Project ID BE-MI2024_11


Co Supervisor 1A Faculty of Dentistry, Oral & Craniofacial Sciences,Centre for Oral, Clinical & Translational SciencesWebsite

Co Supervisor 1B Faculty of Dentistry, Oral & Craniofacial Sciences, Centre for Oral, Clinical & Translational SciencesWebsite

Additional Supervisor David Bartlett

Big Data in Dentistry: Can Morphometrics Facilitate Accurate Automated Registration?

Registering or superimposing sequential intraoral scans to diagnose potentially pathological change between dental appointments is problematic. We have shown that the error introduced by registration can often be greater than pathological disease progression when measuring at a micron level. While this can in part be overcome by manually identifying and registering on areas which are least likely to have changed, this is subjective, time consuming and difficult to automate. These limitations make assisted computer-generated diagnosis inaccessible for use in dental primary care. This project aims to
1. Create the first high-quality, open access, large oral datasets.
2. Develop frameworks for automated registration of sequential oral scans using big data morphometrics
3. Compare the developed frameworks with human operators with different scanner resolutions and assess each system accuracy. This will provide a toolkit with cross-fertilisation potential to a variety of systems and scanners within medical imaging.
The student will learn a diverse range of skills relating to both big medical image data analytics and morphometrics. Interacting with clinicians, they will learn how to acquire and de-identify clinical images as part of data preprocessing. They will become proficient in applying multivariate statistical techniques, machine learning algorithms, data manipulation and data visualization. They will also learn what is required to translate their findings and developments to clinical use.

Year 1: Creation of high-quality large datasets: At King’s we have access to one of the largest dental clinical databases globally in addition to validated data access SOP’s and both data scientist and clinical academic mentors.
Year 2: Assessing and quantifying different morphometric measures with an aim to detect areas of relative intraoral stability.
Year 3: Development of registration/quantification tool to quantify change in sequential scans.
Year 4: Assessing accuracy compared to gold standard under different scanning parameters and preparing for translational into clinical use.

Representative Publications

1. ‘O’Toole S, Osnes C, Bartlett D, Keeling A. Investigation into the accuracy and measurement methods of sequential 3D dental scan alignment. Dent Mater 2019 Mar;35(3):495-500. doi: 10.1016/

2. O’Toole S, Charalambous P, Almatrafi A, Mukar S, Elsharkawy S, Bartlett D. Progress and limitations of current surface registration methods when measuring natural enamel wear. J Dent. 2021 Sep;112:103738. doi: 10.1016/j.jdent.2021.103738. Epub 2021 Jun 25.

3. O’Toole S, Bartlett D, Keeling A, McBride J, Bernabe E, Crins L, Loomans B.Influence of scanner precision and analysis software in quantifying three-dimensional intraoral changes: two-factor factorial experimental design J Med Internet Res. 2020 Nov 27;22(11):e17150. doi: 10.2196/17150.

1. Charalambous P, O’Toole S, Bull T, Bartlett D, Austin R. The measurement threshold and limitations of an intra-oral scanner on polished human enamel. Dent Mater. 2021 Apr;37(4):648-654. doi: 10.1016/

2. Jadeja S, Austin R, Bartlett D. Use of polyvinyl siloxane impressions to monitor sub-5-μm erosive tooth wear on unpolished enamel. J Prosthet Dent. 2023 May 25;S0022-3913(23)00276-7. doi: 10.1016/j.prosdent.2023.04.020.

3. Charalambous P, O’Toole S, Austin R, Bartlett D. The threshold of an intra oral scanner to measure lesion depth on natural unpolished teeth. Dent Mater. 2022 Aug;38(8):1354-1361. doi: 10.1016/