Segmentation of the human tongue musculature using MRI: Field guide and validation in motor neuron disease.
Authors
Affiliations (17)
Affiliations (17)
- School of Electrical Engineering and Computer Science, The University of Queensland, St Lucia, QLD, Australia; Centre for Advanced Imaging, Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, St Lucia, QLD, Australia; Department of Neurology, Royal Brisbane and Women's Hospital, Herston, QLD, Australia. Electronic address: [email protected].
- School of Electrical Engineering and Computer Science, The University of Queensland, St Lucia, QLD, Australia.
- School of Electrical Engineering and Computer Science, The University of Queensland, St Lucia, QLD, Australia; Griffith School of Medicine and Dentistry, QLD, Australia; Queensland Health, Queensland, Australia.
- School of Health and Rehabilitation Sciences, The University of Queensland, St. Lucia, QLD, Australia.
- School of Electrical Engineering and Computer Science, The University of Queensland, St Lucia, QLD, Australia; Queensland Digital Health Centre (QDHeC), The University of Queensland, Herston, QLD, Australia.
- School of Biomedical Sciences, The University of Queensland, St Lucia, QLD, Australia.
- School of Electrical Engineering and Computer Science, The University of Queensland, St Lucia, QLD, Australia; School of Psychology, The University of Queensland, St Lucia, QLD, Australia.
- Centre for Advanced Imaging, Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, St Lucia, QLD, Australia; ARC Training Centre for Innovation in Biomedical Imaging and Technology (CIBIT), Centre for Advanced Imaging, The University of Queensland, St Lucia, QLD, Australia.
- Department of Medical Imaging, Royal Brisbane and Women's Hospital, QLD, Australia; School of Medicine, University of Queensland, QLD, Australia.
- Department of Neurology, Royal Brisbane and Women's Hospital, Herston, QLD, Australia; Centre for Clinical Research, The University of Queensland, Herston, QLD, Australia.
- Neuroscience Research Australia, Sydney, NSW, Australia; Neuroscience, The University of NSW, Department of Neurology, Southeastern Sydney Local Health District, Sydney, NSW, Australia.
- Google DeepMind, London, United Kingdom.
- Department of Neurology, Royal Brisbane and Women's Hospital, Herston, QLD, Australia; Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, QLD, Australia.
- School of Health and Rehabilitation Sciences, The University of Queensland, St. Lucia, QLD, Australia; Speech Pathology, Royal Brisbane and Women's Hospital, Herston, QLD, Australia.
- Department of Neurology, Royal Brisbane and Women's Hospital, Herston, QLD, Australia; School of Biomedical Sciences, The University of Queensland, St Lucia, QLD, Australia.
- Neuroscience Research Australia, Sydney, NSW, Australia; Brain and Mind Centre, Faculty of Medicine and Health, University of Sydney, NSW, Australia.
- School of Electrical Engineering and Computer Science, The University of Queensland, St Lucia, QLD, Australia; Centre for Advanced Imaging, Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, St Lucia, QLD, Australia; ARC Training Centre for Innovation in Biomedical Imaging and Technology (CIBIT), Centre for Advanced Imaging, The University of Queensland, St Lucia, QLD, Australia.
Abstract
This work addresses the challenge of reliably measuring the muscles of the human tongue, which are difficult to quantify due to complex interwoven muscle types. We introduce a new semi-automated method, enabled by a manually curated dataset of MRI scans to accurately measure five key tongue muscles, combining AI-assisted, atlas-based, and manual segmentation approaches. The method was tested and validated in a dataset of 178 scans and included segmentation validation (n = 103) and clinical application (n = 132) in individuals with motor neuron disease. We show that people with speech and swallowing deficits tend to have smaller muscle volumes and present a normalisation strategy that removes confounding demographic factors, enabling broader application to large MRI datasets. As the tongue is generally covered in neuroimaging protocols, our multi-contrast pipeline will allow for the post-hoc analysis of a vast number of datasets. We expect this work to enable the investigation of tongue muscle morphology as a marker in a wide range of diseases that implicate tongue function, including neurodegenerative diseases and pathological speech disorders.