AI development in medical imaging: Institut Curie, member of the consortium AI DReAM led by GE Healthcare
On September 2020, GE Healthcare announced the launch of the AI DReAM project, a consortium that gathers Institut Curie & other research labs and clinical centers, French SMEs, and start-ups to help accelerate the development of AI in medical imaging. Supported with €13 million by the “Investments for the Future” program (PIA), led by the French General Secretariat for Investment and operated by Bpifrance, AI DReAM aims at improving diagnosis accuracy, patient’s care pathway and efficiency of healthcare systems.
The AI DReAM Consortium
Led by GE Healthcare, the AI DReAM consortium brings together Institut Curie, SME Evolucare, Start-ups Therapanacea and Pixyl, the AP-HP, Gustave Roussy and the Foundation Saint Joseph Hospital, who will partner on several clinical case studies to help build the platform and tools to accelerate AI projects.
|AI DReAM’s objectives:|
|Facilitate the creation of standardized and structured databases for the implementation of AI projects.|
|Accelerate the review of remote imaging exams. Enable quality annotation on a large scale while respecting data security.|
|Facilitate the creation of AI algorithms and their integration into medical applications by opening the platform to third-party developments.|
Accelerate market access for new AI applications by providing tools and services for regulatory registration.
At Institut Curie, it is Irène Buvat, head of the Translational Oncology Imaging Laboratory (Institut Curie/Inserm), who will be leading the AI DReAM project, which is proving to be particularly promising and useful for an infinite number of medical applications:
This is a pioneer project because it brings together a set of key players to develop a generic platform that will accelerate the deployment and validation of AI methods for imaging. Instead of focusing on a specific application, it aims to build a whole ecosystem to easily create and validate AI algorithms and thus facilitate their implementation in hospital imaging departments.
Within this project, GE Healthcare will develop an advanced visualization platform with 2D/3D annotation tools to quickly analyze images generated by medical imaging equipment, to enable faster diagnosis and appropriate therapy while ensuring better patient follow-up. "The goal is not only to provide AI methods that will help radiologists interpret images, but also to use AI to better understand the disease from the images", adds Irène Buvat.
A strong need for AI development in medical imaging
Advances in medical imaging constantly increase our knowledge of diseases and their treatments, creating a surge in the amount of data generated. For each patient to benefit from the promises of personalized medicine, new AI-based tools are needed to aggregate, standardize and make sense of this data quickly.
However, the development of AI applications in medical imaging requires access to large amounts of high-quality annotated data, advanced design visualization capabilities and the ability to trace the development and learnings – which can be a challenge for smaller AI players. This is the challenge that the AI DReAM project aims to address, leveraging existing AI know-how and capabilities to accelerate the development of AI solutions and help structure an AI healthcare ecosystem.