Artificial and Human Intelligence in Healthcare
Who is this presentation for?Researchers, Industry looking at interaction of AI systems in healthcare,
With the fundamental breakthroughs in Artificial Intelligence and the significant increase of digital healthcare data, there has been enormous interest in AI for healthcare applications. One rapidly developing area is the use of deep neural networks for medical imaging, with applications ranging from diagnosing chest X-rays to early detection of Alzheimer’s to identifying cancer in pathology slides.
Despite this variety of applications, there remain some crucial unanswered questions. On the methods side, there has been a premature convergence on a specific model development strategy: deep neural networks are first trained on natural image data, and then finetuned (transfered) to work on the medical data. In the first part of the talk, I explore this process, showing that contrary to conventional wisdom, this standard method of model development is not guaranteed to provide the benefits it is believed to, and suggest simple and effective alternate methodologies.
On the applications side, there has been little exploration of the interaction of these medical AI algorithms with human experts, with existing literature typically evaluating the algorithm in isolation and the human experts in isolation — vastly different to a realistic deployment scenario. In the second part of the talk, I show that the essential question on the role of human experts provides new, crucial, prediction problems to study, and significant benefits, through effective combination of artificial and human intelligence.
Prerequisite knowledgeA little knowledge of developing machine learning models (specifically deep neural networks) would be helpful.
What you'll learnAudience will learn about the standard methods for developing algorithms (deep neural networks) for medical imaging applications and ways to improve on these, as well as places where conventional beliefs might be misleading. They will also learn about typical possible deployment scenarios for these technologies, and the kinds of challenges and benefits that arise through interaction with human experts.
Cornell University/Google Brain
Maithra Raghu is a PhD Candidate at Cornell University and a Research Scientist at Google Brain. Her research focuses on developing tools to understand deep neural networks and using these insights in healthcare applications. She has been named as one of the Forbes 30 Under 30 in Science, and an EECS Rising Star by MIT.
Leave a Comment or Question
Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?
Join the conversation here (requires login)
Diversity and Inclusion Sponsor
For conference registration information and customer service
For more information on community discounts and trade opportunities with O’Reilly conferences
For information on exhibiting or sponsoring a conference
View a complete list of O'Reilly AI contacts