Presented By O’Reilly and Intel AI
Put AI to work
Sep 4-5, 2018: Training
Sep 5-7, 2018: Tutorials & Conference
San Francisco, CA

Building intelligent mobile applications in healthcare

Xiaoyong Zhu (Microsoft), Wilson Lee (CLOUD AI) (Microsoft), Ivan Tarapov (Microsoft), Mazen Zawaideh (University of Washington Medical Center)
1:30pm-5:00pm Wednesday, September 5, 2018
Implementing AI, Interacting with AI, Models and Methods
 Location: Union Square 22
Secondary topics:  Computer Vision, Edge computing and Hardware, Health and Medicine

Who is this presentation for?

  • Data scientists and AI application developers

Prerequisite knowledge

  • A basic understanding of AI concepts and methods (unsupervised learning, CNNs, image classification, etc.)

Materials or downloads needed in advance

  • A laptop

What you'll learn

  • Explore skills to use for basic medical image classification problems such as chest X-rays, where the image set is usually highly imbalanced
  • Learn how to reproduce and adapt this work for your own use cases

Description

Artificial intelligence has emerged as one of the most disruptive forces behind the digital transformation that is currently revolutionizing the way we live and work. In the field of healthcare and medicine, AI is accelerating change and empowering physicians to achieve more. Microsoft’s Health NExT project is looking at innovative approaches to fuse research, AI, and industry expertise to enable a new wave of healthcare innovations and empower every developer to innovate and accelerate the development of intelligent apps. AI-powered experiences augment human capabilities and can potentially allow us to lead healthier lives.

Take the task of detecting diseases from chest X-ray images, for instance. This is a challenging task that requires consultation with an expert radiologist. However, two-thirds of the world’s population lacks access to trained radiologists, even when imaging equipment is readily available. The lack of image interpretation by experts may lead to delayed diagnosis and could potentially increase morbidity or mortality rates for treatable diseases like pneumonia.

Xiaoyong Zhu, Gheorghe Iordanescu, Wilson Lee, and Ivan Tarapov walk you through a working example that helps clinicians in areas with less access to radiologists identify possible lung diseases, inspired by the CheXNet work done by Stanford University ML Group, and explain how data scientists can leverage the Microsoft AI platform and open source deep learning frameworks like Keras or PyTorch to build an intelligent disease prediction deep learning model.

Topics include:

  • Introduction to medical image classification
  • State-of-art image classification algorithms and architectures and how they can be applied to medical image classification problem
  • Training techniques on medical image classification
  • Operationalizing the model in the cloud and on devices
  • Building intelligent applications for iOS, Android, and Windows, consuming a local model without requiring data to leave the device
Photo of Xiaoyong Zhu

Xiaoyong Zhu

Microsoft

Xiaoyong Zhu is a senior data scientist at Microsoft, where he focuses on distributed machine learning and its applications.

Photo of Wilson Lee (CLOUD AI)

Wilson Lee (CLOUD AI)

Microsoft

Wilson Lee is a senior software engineer in the AI CTO Office at Microsoft, where he works with teams to envision and innovate new end-to-end experiences that illustrate what is possible with the present and the future of Microsoft AI. Wilson strongly believes that story-driven innovation paired with great software architecture can create real change in the world, making the impossible possible. He holds a bachelor of computer science from the University of Waterloo.

Photo of Ivan Tarapov

Ivan Tarapov

Microsoft

Ivan Tarapov is a senior program manager at Microsoft Research working on Project InnerEye, which is focused on using state-of-the-art machine learning to build a platform that will transform the way doctors interact with medical images. Previously, Ivan worked for a global software consultancy, contributing to high-risk medical software projects such as implantable pacemakers, external defibrillators, and insulin pumps in the areas of software design and development, engineering processes, and software architecture. Ivan is a certified Scrum Product Owner and excels at execution; he has a proven track record of building high-performing Agile software development teams and driving them to build great products. He holds a master’s degree in applied mathematics.

Photo of Mazen Zawaideh

Mazen Zawaideh

University of Washington Medical Center

Mazen Zawaideh is Chief Radiology Resident at the University of Washington. He is also co-founder and co-instructor of imagedeep.io, an intensive course for radiology residents designed to bridge the gap between medical imaging and AI education. He received his BS and MD from UCSD.