Work

Real outcomes, clearly measured.

Real projects across sleep-medicine AI, clinical automation, and population research — each with outcomes we can point to and a detailed write-up of how we got there.

UK Biobank study figure: population-scale accelerometry and ML-derived RBD score (methods) alongside hazard ratios by RBD-risk stratum, survival curves, and PD polygenic-risk results.

Sleep Medicine / NeurodegenerativeUK Biobank Cohort Study

Prodromal RBD Risk Stratification

Population-level risk model identifying prodromal neurodegeneration from wrist actigraphy

Parkinson's Disease
Wearables
Sleep Medicine
UK Biobank

Read case study →

Workflow automation diagram showing the automated sleep study reporting pipeline

Clinic OperationsAnonymous Sleep Clinic

Sleep Clinic Workflow Automation

Automated reporting pipeline eliminating 15+ hours of manual work per week

Clinic Automation
Sleep Medicine
EHR
Workflow

Read case study →

ML classification dashboard showing cataplexy feature importance and NT1 risk stratification

Sleep MedicineStanford Sleep Medicine Center, USA

Optimizing ML Classification of Narcolepsy Type 1 Using Cataplexy Questionnaire & HLA Biomarker

ML-based screening pipeline enables 94% sensitive, 98% specific NT1 detection from brief questionnaire + genetic marker

Sleep Medicine
Machine Learning
Narcolepsy
Cataplexy

Read case study →