Related Experiment Video
Updated: May 13, 2026
07:31
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
8.2K
Understanding Responsible Development in AI-Based Clinical Prediction Models for Mortality: Protocol for a Scoping
Riley Martens1, Jessalyn K Holodinsky1,2,3,4, Jessica Simon1,5,6
1Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, 3280 Hospital Dr NW, Calgary, AB, T2N 5A1, Canada, 1 (403) 220 6940.
JMIR Research Protocols
|March 11, 2026
Summary
Artificial intelligence-based clinical prediction models (AIPMs) can improve mortality prognostication but risk worsening health inequities. This review synthesizes literature on AIPM development and application to promote responsible innovation in healthcare.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Healthcare
- Clinical Decision Support
Background:
- Prognostic inequity creates barriers to end-of-life care for underrepresented groups.
- Artificial intelligence-based clinical prediction models (AIPMs) offer potential for accessible mortality prognostication but may exacerbate existing health inequities due to biased data and opacity.
- Ethical considerations are crucial for the responsible development and deployment of AIPMs.
Purpose of the Study:
- To synthesize peer-reviewed literature on the creation and application of AIPMs for mortality prognostication in adult acute care settings.
- To provide insights into responsible and ethical model development for AIPMs.
- To identify key elements in the development of AIPMs for mortality prediction.
Main Methods:
- A transdisciplinary search strategy across multiple academic databases (Medline, Embase, IEEE Xplore, ACM Digital Library, Compendex, Scopus) was employed.
- Literature screening involved two rounds (titles/abstracts, then full texts) with specific eligibility criteria.
- Data analysis will utilize descriptive, summary, and qualitative synthesis informed by the responsible research and innovation (RRI) framework.
Main Results:
- The literature search was completed on March 25, 2025, with screening initiated in May 2025.
- Results are anticipated by January 2026.
- This review will detail the specific elements included in the development of AIPMs for mortality prediction.
Conclusions:
- This review will offer a comprehensive summary of AIPMs for mortality prediction.
- The study will analyze AIPM development through the lens of the responsible research and innovation (RRI) framework, focusing on anticipation, reflexivity, inclusion, and responsiveness.
- Emphasis will be placed on interdisciplinary collaboration, computational and clinical ethics, and stakeholder engagement for responsible innovation.
