SCCM Resource Library-The Role of AI/ML in Prediction Modeling

Video Transcription

Video Summary

In the webcast titled "The Role of AI in Machine Learning and Prediction Modeling," Megan Zingler, a clinical pharmacy specialist, moderates a discussion focusing on the application and future of AI in critical care, particularly in prediction modeling. The webcast offers insights into the strengths and limitations of traditional regression-based methods compared to AI and machine learning approaches.<br /><br />Speakers Andrea Sikora and Brian Murray from the Universities of Georgia and Colorado, respectively, elucidate three key objectives:<br />1. Reviewing the strengths and limitations of traditional medicine-based modeling.<br />2. Discussing prediction modeling approaches based on machine learning.<br />3. Exploring AI algorithms for intensive care unit (ICU) event prediction.<br /><br />The presenters illustrate the utility of AI through examples such as sepsis early warning systems and the Epic Deterioration Index, emphasizing how AI can predict critical events like ICU transfers and patient deterioration. They also highlight the importance of external validation of models and ensuring they are thoroughly tested against established clinical practices.<br /><br />Key principles in evaluating AI performance are explained, including model sensitivity, specificity, and the area under the receiver operating characteristic (AUROC). Additionally, the importance of understanding and handling large, complex data sets typically found in ICU settings is underscored, along with the necessity of interdisciplinary collaboration involving computer scientists, biostatisticians, and clinicians.<br /><br />The webcast concludes with a discussion on the challenges of implementing AI, such as data quality and interpretability, and the critical need for implementation science to study how AI tools impact patient care in real-world settings. Questions from the audience further engaged with topics like data preprocessing, multidisciplinary teamwork, and causal inference.

Asset Subtitle

Professional Development and Education, 2024

Asset Caption

Review recent computing trends with artificial intelligence (AI), machine learning (ML), and natural language processing as they relate to research, education, and clinical practice in the intensive care unit. This webcast explores how to appraise this technology, the prose and cons of AI, the potential for biased algorithms based on training data, appropriate critiques of AI/ML methods, and future methodologic advances.

Learning Objectives
Review strengths and limitations of traditional regression-based prediction modeling
Discuss advantages and key concepts of assessing prediction modeling approaches based on machine learning
Explore use cases of artificial intelligence algorithms for intensive care unit event prediction

Meta Tag

Content Type Webcast

Knowledge Area Professional Development and Education

Membership Level Professional

Membership Level Select

Tag Innovation

Tag Professional Development

Year 2024

Keywords

Webcast

Professional Development and Education

Innovation

Professional Development

2024

Professional

Select

AI in healthcare

prediction modeling

machine learning

critical care

sepsis early warning

Epic Deterioration Index

model validation

ICU event prediction

interdisciplinary collaboration

implementation science

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.