Reducing hospital-acquired infections with artificial intelligence
SAS helps clinicians obtain reliable insights to promote patient well-being.
Improved patient safety
via AI technologies
Region of Southern Denmark achieved this using • SAS® for Machine Learning and Deep Learning • SAS® Visual Text Analytics
Hospitals in the Region of Southern Denmark aim to increase patient safety using analytics and AI solutions from SAS
The Region of Southern Denmark, with help from SAS, has become the first place in the world to implement a complete system for monitoring hospital-acquired infections. Professor Jens Kjølseth Møller at Lillebaelt Hospital is the brain behind the new system, which is made possible by SAS® Analytics. Kjølseth Møller expects the system to reduce the number of infections during hospitalization by one-third, significantly increasing patient safety.
“It is unsatisfying that patients admitted to Danish hospitals are at risk of further illness,” says Peder Jest, Medical Director at Odense University Hospital. “The work of providing a high degree of patient safety and good infection hygiene is, therefore, a key focus area for the Region of Southern Denmark. With SAS, we now have the ability to monitor and predict the risk of hospital-acquired infections at a patient level.”
The challenge: An estimated one in 10 patients acquires an infection while hospitalized, and over 3,000 patients in Denmark die per year because of their infections. But until now, no one knew the actual infection rates, or exactly when and where to make a precise effort to prevent them. Now all hospitals in Southern Denmark have access to the new monitoring system that provides a complete picture of hospital-acquired infections at a patient level.
The system is the result of pioneering Danish work that connects many years of clinical experience with modern technology. Danish hospitals found it unacceptable that admitted patients were at risk of further illness. The work of providing a high level of patient safety and good infection hygiene became key focus areas.
“Now we have a tool that can monitor hospital-acquired infections. With that we can make sure – and even make it transparent and well documented – that we do everything we can to prevent these unwanted infections,” Jest says. “This means that our clinical managers are enabled to monitor their efforts and create better results because the system tells them where to look. This is a new era in the work of reducing hospital-acquired infections.”
SAS presents the information in a way that both clinicians and administrators can understand and act on … The knowledge we receive about patients today is the knowledge that will help prevent infections for patients tomorrow.Jens Kjølseth Møller Professor Lillebaelt Hospital
The first complete system for monitoring infections
Kjølseth Møller had earlier developed a system that could present an overview of the number of infections. This system used local data from one hospital only in the region. But it was not complete, because it couldn’t include unstructured clinical information on infections recorded in clinician notes by doctors and nurses.
The Region of Southern Denmark decided to work with SAS primarily for two reasons: great experience with high-quality artificial intelligence (AI) technologies and deep insights into the health care sector.
The technology dictates rules and clinical terms. These are the so-called “triggers” that can identify signs of infection in patients. It is the first time worldwide that a system based on AI has provided a complete overview of hospital-acquired infections.
In the future, all patients admitted to hospitals in Southern Denmark will be scored for their risk of developing a urinary tract infection while admitted to the hospital. This will enable doctors and nurses to respond in real time and prevent hospital infections. The risk models are developed with AI based on 284,000 previous patient cases in the region.
“We know how many infections there have been for a specific period in a particular department,” Kjølseth Møller says. “We changed our work procedures from handling much of the data ourselves to using a data management solution from SAS, which is accessible by both clinicians and administrators. We deliver raw data to the SAS system, and SAS presents the information in a way that they can understand and act on.”
From data management to model development and deployment, everyone works in the same integrated environment. Within SAS for Machine Learning and Deep Learning, AI helps predict which patients have an increased risk of developing hospital-acquired infections during hospitalization. The infection-monitoring overview is displayed through the region's management information portal, where doctors and departmental managers are accustomed to retrieving other information.
Region of Southern Denmark – Facts & Figures
284,000
previous patient cases were used to develop risk models
Lillebaelt Hospital
is one of five hospitals in Southern Denmark
Public health care
is the region’s main responsibility
More hospitals want to monitor their hospital-acquired infection rates
In 2018, clinicians in Southern Denmark hospitals could, for the first time, draw reports of all infections. SAS allows hospital and departmental management to monitor the development of hospital-acquired infections with overview and development reports over time at the hospital, departmental and sectional levels.
“The knowledge we receive about patients today is the knowledge that will help prevent infections for patients tomorrow,” Kjølseth Møller says. “It will help us also work much more actively with what we call intervention, where we try to change routines to see if we can do better than what we do now.”
本文中所展示的成果取决于文中所述的特定场景、业务模式、数据输入和计算环境。每位SAS客户的体验都因其业务和技术特性而不同，请勿将本文观点视为通用观点。实际的成本节约、成果和效果最终取决于每位客户的实际配置和条件。SAS不保证每位客户都能取得本文类似的成果。SAS仅对SAS的产品和服务提供保证，请参阅SAS的产品服务质保条款，本文中提及的内容不能视为质保条款。客户可以按照合约商定的条款分享SAS软件实施项目的成功案例，相应的品牌和产品名称归属相应的公司所有。