Wearables and AI Framework for Inflammatory Arthritis Assessment and Management

Using wearable sensors and machine learning to analyze real-world movement and sleep data, this project aims to predict treatment response earlier and enable more personalized care for inflammatory arthritis.
Ongoing Autoinflammatory Diseases Rheumatoid Arthritis Systemic Diseases Data-Driven & Quantitative Early Disease Detection Experimental Platforms & Models Human Cohorts Machine Learning & AI Population & Patient-Centered Precision Medicine Real-world Evidence Translational & Clinical

Funding Year: 2024

Project Overview

Inflammatory arthritis (IA) includes chronic conditions that cause joint pain, stiffness, and reduced mobility. While new treatments are available, clinicians still lack reliable ways to predict which therapies will work for individual patients.This project explores a non-invasive solution using wearable devices that continuously collect data on physical activity, sleep, and movement patterns. Thirty patients with active IA undergoing a change in therapy — alongside five healthy controls — wear an Actigraph LEAP device for 90 days, with wearable data compared directly to in-clinic outcomes. Machine learning algorithms are then applied to identify clinically meaningful metrics for arthritis management and to detect early signals of treatment response over time.

Impact & Innovation

Smarter monitoring, more personalized care

 

Using an Actigraph LEAP wearable worn by 30 patients over 90 days, this project applies machine learning to real-world movement data to predict treatment response and detect flares earlier than clinic visits allow.

  • Addresses the lack of reliable biomarkers for predicting which IA therapies will work for individual patients
  • Generates IP potential through ML algorithms for patient-specific, digital prediction of treatment response
  • Advances the Consortium’s Integrated Data and Discovery Platforms pillar, establishing continuous wearable monitoring as a new frontier in precision autoimmune care
Research Approach

A framework designed for discovery

This project integrates wearable sensor technology with machine learning to study disease activity in inflammatory arthritis. Participants wore a research-grade device continuously over a 90-day period, enabling passive, real-world data collection. Computational models were applied to detect patterns in movement, sleep, and activity that correlate with treatment response.

Continuous monitoring using the Actigraph LEAP wearable device, combined with machine learning analysis of longitudinal activity and sleep data.

Longitudinal wearable data (physical activity, sleep, gait, balance, vital signs) paired with direct clinical assessments from in-clinic visits; machine learning algorithms used to model temporal patterns and identify digital biomarkers.

Development of predictive models to identify early signals of treatment response and disease flare, supporting more timely, personalized treatment decisions and enabling remote monitoring alongside traditional care.


Investigators & Institutions

Powering the science

Principal Investigators

Rebecca Haberman, MD, Colton Consortium Member

Assistant Professor, Department of Medicine, NYU Grossman School of Medicine / NYU Langone Health

Souptik Barua, PhD, Colton Consortium Member

Assistant Professor, Department of Medicine, NYU Grossman School of Medicine / NYU Langone Health

Colton Centers & Institutions

Judith & Stewart Colton Center for Autoimmunity at NYU Langone Health

NYU Grossman School of Medicine

Research Outputs

From insight to impact

Publications

Spatial transcriptomics stratifies psoriatic disease severity by emergent cellular ecosystems

Science Immunology
Castillo, RL; Sidhu, I; Dolgalev, I; Chu, T; Prystupa, A; Subudhi, I; Yan, D; Konieczny, P; Hsieh, B; Haberman, RH; Selvaraj, S; Shiomi, T; Medina, R; Girija, PV; Heguy, A; Loomis, CA; Chiriboga, L; Ritchlin, C; Garcia-Hernandez, MDLL; Carucci, J; Meehan, SA; Neimann, AL; Gudjonsson, JE; Scher, JU; Naik, S June 2023
BioinformaticsData-Driven & QuantitativeDisease SubtypingExperimental Platforms & ModelsMulti-omics IntegrationSingle Cell TechnologiesSpatial BiologyTranslational & ClinicalAutoinflammatory DiseasesDermatologic DiseasesPsoriasisPsoriatic ArthritisNew York University

Racial and ethnic determinants of psoriatic arthritis phenotypes and disease activity

Rheumatology
Haberman, RH; Ahmed, T; Um, S; Zhou, YY; Catron, S; Jano, K; Felipe, A; Eichman, S; Rice, AL; Lydon, E; Moussavi, S; Neimann, AL; Reddy, SM; Adhikari, S; Scher, JU February 2025
Autoimmune EpidemiologyDisease SubtypingExperimental Platforms & ModelsHealth DisparitiesHuman CohortsPopulation & Patient-CenteredReal-world EvidenceTranslational & ClinicalDermatologic DiseasesPsoriasisPsoriatic ArthritisSystemic DiseasesNew York University