DETECT-ARDS

Overview

DETECT-ARDS is an analytic for detecting Acute Respiratory Distress Syndrome (ARDS). It enables early and accurate detection of the acute respiratory distress syndrome by leveraging artificial intelligence (AI) to interpret and understand visual images with human-level accuracy.

Value Proposition to Michigan Medicine

DETECT-ARDS is a new approach for identifying ARDS findings on chest x-rays. With ARDS often missed or under-diagnosed, DETECT-ARDS has the potential to transform patient outcomes for the better. By training powerful algorithms called deep convolutional neural networks (CNNs), the system can identify findings consistent with ARDS with high accuracy.

Principal Investigator(s)

Michael Sjoding, MD
Sardar Ansari, PhD

Clinical Champion

MM Pulmonary and Critical Care

Digital Health Innovation Support

Digital Health Innovation (DHI) supported the Weil Institute by developing the data pipeline to pass CXR images into the model. In collaboration with CIC and HITS, DHI helped deploy the model to MiChart (Epic Nebula) to prospectively evaluate model performance.

Partnerships

Weil Institute 

Clinical Intelligence Committee (CIC)

Health Information Technology and Services (HITS)

MM Pulmonary and Critical Care

Publications

https://pubmed.ncbi.nlm.nih.gov/33893070/

Intellectual Property

Invention Disclosure # 2020-026

Patent Issued: