By: Sharon Mah
Faculty of Health Sciences (丁香园AV FHS) assistant professor, Jean-Christophe B茅lisle-Pipon is co-leading a team with professeure titulaire from Universit茅 de Montr茅al that will contribute to developing new knowledge, resources and guidance and establishing standards for 鈥榚thical trustworthiness鈥� for two projects within Bridge to Artificial Intelligence (Bridge2AI): Voice as a Biomarker of Health (鈥榁oice鈥�), and Cell Maps for Artificial Intelligence (CM4AI).
that will invest funds over a four year period to accelerate the widespread use of artificial intelligence (AI) by the biomedical and behavioural research communities.
鈥淰oice-data and human cell maps research are both emerging research fields, and at present, there is little to no guidance with respect to the ethical, legal and social implications of this work,鈥� says B茅lisle-Pipon. 鈥淲ithin the Bridge2AI collaboration, there will be an opportunity for Dr. Ravitsky and myself to identify, anticipate, address, and provide guidance to other researchers creating datasets that will be compiled for use in AI applications. Our work will anticipate and address ethical challenges such as inclusion, diversity, privacy, consent, data ownership and sharing, AI transparency, and potential bias.鈥�
鈥淲e intend to develop this module by using the approach of ethics inquiry through a continuum, starting from data generation and AI research and development, continuing into clinical adoption of the datasets, and extending to downstream patient health decisions and outcomes,鈥� says Ravitsky.
During the first year of their work on the ethics module for Bridge2AI, B茅lisle-Pipon and Ravitsky will establish the groundwork for voice-specific and cell maps-specific bioethical conceptual frameworks that will be informed by patient and stakeholder preferences and values. This investigation will be anchored in consensus-building, and will involve extensive collaboration with leading researchers in approximately 20 institutions across North America involved in Voice and CM4AI.
鈥淭his is an extraordinary opportunity to root this important health AI research in a strong ethical grounding so that it doesn鈥檛 perpetuate health inequities or ethical problems that may occur during data collection, analysis, or deployment,鈥� observes B茅lisle-Pipon. 鈥淎I has incredible potential to address and possibly solve some of our most pressing health challenges. It鈥檚 essential that diversity is integrated into the development of these tools and resources so that we can increase representation of historically marginalized people in the data. This will reduce biases and ultimately improve the effectiveness of the technologies that will be derived from this research.鈥�
Each project within this initiative brings together team members from diverse disciplines and backgrounds from across the US and Canada to generate ethically-sourced tools, resources and richly detailed state-of-the-art data sets that are responsive to AI approaches. The Voice as a Biomarker for Health module is being co-led by assistant professor Yael Bensoussan with the University of South Florida, and professor Olivier Elemento of Weil Cornell Medicine. Cell Maps for Artificial Intelligence is being headed by professor Trey Ideker of UC San Diego.
To learn more about the Bridge2AI program, read 鈥�,鈥� a blog for the National Library of Medicine, and watch .