A new AI system could help hospitals keep their expensive medical equipment healthy by scanning for problems before they become expensive to fix.
Data analytics company Glassbeam has announced a series of new AI applications that will help healthcare providers to identify parts failures in hospital MRI and CT scanners.
Using the new systems, doctors and other medical professionals can tap into machine learning algorithms to ensure that expensive, life-saving systems are kept in constant working order. The applications also offer cloud-based dashboards and alerts to transform the equipment maintenance process.
Glassbeam said it wants to help healthcare organisations “deliver better and more efficient patient care”.
It can be an expensive and lengthy process for medical technicians to take systems offline and repair them internally. Meanwhile, predicting or planning for failures can be a challenge in environments where investment is tight, time is critical, and lives are at stake.
Corey Holtman, president at Gateway Diagnostic Imaging, said medical imaging systems are becoming increasingly complex. As a result, AI and IoT systems could revolutionise the healthcare sector, he said.
“The management of medical machines such as MRI and CT scanners has taken on a new level of complexity in recent years, due in part to the increased sophistication of equipment and the ever-increasing requirements for compliance, safety, reliability, and accuracy,” he said.
“Predicting machine health and utilisation patterns – with help from the latest techniques in artificial intelligence and machine learning – is the next frontier to improve operations in clinical engineering functions.”
Changing the industry
Glassbeam plans to develop the new systems in phases. The first of these will focus on CT scanners, which are used to create cross-sectional views inside the human body without the need for surgery.
CT systems can cost up to $2.5 million apiece, and when they malfunction or need parts replacing, healthcare providers risk running up six-figure bills, while taking life-saving systems offline.
Glassbeam said that hospitals and clinics are missing out on the predictive capabilities offered by machine learning and big data analytics.
Its new system can warn professionals about problems a week before they might occur. This “can alert clinical engineering staff to become proactive in avoiding unplanned downtime, saving costs, and averting patient re-scheduling at the last minute”, said the company.
Counting the benefits
“The parts replacement industry for the global installed base of medical imaging equipment is slated to be a $3.6 billion market in 2020,” said Puneet Pandit, co-founder and CEO of Glassbeam.
“With AI and ML applications based on analysing millions of sensor readings captured in the Glassbeam cloud each day, even with a modest 10 percent savings we are ready to make a significant dent on the underlying inefficiencies.”
Lise Getoor, professor of computer science at the University of California in Santa Cruz, praised the work being done by Glassbeam, which is a technology partner at UCSC’s D3 (Data, Discovery, Decisions) Center.
“The number of signals coming from connected machines in the IoT market surpassed the ability of humans to keep track of them years ago,” she said.
“I’m excited to see Glassbeam taking a leadership role in leveraging artificial intelligence to change the rules of the game for the healthcare market.”
Internet of Business says
AI’s predictive capabilities, together with technologies such as big data analytics, digital twins, and enterprise asset management (EAM) systems, could be transformative across many sectors as the IoT spreads.
CERN’s Large Hadron Collider, the largest machine ever built, is perhaps the leading example of the technology’s potential. Every single component in the CERN campus is logged in an EAM system as a digital twin, and predictive analytics help engineers predict failures and plan downtime for essential maintenance.
Just as important, the system tells them exactly where the problem lies: an important factor in large, complex systems. With lives at stake as well as big science, these connected technologies have significant potential.