Skip to content
Division of Health AIDivision of Health AI
TeamResearchPublicationsJoin
TeamResearchPublicationsJoin
Theme
Division of Health AIDivision of Health AI

Site

  • Team
  • Research
  • Publications
  • Join

Affiliated institutions

  • Feinstein Institutes (opens in new tab)
  • Northwell Health (opens in new tab)
  • Zucker School of Medicine (opens in new tab)

© 2026 Division of Health AI, Northwell Health. All rights reserved.

Cross-section of the human vagus nerve with individual fibers mapped in red and green

Research

Human vagus nerve fiber map, 60 donor nerves · Photo: Northwell Health

Federally funded programs

Federally funded

Health AI Assurance

FDA-funded validation of electrocardiogram devices and independent evaluation of clinical AI on Northwell's own patients, before and after a tool reaches the bedside.

Federally funded · R01NR020774

In-hospital deterioration prediction

Deep learning models that estimate a hospitalized patient's risk of deterioration continuously, from the electronic health record and from wearable sensors, so that clinicians are warned hours before a rapid response or ICU transfer.

Federally funded · U01OH012050

Transcutaneous auricular vagus nerve stimulation for PTSD in World Trade Center responders

A feasibility and acceptability study of daily, non-invasive auricular vagus nerve stimulation to reduce PTSD symptoms among 9/11 responders, twenty years after the attacks.

ONSET

Federally funded · R01AG100193

Delirium identification and prediction

Machine learning that identifies delirium in hospitalized older adults from clinical notes, using large language models, and predicts who is at risk before it develops.

Federally funded · SPARC

REVA: the human vagus nerve atlas

A comprehensive anatomical dataset of 60 human vagus nerves (30 left, 30 right), spanning millions of micro-CT images across hundreds of terabytes of data. Using 3D nnU-Net segmentation, this project builds a detailed vagus digital twin to guide the design of selective vagus nerve stimulation therapies, part of a $6.7 million NIH SPARC award in collaboration with the TNP Lab.

Elmezzi Scholar projects

Elmezzi Scholars

Sleep apnea screening in women

A machine learning approach to screening women for moderate-to-severe obstructive sleep apnea, a condition underdiagnosed in women because standard screening was built around male presentation.

Elmezzi Scholars

Uveal melanoma screening from fundus photographs

A systematic review and meta-analysis of artificial intelligence that distinguishes uveal melanoma from benign nevus on fundus photographs, the basis for automated screening in ophthalmology.

Elmezzi Scholars

Predicting response to vagus nerve stimulation in drug-resistant epilepsy

Machine learning models can predict which patients with drug-resistant epilepsy will respond to vagus nerve stimulation therapy, achieving an average area under the receiver operating characteristic curve (AUROC) of 0.84 across 12 studies comprising 535 patients.

Elmezzi Scholars

Deep learning in pancreatic cancer and PET interpretation

Machine learning to predict completion of treatment for pancreatic cancer, and systematic reviews of deep learning for PET image interpretation and for pancreatic cancer care.

Division projects

Division project

Computational models of vagus nerve stimulation

Histologically realistic models of the cervical vagus nerve that predict which fibers a stimulation pattern will activate, used to design selective, organ-specific vagus nerve stimulation.

Division project

Decoding the vagus nerve: preclinical recording and interpretation

Recording from the mouse vagus nerve, chronically and wirelessly, and decoding what its neurons report about inflammation and metabolism, toward bioelectronic devices that diagnose and treat from neural signals.

Division project

Nursing workforce forecasting

Machine learning models using DeepAR probabilistic forecasting predict nursing workforce demand across Northwell's hospital units up to 12 months ahead, supporting preemptive hiring and staffing decisions across diverse specialties.

Division project

ECG deep learning

Generative and predictive models over the electrocardiogram: reconstructing the full twelve-lead ECG from a minimal set of leads, and stratifying acute coronary syndrome risk from the admission ECG.

Division project

Maternal fever / neonatal sepsis prediction

Continuously monitored vital signs and heart rate variability during labor predict maternal fever 2-3 hours before clinical onset, with area under the curve of 0.748, enabling early detection of mothers at risk for neonatal early-onset sepsis.

Concluded work

  • Cardiac electrophysiology: arrhythmia prediction, VNS for ectopy, and closed-loop CPRPredicting atrial fibrillation from implantable defibrillator transmissions, suppressing premature ventricular contractions with transcutaneous vagus nerve stimulation, and modeling carotid blood flow for personalized mechanical CPR.2019 to 2025Recent paper
  • COVID-19 clinical decision supportMachine learning models built and deployed during the pandemic: clinical phenotyping, deterioration and respiratory failure prediction, and self-updating prognostic models, studied across cohorts in New York and the Netherlands.2020 to 2023
  • Ambulatory no-show predictionA machine learning model, developed and validated on Northwell ambulatory visits, that predicts which appointments will not be kept so that clinics can act in advance.2022 to 2023
  • Lung opacity estimation from chest radiographsA deep transfer learning framework adapted to estimate lung opacity burden from chest X-rays.2022 to 2023
  • Autonomic function quantification and taVNS physiologyMethods to quantify autonomic nervous system function in healthy adults and to measure physiological responses to transcutaneous auricular vagus nerve stimulation, including a clinical study of taVNS for pain and fatigue in lupus.2020 to 2021

More

Read the papersMeet the teamInternships and PhD training