Desktop software for 3D facial analysis to assist in equitable diagnosis, treatment monitoring and surgical planning.
Facial appearance has long been understood to offer insight into a person’s health. To an experienced clinician, atypical facial features may signify the presence of an underlying rare or genetic disease. Further changes in facial features can also be observed in pharmacological treatments.
Historically, measuring the size and shape of a patient’s facial features (facial morphometrics) in clinical diagnosis has relied upon the use of manual measuring aids such as callipers, rulers or tape measures, and angle measuring tools such as protractors which are held against the appropriate part of a patient’s face to read off a measurement.
The correct use of such measuring tools requires training to ensure that the correct part of the face is being measured, and a steady hand (and cooperative patient) to reliably make precise enough measurements. In recent years, 3D imaging systems have enabled clinicians to take facial measurements with increased precision, accuracy, and reliability. This research aimed to support and further the integration of 3D facial measurement and analysis for enhanced clinical utility by showing that a tool such as Cliniface offers increased accuracy, precision, and reliability over existing manual methods of measurement and monitoring.
Cliniface engaged with multiple organisations, including researchers, clinicians, pharmaceutical companies, clinical trial companies and state and national health agencies to deliver a multifaceted program of work.
FrontierSI has supported the Cliniface project since its inception and continues to provide project management and end-user engagement support as the project continues to scale. Clinical lead, Prof Gareth Baynam and his team at Genetic Services WA have successfully raised the awareness of 3D facial analysis through national and international networks in the rare and undiagnosed disease realm. This has enabled significant support for continued development of the software and importantly assisted in transitioning the research into clinical practice.
Cliniface has been developed by a multidisciplinary team from Curtin University and is a free and open-source desktop software application that provides a suite of tools to visualise, measure, and analyse 3D facial images. Cliniface can automatically place landmarks on a face, extract and show standard facial measurements, and then make inferences about atypical features of clinical significance, that when used together with other phenotypic information about a patient may assist in the diagnosis of an underlying rare disease. In addition, more recent developments have enabled Cliniface to detect changes as well as differences in facial features which can now be applied to treatment monitoring, clinical trials and facial surgery.
This research showed that non-experts with minimal guidance can use Cliniface to extract facial anthropometrics from a 3D facial image at a level of accuracy comparable to an expert. Further, it showed that Cliniface itself can extract the same measurements at a similar level of accuracy – completely automatically.
An additional and vital aspect to the Cliniface project is the collection of facial data of underrepresented populations in order to build population specific norms for analysis. Prior to Cliniface, analysis of Aboriginal people was performed against African America populations norms which skewed results significantly. With population specific norms, analysis can be significantly more accurate, avoiding incorrect diagnosis.