Related Experiment Video
Updated: Jun 18, 2025
09:52
Setting Up a Stroke Team Algorithm and Conducting Simulation-based Training in the Emergency Department - A Practical Guide
Published on: January 15, 2017
17.2K
An App for Navigating Patient Transportation and Acute Stroke Care in Northwestern Ontario Using Machine Learning:
Ayman Hassan1,2,3, Rachid Benlamri4, Trina Diner1
1Thunder Bay Regional Health Sciences Centre, Thunder Bay, ON, Canada.
JMIR Formative Research
|August 1, 2024
Summary
A new mobile app, NWO Navigate Stroke, helps healthcare providers in Northwestern Ontario make faster, more informed decisions for acute stroke patient transfers. This tool optimizes care access and improves patient outcomes in a region with dispersed resources.
Area of Science:
- Medical Informatics
- Health Services Research
- Machine Learning in Healthcare
Background:
- Coordinated care is crucial for timely acute stroke treatment, especially in widespread regions like Northwestern Ontario (NWO).
- Limited resources and dispersed facilities in NWO necessitate patient transfers for optimal stroke care.
- Healthcare providers unfamiliar with the region may face challenges in accessing timely stroke care, leading to suboptimal decisions and increased costs.
Purpose of the Study:
- To develop a decision-support tool for healthcare providers to optimize stroke patient transfer options in NWO.
- To create a mobile application integrating geomapping and machine learning for efficient stroke care navigation.
- To utilize key stroke timelines, patient location, and facility resources for informed transfer decisions.
Main Methods:
- A machine learning prediction model was developed using historical air (Ornge) and land medical transport data (2008-2020).
- Data preprocessing and cleaning were performed, with merging of transport data when both air and land services were involved.
- The dataset was divided for training (35%), testing (35%), and validation (30%) of the prediction model.
Main Results:
- Over 70,000 records from Ornge and land transport services were used to build the prediction model.
- The decision tree model demonstrated superior accuracy compared to other machine learning models analyzed.
- The developed "NWO Navigate Stroke" system, based on the decision tree model, showed accurate results in simulations.
Conclusions:
- The NWO Navigate Stroke system offers a data-driven, reliable, and accurate prediction model for acute stroke management.
- The mobile application serves as a significant tool for healthcare providers navigating stroke care logistics in NWO.
- Usability testing with end-users is the planned next step following historical data validation.
Keywords:
OntarioTIAacute strokeappapplicationsappsbraincardiovascularcoordinated caregeographicalgeographygeomappinglocationmHealthmachine learningmobile healthmodelsnavigatingnavigationneurosciencenorthwesternpredictpredictionpredictionspredictivespatialstrokestroke caretransient ischemic attack
