2026 American Academy of Neurology Abstract Website

Kinan Muhammed¹, Francis Chmiel¹, Ledia Agley¹, Krystal Hattar¹, Michele Tagliati², Bridget Frommel², Cheri Phillips², Shreya Rai³, Diego Rodriguez³, Olivia Laun³, Jennifer Liu³, Widad Abou Chaar³, Barbara Changizi³, Daniel El Kodsi³, Vikram Khurana³, Sheeba Mason⁴, Andrew Graham⁴, Teresa Clarke⁴, Rachel Rich⁴, Jeremy Cosgrove⁵, Stephen Butterworth⁵, Amanda Hulejczuk⁶, Mary Burt⁶, Rebecca Heslop⁶, Antonella Macerollo⁷, Richard Ellis⁷, Joanne Bromley⁸, Nicola Findlay⁸, Damian Jenkins⁸, Nagaraja Sarangmat⁸, Michele Hu⁹
¹Kneu Health, ²Cedars-Sinai Medical Center, ³Division of Movement disorders and American Parkinson’s Disease Association Center for Advanced Research, Brigham and Women’s Hospital, ⁴East Suffolk and North Essex NHS Foundation Trust, ⁵The Leeds Teaching Hospitals NHS Trust, ⁶University Hospitals Sussex NHS Foundation Trust, ⁷The Walton Centre NHS Foundation Trust, ⁸Oxford University Hospital NHS Foundation Trust, ⁹University of Oxford

Objective:

To evaluate real-world use of a multi-modal digital platform (Kneu Health) for remote monitoring of idiopathic Parkinson’s disease (PD) across UK and US health systems.

Background:

PD prevalence is rising, creating substantial morbidity and strained healthcare resources. Challenges include limited capacity for frequent monitoring, objective medication management, and access to care. Scalable, objective tools to assess and aid patient management are needed internationally.

Design/Methods:

A smartphone based platform was developed to objectively quantify PD signs and symptoms via digital assessments of voice, balance, gait, finger dexterity, reaction time, rest/postural tremor, and cognitive function. The platform was deployed across 11 UK NHS hospitals and 2 US healthcare providers. Use per site ranged from 3-18 months. 726 patients completed smartphone assessments in a real-world setting. Machine learning models were used to estimate scores including MDS-UPDRS, and key metrics were displayed on a clinician dashboard. Use in clinics was evaluated, including clinical utility, efficiency and service impact. An additional 515 PD patients provided paired clinical data to train models.

Results:

Over 800,000 digital measures were completed. Smartphone derived motor features correlated with corresponding MDS-UPDRS components (r=0.12 to 0.51, p<0.05) and were sensitive to daily symptom fluctuations. 93% of patients found the technology easy to use and sustained improvements of up to 30% were observed across patient empowerment, confidence, and knowledge. Dashboard data supported medication optimization in 53% of clinical consultations and saved on average 20% consultation time. At the most mature UK site, PD related emergency admissions decreased by 1.2% versus an 8.6% national increase. In the US, 11% of patients reported avoiding an unplanned hospital or emergency visit within 3-6 months of use.

Conclusions:

Remote smartphone-based assessments enable clinically actionable longitudinal monitoring of PD in UK and US healthcare settings. The approach was acceptable to patients, improved clinical decision-making, and demonstrated real-world impact on service outcomes.