Скачать с ютуб AI and Bias In Healthcare в хорошем качестве

AI and Bias In Healthcare

Continuing our series on AI in Healthcare; In our last two events, we saw examples of how AI helped address challenges in Healthcare. We covered numerous applications, from using NLP to help doctors and better communicate to innovative new diagnostic tools that automate tasks and reduce costs. However, an issue that popped up numerous times during our discussion was Bias in AI. Bias in AI can have many sources- the data, the labeling, even the algorithm. It can widen inequalities when left unchecked and can even be fatal in the context of Healthcare. (See this for example https://www.scientificamerican.com/ar...) In this panel discussion, we discussed some of the cases where Bias in AI is observed, some of Bias's sources, and what we can do to mitigate, or if not eliminate, the bias. Panelists Andrew Fairless, Ph.D., Principal Data Scientist at Geneia:   / andrew-fairless-7a32593   Event: AI and Bias in Healthcare Date: Tuesday, December 15th, 2020 @ 5 PM PST Agenda 5:00-5:10 Introduction 5:10-5:30 Speaker 1 5:30-5:50 Speaker 1 5:50-6:00 Q&A Papers: Here are some of the papers mentioned by Dr. Fairless. https://arxiv.org/abs/1808.00023 The Measure and Mismeasure of Fairness: A Critical Review of Fair Machine Learning Sam Corbett-Davies, Sharad Goel https://pubmed.ncbi.nlm.nih.gov/31649... Science. 2019 Oct 25;366(6464):447-453. doi: 10.1126/science.aax2342. PMID: 31649194 Dissecting racial bias in an algorithm used to manage the health of populations Ziad Obermeyer, Brian Powers, Christine Vogeli, Sendhil Mullainathan