Andrea Bandini 1, Jordan R. Green 2, Babak Taati 1,3,4, Silvia Orlandi 5, Lorne Zinman 6,7, Yana Yunusova 1,7,8
1 UHN-Toronto Rehabilitation Institute, Toronto, Canada
2 MGH-Institute of Health Professions, Boston, USA
3 IBBME, University of Toronto, Canada
4 Department of Computer Science, University of Toronto, Canada
5 BRI, Holland Bloorview Kids Rehabilitation Hospital, Toronto, Canada
6 Neurology, Sunnybrook Healthy Sciences Centre, Toronto, Canada
7 Brain Sciences, Sunnybrook Research Institute, Toronto, Canada
8 Department of Speech-Language Pathology, University of Toronto, Canada
Abstract
The analysis of facial movements in patients with amyotrophic lateral sclerosis (ALS) can provide important information about early diagnosis and tracking disease progression [1-2]. The use of expensive motion tracking systems [3] has limited the clinical utility of the assessment. We propose a video-based approach to discriminate patients with ALS from neurotypical subjects. Facial movements were recorded using a depth sensor (Intel® RealSense™) during speech and non-speech tasks. A small set of kinematic features of lips was extracted using the Supervised Descent Method for face alignment to mirror the perceptual evaluation performed by clinicians, considering the following aspects: range of motion, speed of motion, symmetry, and shape. Our results demonstrate that it is possible to distinguish patients with ALS from neurotypical subjects with high accuracy (~89%) during repetitions of sentences, syllables, and labial non-speech movements.
This study provides strong rationale for the development of automated systems to detect neurological diseases from facial movements. Patients can benefit from this automated assessment, as an early diagnosis implies earlier interventions to maintain acceptable standards of quality of life during disease progression. This work opens new possibilities to develop intelligent systems to support clinicians in their diagnosis, introducing novel standards for assessing the oro-facial impairment in ALS, and tracking disease progression remotely from home.
[1] Y. Yunusova et al., “Kinematics of disease progression in bulbar ALS,” Journal of Communication Disorders, vol. 43, no. 1, Jan-Feb. 2010, pp. 6-20.
[2] P. Rong et al., “Predicting early bulbar decline in amyotrophic lateral sclerosis: A speech subsystem approach,” Behavioural Neurology, Jun. 2015.
[3] A. Bandini et al., “Kinematic features of jaw and lips distinguish symptomatic from pre-symptomatic stages of bulbar decline in amyotrophic lateral sclerosis (ALS),” Journal of Speech, Language, and Hearing Research, vol. 61, 2018, pp. 1118-1129.