PICTURE: Pediatric General Floor Deterioration Prediction

Written By Phil Jacokes

PICTURE is a suite of machine learning algorithms that utilize electronic health record (EHR) data to predict patient deterioration in hospital settings. The Pediatric General Floor Analytic is modeled after a pediatric data set and is used specifically for pediatric inpatients on a general floor.

Advancing Pediatric Healthcare

Early Detection for Better Outcomes
Identifying patient deterioration early can significantly improve health outcomes by reducing mortality, shortening hospital stays, and decreasing healthcare costs. PICTURE provides clinicians with the tools they need to detect these changes promptly and efficiently.

Transparent and Trustworthy Predictions:
PICTURE not only predicts but also explains. Each prediction comes with a clear breakdown of contributing factors, giving clinicians the transparency they need to trust and act on the insights provided.

Designed for Pediatric Excellence:
Tailored specifically with a pediatric focus, PICTURE’s General Floor Analytic incorporates vital age-specific features, such as capillary refill time, and caters to patients aged 1-22. This ensures a targeted approach to pediatric care that respects the nuances of young patients’ physiology.

Key Features

  • Explanatory Insights: Every prediction is accompanied by detailed explanations, enhancing understanding and application in clinical settings.

  • Seamless Integration: Utilizes existing routine medical data, smoothly integrating into clinical workflows without necessitating changes, even amidst evolving healthcare standards.

  • Customizable Precision Settings: Individual hospital units can customize precision thresholds to match their specific needs, aligning with care goals.

  • Novel Physiology-Based Modeling: By focusing on physiological indicators rather than clinician behavior, PICTURE ensures the alarms provide novel information to care providers.

Collaboration and Ongoing Research

AI & Digital Health Innovation is collaborating with the Weil Institute and the Michigan Institute for Clinical and Health Research (MICHR) to conduct 10 human-centered design sprints, which engage over 40 professionals from Michigan Medicine to optimize workflows and interface designs. Following these optimizations by the Clindoc team, a 3-month randomized control trial will rigorously test PICTURE’s real-world effectiveness.

 

Principal Investigators
Sardar Ansari, PhD
Kevin Ward, MD

Intellectual Property
Invention Disclosure # 2019-094, 2021-244, 2021-245, 2021-250, 2021-251
Patent Issued: #11587977

Phil Jacokes
Previous

Ambulatory Surgery Center Model
Next

PICTURE: Adult General Floor Deterioration Prediction