AI/ML Implementation & Enabling Infrastructure - Page 3 of 6

DIHI has developed a diverse portfolio of Augmented Intelligence & machine learning models and is responsible for the underlying infrastructure to facilitate their development and implementation into Duke Health. DIHI applies devops principles to the entirety of its infrastructure design to achieve reproducible, monitored, automated, and tested deployments of innovations at-scale for Duke.

a university building in the background, with a small lake and palm trees in the foreground

World Health AI Partnership

Focused on capacity building, best practice diffusion, and AI solution scaling between Duke Health and Aga Khan University in Kirachi, Pakistan.

EASY Telehealth Care: Optimizing telehealth in rheumatology

Enable rheumatology providers to identify patients who could be seen by telehealth rather than in-person at their next follow-up appointment.

patient with lower back pain being seen by a nurse

Using Machine Learning to Appropriately Triage Patients with Low Back Pain (LBP)

Develop and validate a machine learning algorithm capable of predicting patient need for spinal surgery or non-operative management utilizing elements from the electronic health record.

nurse talking with patient getting ready to perform a pulmonary function test

Machine Learning for Interpreting PFTs: Improving operational efficiency of pulmonary medicine

Automating PFT interpretation

abstract technological image depicting data science

Automating CMS Quality Measure Curation

Optimize the process and enhance the efficiency of quality measure reporting to removes unnecessary burden from clinical staff.

dwarf weeping Japanese maple tree in the foreground which arches over a park bench in the distance

Predict Behavioral Emergencies in the Hospital

Identifying patients at risk of having behavioral emergencies can help to optimize care delivery.

Enhanced Decision Support for Periop Care

Challenges related to hospital access, staffing constraints, and patient-related factors require a refined risk prediction model to enable risk-mitigation strategies earlier in the process.

water fountain sculpture in a botanical garden on a bright, sunny day

Improving Equity and Value of PAD Care at a Population Level

Identifying PAD patients has historically been difficult because diagnosis codes work very poorly for PAD cohorts.

Machine Learning Approach to Pediatric Sepsis Recognition

The Problem Mitigation of mortality in pediatric sepsis patients require early detection of sepsis warning signs, however vital signs and lab abnormalities are more ambiguous