Siemens Healthineers and the European Society of Radiology (ESR), which represents more than 10,000 members worldwide, are launching a collaborative arrangement to promote digitalization within the radiology community.

WHY IT MATTERS
The partnership includes discussions about artificial intelligence, machine learning, neural networks and will include webinars and webcasts for information and discussion with radiologists.

“Digitalization affects all of us, and will fundamentally change our lives and the healthcare industry,” Christoph Zindel, president of diagnostic imaging at Siemens Healthineers, said in a statement. “Companies must actively promote and understand the topic if they are to shape this digital future.”

Zindel said that makes it all the more important to have an effective discussion with the radiology community, so partners can work together to identify effective solutions and promote further dialogue.

The partnership started with a first course on AI, which was held in Barcelona earlier this month as part of the European Congress of Radiology (ECR), where it focused on the digital future of radiology.

Outside the traditional radiology activities of image interpretation, AI is estimated to impact on radiomics, imaging biobanks, clinical decision support systems, structured reporting, and workflow.

ON THE RECORD

“We are very pleased that Siemens Healthineers has not only put the forward-looking themes of radiology, digitalization, and artificial intelligence at the heart of its corporate interests, but is also working with ESR to determine how we can best make these themes accessible to our members,” ESR chair for the board of directors Lorenzo Derchi said in a statement.

WHAT ELSE TO KNOW

In the future, Siemens Healthineers, which is currently developing digital health services and enterprise services, will regularly publish texts and interviews on the subject of AI on ESR’s recently introduced blog.

An ESR whitepaper published earlier this month said the key factor of AI performance is training with big and high-quality data to avoid overfitting and underfitting. The paper noted the best solution for reducing overfitting is to obtain more training data.

“Multiple rounds of training and testing on different datasets may be performed, gradually improving network performance, and permitting assessment of the accuracy and generalizability of the algorithm, before the algorithm is released for general use,” the paper said.

Another solution would be the so-called “data augmentation”, which means modifying the training data by adding some variability so that the model will not see the exact same inputs from the training data set during the training iterations.

The paper also noted laws of robotics could be applied to radiology where the “robot” is the “AI medical imaging software,” and explained that if AI is used in clinical practice, the main medico-legal issue that then arises is “who is responsible for the diagnosis.”

Nathan Eddy is a healthcare and technology freelancer based in Berlin.

Email the writer: nathaneddy@gmail.com

Twitter: @dropdeaded209 

Healthcare IT News is a HIMSS Media publication.