Use cases and implementation strategies for artificial intelligence in radiology

Charles E. Kahn, Jr., MD, MS, editor of the journal RSNA Radiology: Artificial Intelligence, and Professor and Associate Chair in Radiology at the University of Pennsylvania Perelman School of Medicine. He was heavily involved in radiology informatics and witnessed the development of radiology towards deeper integration with artificial intelligence (AI).

Kahn explains that it takes a lot of work to integrate AI into radiological systems. He also said that as the US faces a growing shortage of radiologists, the role of AI is becoming more important, and the technology can help radiologists do more and improve patient care.

“Every time someone comes in and asks to install an AI application in the radiology department, it means someone has to do the legal agreements and all the contracts, but then you have to connect them to your systems,” Kahn said.

This ideally includes the connection to EMR, PACS and other systems used in radiology. For this reason, several vendors are opting for an app store concept, where a single vendor could serve as a gatekeeper for easy integration of specific AI into an existing PACS system architecture.

“For departments that want to start researching these tools, it’s an expensive endeavor and requires quite a bit of resources, not just in terms of cash to purchase or license the system, but also in terms of IT support for building and maintaining the system connections,” explained Kahn.

Another question that radiology departments need to ask is the reason for adopting a particular AI algorithm. Use cases proposed for AI include a way to scale up screening programs or augmented first-pass interpretations of images in rural hospitals and underserved and resource-poor communities. It was suggested a few years ago that AI could replace radiologists, but that seems decades in the future, if ever, Kahn said. Instead, there is a growing shortage of radiologists, and AI can play a role in assisting radiologists so they can focus on cases with suspected disease or more complex cases.

Kahn also said that AI could play a key role in addressing health disparities in the years to come.

“At some level we need to find ways to provide care that is cost effective, reaches all the people we need to reach and provides equitable healthcare, and the hope is that we can use AI to extend that reach.” to expand what we do and improve the quality of it,” Kahn said.

Related content on AI in Radiology:

VIDEO: Assessing AI in Radiology and Understanding Programmatic Bias – Interview with Charles E. Kahn, Jr., MD

VIDEO: 6 Key Trends in PACS and Radiology Informatics Observed by KLAS – Interview with Monique Rasband, VP Imaging, KLAS

Artificial intelligence in radiology: friend, not foe, say experts concerned about students’ perceptions of AI

VIDEO: 9 key areas where AI is being implemented in healthcare – Interview with Julius Bogdan, HIMSS

VIDEO: Where do we stand with the introduction of AI in radiology? — Interview with Bibb Allen, MD

VIDEO: Radiology AI Validation Monitoring to Ensure Accuracy – Interview with Bibb Allen, MD

Radiologists can reclaim an hour every day with AI support

VIDEO: Radiology AI Overview – Keith J. Dreyer, DO, CSO, ACR Data Science Institute

VIDEO: Artificial Intelligence in Radiology Market Segmentation by Function – Interview with Keith J. Dreyer, DO

VIDEO: Where will radiology AI be in 5 years? — Interview with Keith J. Dreyer, DO

What do radiologists really think of the introduction of AI? New data offers insights

Less experienced radiologists benefit from deep learning models when searching for intracranial aneurysms

Legal consequences to consider when integrating AI into daily radiological practice

Source link