Clinical Evaluation of AI for Health: Working Group Collaboration Overview
This document presents updates from the FG-AI4H meeting discussing the clinical evaluation of AI in healthcare. It includes details on collaborations, guidance for best practices, principles evaluation, and special considerations for LMIC settings. The presentation covers the objectives, key participants, and timelines of the Working Group on Clinical Evaluation.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
E N D
Presentation Transcript
FGAI4H-O-038-A01 Berlin, 31 May 02 June 2022 Source: Editors DEL7.4 Title: Updated DEL7.4: Clinical evaluation of AI for health - Att.1: Presentation Discussion Naomi Lee Shubs Upadhyay Eva Weicken E-mail: eva.weicken@hhi.fraunhofer.de Purpose: E-mail: naomi.lee@lancet.com E-mail: shubs.upadhyay@ada.com Contact: Abstract: This PPT contains a presentation of: Status update Working Group on Clinical Evaluation Updates on DEL 7.4
Working Group on Clinical Evaluation Deliverable 7.4 FG-AI4H meeting O , 30 May 02 June 2022
Objective WG-CE Collaboration Collaboration with other expert groups & global stakeholders Guidance for best practice evaluation best practice evaluation & principles evaluation (relevant across all countries) Special consideration LMIC settings LMIC settings (WHO SDGs) Framework for Framework for: clinicians, researchers, developers, regulators, policy makers & patients, public Intended to be used together with other FG principles of FG- -AI4H guidance AI4H guidance
Collaboration 65+ members AbouElkhir Osama, Akogo Darlington, Allen Megan, Alsalamah Shada, Balachandran Pradeep, Bastawrous Andrew, Bathke Arne, Chiavegatto Filho Alexandre, Darkoh Ernest, Ehrenfeld Jesse, Fehr Jana, F rstenau Daniel, Gaudin Robert, Gilbert Stephen, Greaves Felix, Gupta Saurabh, G tter Zdenek, Hatton Grace, Ho Dean, Ibrahim Hussein, Islam Shariful, Jarral Reza, Jeon Jonghong, John Oommen, Kadam Rigveda, Kherif Ferath, Kuku Stephanie, Kurtys Michal, Lap o Lu s, Linder Nina, Loh Irv, Loveys Kate, Magrabi Farah, Mahajan Arnav, Malpani Rohit, Mamun Khondaker, Masud Jakir Hossain Bhuiyan, Matin Rubeta, Matthew Fenech, McCradden Melissa, Menezes Audrey, Murchison Andrew, Murphy Lisa, Nakasi Rose, Oala Luis, Pankova Natalie, Piekut Agata, Porras Lina, Reddy Sandeep, Salim Ally, Schwendicke Falk, Sethi Tavpritesh, Sood Harpreet, Sousa In s, Srivastava Manish, Starlinger Johannes, Wasswa William, Werneck Leite Alixandro Writing group: Carolan Jane, Denniston Alastair, Lee Naomi, Liu Xiao, Karpathakis Kassandra, Upadhyay Shubs, Weicken Eva, Wilkinson Tommy
Timeline Feb + May 22 FG-AI4H meetings N & O
Activities since meeting N AI Auditing Extracted a summarised guidance to be used by auditing teams Added terms relevant for CE to FG-AI4H Common unified terms (TBC) Added feedback from WG-Regulation (Shada Alsalamah)
Deliverable 7.4_Table of content Executive Summary Introduction Model design and suitability Algorithmic validation Clinical validation Deployment and ongoing monitoring Economic evaluation Conclusions and recommendations for future action References posted on collaboration site: Del 7.4/O-038
Conclusions & Recommendations for future action (I) Guidance provides a framework health technologies health technologies that if applied will work towards ensuring AI systems are effective, safe, cost effective, ethical, inclusive, fair are effective, safe, cost effective, ethical, inclusive, fair (urgent global priority) framework for current best current best- -practice evaluation of AI practice evaluation of AI ensuring AI systems This requires input from a broad group of stakeholders range of considerations required broad group of stakeholders to understand the Evaluation must be transparent, results open and accessible transparent, results open and accessible to build trust and enable stakeholders having evidence having evidence to assess the safety, effectiveness and value of AI and its performance build trust
Conclusions & Recommendations for future action (II) Areas of future actions High quality datasets High quality datasets to train & evaluate the performance of health AI systems Clinical Clinical studies studiesare essential (assess whether in silico performance translates into measurable benefit in the real world) Acquisition of relevant relevant health healtheconomic economic data underpin public trust) Much more active role for health healthtechnology addition to regulators data (to support decisions & technologyassessment assessment (HTA) (HTA) in
Next steps & publication plans Share the draft with WHO for review and publication possibly together with WG-RC guidance Discuss implementation of WG-CE guidance in real-world
Thank Thank you you! ! AbouElkhir Osama, Akogo Darlington, Allen Megan, Alsalamah Shada, Balachandran Pradeep, Bastawrous Andrew, Bathke Arne, Chiavegatto Filho Alexandre, Darkoh Ernest, Ehrenfeld Jesse, Fehr Jana, F rstenau Daniel, Gaudin Robert, Gilbert Stephen, Greaves Felix, Gupta Saurabh, G tter Zdenek, Hatton Grace, Ho Dean, Ibrahim Hussein, Islam Shariful, Jarral Reza, Jeon Jonghong, John Oommen, Kadam Rigveda, Kherif Ferath, Kuku Stephanie, Kurtys Michal, Lap o Lu s, Linder Nina, Loh Irv, Loveys Kate, Magrabi Farah, Mahajan Arnav, Malpani Rohit, Mamun Khondaker, Masud Jakir Hossain Bhuiyan, Matin Rubeta, Matthew Fenech, McCradden Melissa, Menezes Audrey, Murchison Andrew, Murphy Lisa, Nakasi Rose, Oala Luis, Pankova Natalie, Piekut Agata, Porras Lina, Reddy Sandeep, Salim Ally, Schwendicke Falk, Sethi Tavpritesh, Sood Harpreet, Sousa In s, Srivastava Manish, Starlinger Johannes, Wasswa William, Werneck Leite Alixandro, Carolan Jane, Denniston Alastair, Lee Naomi, Liu Xiao, Karpathakis Kassandra, Upadhyay Shubs,Weicken Eva, Wilkinson Tommy
