15 May 2020

One of the world's most powerful systems for artificial intelligence (AI) computation has landed at Linköping University. The system will serve as a national resource for projects within the "Analytic imaging diagnostics arena" (AIDA). The NVIDIA system is the second of its kind in Sweden and a unique resource for AI research in medical imaging.

Unveiling of the DGX-2
Unveiling of the NVIDIA DGX-2 AI system during the AIDA Days at CMIV in March 2020.

-This is a unique type of resource that we, in collaboration with NVIDIA and Proact, have shaped according to the needs of our research groups, says Claes Lundström, Professor at Linköping University and Arena Leader for AIDA.

 

The system, named NVIDIA DGX-2, has 16 GPUs, which means that it can receive and process extensive volumes of data in a short period of time. This is a prerequisite for large-scale computation in AI projects, especially in medical imaging as extremely large amounts of data are often generated. This investment is not possible for individual research groups. Through AIDA they will now have the opportunity to realize highly challenging projects.

 

AIDA has its physical base at the Center for Medical Image Science and Visualization, CMIV, at Linköping University and this is also where the GPU system has been installed.

 

-CMIV has many years of experience in bringing medicine and technology together to create patient benefits. The technical competence to administer a system of this size can also be found there, explains Claes Lundström.

 

The projects within AIDA all develop AI solutions that are intended to become helpful tools in healthcare. Transferring technological innovations to clinically useful solutions is a challenge and a close collaboration with healthcare is a prerequisite for success. AIDA's task is to facilitate this process.

 

Using AI, mathematical algorithms can be trained to analyze the information in a medical image. For example, it can be about automatically identifying cancer cells in a microscopy image or identifying and marking organs in an X-ray image.

 

During the past winter and spring, AIDA held a series of seminars on NVIDIA DGX-2 ending with a more practical Hackathon prior to the opening of the system. Proact has delivered the system and NVIDIA has been involved throughout the process and contributed with their AI expertise.

 

The system will function as a timeshare condo where AIDA members can book time for their calculations. There is a great interest in the system, shown among other things by a significant increase in the number of project applications.

Contact

Latest news from LiU

Two men in white lab coats with a computer in a lab.

Improving Alphafold to predict very large proteins

The AI tool Alphafold has been improved so that it can now predict the shape of very large and complex protein structures. Linköping University researchers have also succeeded in integrating experimental data into the tool.

Rinata Kazak looking down at her jacket.

LiU researcher ahead of UN climate summit - "I’m optimistic"

Azerbaijan will host the International Climate Summit this year. Although the country is heavily dependent on its oil production, holding the meeting there could actually be an advantage, according to Rinata Kazak, who will represent LiU.

Two women at a table talking.

Working together for a less biased world

In what ways does modern technology risk giving us a distorted picture of the world? Seeking answers, researchers at Tema Genus are working with colleagues in computer science.