Supercharging MR with AI

Karley M. Yoder

VP & GM, Edison AI GE Healthcare Digital

As we enter a new era of healthcare defined by the emergence of AI, it’s imperative that we rethink not just the technologies that make up our AI solutions, but how we implement these tools. That means developing technology fully integrated with AI from the beginning, embedded into systems and software so it’s a seamless clinical experience.

At GE Healthcare, we’re aiming to become the center of gravity for this AI innovation. We’re evolving our mindset around AI as a set of technologies and productivity tools that have the power to help you transform patient outcomes. We’re powering breakthroughs using AI that can improve workflow so you can maximize productivity and save lives in the moments that matter.

We’re also creating an ecosystem of technical and clinical knowledge, starting with the clinicians who use these systems every day. The best way to understand how to reduce costs, increase access and improve quality are to talk to the clinicians who live this reality. That’s why GE partners with clinical experts at academic institutions and leading healthcare facilities around the world. These connections help us live in your workflows, understand your pain points and then apply this powerful technology to drive better outcomes.

These partnerships have powered advancements like our Edison intelligence platform. Edison helps accelerate the development and adoption of AI technology and empower providers to deliver faster, more precise care. Embedded within existing workflows, Edison applications can integrate and assimilate data from disparate sources and apply analytics or advanced algorithms to transform the data and generate clinical, operational and financial insights.

For example, AIR x^TM, which stands for Artificial Intelligence Prescription, is built on Edison as an AI-based, automated workflow tool. It uses deep learning (DL) algorithms built right into the MR technologist’s workflow to automatically identify anatomical structures to “prescribe” slices for neurological exams, delivering consistent results to facilitate quantitative analysis.

Another example is a new MR image reconstruction method, AIR^TM Recon DL^‡, representing an evolution in digital image reconstruction for reducing traditional image artifacts. It is a DL- based, convolutional neural network to intelligently reconstruct a final MR image with a high signal-to-noise ratio and image sharpness. AIR^TM Recon DL is not a post-processing technique, rather it is embedded directly in the reconstruction pipeline, where the neural network model is applied to acquired input data to remove noise and ringing artifacts prior to final image formation. By operating on raw data within the online reconstruction pipeline, AIR^TMRecon DL benefits from access to the full set of acquired source data to generate an image, compared to after DICOM image conversion where important information may be lost.

To accelerate our AI journey, we have developed strong strategic alliances with several leading AI-related global technology companies. With Amazon Web Services, leveraging SageMaker data science tools helps our GE data scientists build best-in-class AI solutions. The alliances with Intel^®, Hewlett Packard Enterprise and NVIDIA^® help bring the most sophisticated AI technology to our 500,000 imaging devices globally and accelerate the speed at which healthcare data can be processed, allowing clinicians to review critical findings at the point of care to accelerate patient diagnosis.

We believe AI should feel invisible while it works silently in the background to make life easier for clinicians. It’s like a driverless car — we’re not excited because it’s built with AI. We’re excited because it gives us a more relaxing commute, improves safety on the roads and allows us to do more in a day.

That’s our vision for AI in healthcare. It’s why we strive to access and provide unbiased data science that’s integrated straight into your workflow, while delivering a greater impact in patient care pathways.

Karley M. Yoder

‡510(k) pending at the US FDA. Not yet CE marked. Not available for sale.