Objective: Collect routine facial expression data from people with Parkinson’s Disease (PwP) and healthy controls via user-friendly AI-powered eyewear and develop AI algorithms to distinguish PD from healthy controls.

Background: Hypomimia is common in PD.  Separate from its value as a biomarker for PD diagnosis and progression, hypomimia is correlated with overall motor and cognitive symptoms, and is independently linked to poor emotional recognition and social functioning.  While current attempts at PD symptom tracking focus on motor symptoms (tremor, gait), hypomimia is overlooked, partly due to a lack of suitable sensing technologies to monitor facial expression in everyday life.

Method: In collaboration with the Smart Computer Interfaces for Future Interactions (SciFi) Lab at Cornell University, we propose to deploy wearable technology to continuously track facial expressions from PwP and healthy controls.  Dr. Zhang’s lab has developed innovative, minimally-obtrusive low-power eyewear employing acoustic sensing technology to reconstruct users’ facial expression in real time.  We plan to recruit 40 subjects from Weill Cornell Movement Disorders Institute, encompassing PD patients at different stages (early- advanced) and healthy controls.  Subjects will wear prototype glasses, perform facial expressions and standard exercises once daily, three times per week, for eight weeks.  Pre- and post- levodopa facial expression data will also be collected.  In-person and telephone visits will assess data collection and device satisfaction.

Results: We expect to generate a first-of-its-kind dataset of continuous facial expression data in PD patients and healthy controls in daily life, with further qualifications on severity of PD disease state and levodopa response.  Utilizing this data set, we will develop AI-based algorithms to show proof-of-concept in using AI to diagnose and monitor PD through everyday facial expression analysis.  We will further establish the relationship between facial expressivity, PD disease state, and medication response.

Conclusion: Our proposal aims to expand the field of wearable technology with innovative eyewear to further explore and establish the value of facial expression in PD diagnosis, progression and treatment response.  Establishing feasibility and usability among our PD population is paramount to its success for larger studies.

« Back to 2024 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/ai-powered-eyewear-for-routine-facial-expression-analysis-in-parkinsons-disease-study-design-and-goals/