected frames quantified facial asymmetry.


Statistically significant differences were seen regarding the magnitude of facial asymmetry between the UCL group and the non-cleft controls in all four facial expressions.

In maximum smile, asymmetry was prominent in the nasal regions- the nasal tip and alar regions as well as the upper lip vermillion border, the philtrum and columella. Higher average asymmetry in the UCL group in the total face was seen in the 3D frame mid-way between maximum smile and rest (frame 4) followed by the frame at peak expression of maximum smile (frame 3).

In cheek puff, asymmetry in the total face was seen most pronounced in frame 2 as the face was moving into the phase of peak facial expression. In the X direction, asymmetry was most noted in the alar regions and philtrum and upper lip regions. Vertical asymmetries were minimal. Anteroposterior asymmetry was most noted in frame 2 in areas around the ala of the nose and upper lip vermillion and philtrum.

In lip purse, asymmetry in the UCL group was seen to be higher in the peak frame (frame

3) followed by frame 4 and 2 for the total face. Asymmetry in the total face was most pronounced in the alar regions of the nose, the philtrum and upper lip vermillion border.

In grimace, asymmetry in the UCL group was seen to be higher in the mid-way frame (frame 2-mid-way between initial resting frame and peak expression of cheek puff) followed by frame 3 for the total face.

Vertical asymmetries were seen to be minimal in all facial expressions except in the case of lip purse. The nasolabial region showed maximum asymmetry during facial movement.


This study provides a sensitive and innovative tool to assess and quantify the dynamics of facial muscle movement, which correspondingly highlights the anatomical areas of residual asymmetry. This further enables surgeons to understand where the deficiency exists and accordingly plan revision procedures to improve facial symmetry and therefore optimally revive facial form and function. This thesis identifies the limitations and inability of primary lip repair to bring about symmetrical outcomes and helps understand the modifications and refinements required during cleft lip repair which ultimately improves the quality and efficacy of the surgery and the quality of life in patients with cleft lip and palate.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: Cleft, asymmetry, symmetry, quantification, objective, Cleft Lip, UCL, palate.
Subjects: R Medicine > RK Dentistry
Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Medicine, Dentistry & Nursing > Dental School
Supervisor's Name: Ayoub, Professor Ashraf, Ju, Dr. Xiangyang, Bell, Dr. Aileen and Gillgrass, Mr. Toby
Date of Award: 2020
Depositing User: Mr Shyam Gattani
Unique ID: glathesis:2020-81870
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 05 Jan 2021 15:46
Last Modified: 07 Sep 2022 13:21
Thesis DOI: 10.5525/gla.thesis.81870
Related URLs:

Actions (login required)

View Item View Item


Downloads per month over past year