In recent years , a subfield of AI , deep learning , has delivered a significant increase in accuracy by using new learning approaches , specialized hardware , and significantly larger datasets to find more complex and subtle patterns within the data .” tion : “ Because AI can identify meaningful relationships in raw data , it can support diagnostic , treatment and prediction outcomes in many medical situations . It allows medical professionals to embrace the proactive management of disease onset . Additionally , predictions are possible for identifying risk factors and drivers for each patient to help target healthcare interventions for better outcomes ,” according to Secinaro , et al . ( 2021 ).
Clinical decision-making is a key area for AI application , but proceeding with caution seems to be the presiding sentiment for now . Jiang , et al . say that AI can help physicians make better clinical decisions or even replace human judgment in healthcare-specific functional areas , with Bennett and Hauser noting that algorithms can benefit clinical decisions by accelerating the process and the amount of care provided , positively impacting the cost of healthcare delivery . The enthusiasm for AI should be balanced with measures to ensure bio-ethical considerations have been weighed fully , they add .
Infection preventionists ( IPs ) and clinicians in general process significant amounts of data throughout their workday , and some experts believe that machine learning is the optimal way to harness and leverage this data , especially with the goal of detection of infectious organisms that can lead to outbreaks .
As Fitzpatrick , et al . ( 2020 ) explain , “ AI uses mathematical tools , machine learning , to iteratively learn patterns within training data and when these patterns are found in new data , the AI translates this into a decision , for example , cancer versus not cancer . In recent years , a subfield of AI , deep learning , has delivered a significant increase in accuracy by using new learning approaches , specialized hardware , and significantly larger datasets to find more complex and subtle patterns within the data .”
The researchers confirm further , “ As healthcare IT systems produce vast quantities of data from disparate sources and become increasingly integrated , AI systems will be able to detect patterns in the data accelerating the detection of outbreaks and providing richer datasets for subsequent analysis . AI can support the case for system change by identifying the cost of inaction , modeling solutions by simulating the behavior of different types of agents within a complex system and supporting change by gathering data and producing analytics . Social graph analysis can identify influencers of hand hygiene programs and explore patient safety culture . In IP & C education and training programs , AI-based simulations can provide a bridge to authentic experience that does not compromise patient safety and provide the repeated cycles of objective evaluation and feedback that are key to the learning process .”
In a commentary , Parums ( 2023 ) observes , “ Local hospital-based infection outbreaks are controlled by tracing and controlling the identified transmission routes . Hospitals use local and national guidelines and infection control procedures to detect local infection outbreaks . Localized hospital-based outbreak detection may require a manual review of patient notes and contact tracing , which can be labor-intensive , and early detection and control may not be possible . Whole-genome surveillance sequences can identify genetic similarities in infectious organisms but do not identify the source and spread of infection .”
Implications for Clinicians , and IPs Specifically Will medical technology like ML / AI “ de-skill ” or even replace clinicians ?
As Secinaro , et al . ( 2021 ) observe , “ Studies have shown that healthcare personnel are progressively being exposed to technology for different purposes , such as collecting patient records or diagnosis …. [ Research ] indicates that the excessive use of technology could hinder doctors ’ skills and clinical procedures ’ expansion . Among the main issues arising from the literature is the possible de-skilling of healthcare staff due to reduced autonomy in decision-making concerning patients .”
Marra , et al . ( 2023 ) state that while AI can revolutionize infection prevention and healthcare epidemiology through improved surveillance and heightened vigilance for adverse events , but there are some concerns about how AI will impact the healthcare workforce .
“ AI can also facilitate predictive modeling of healthcare-associated infections and outbreaks , enabling hospitals to prioritize infection prevention efforts and allocate resources effectively ,” they emphasize . “ The integration of AI technologies in infection prevention and healthcare epidemiology has the potential to revolutionize the way healthcare is delivered . However , this paradigm shift could also raise concerns about the potential displacement or alteration of roles traditionally performed by microbiology technicians , infection prevention and control practitioners , and antimicrobial stewardship clinicians . As AI is increasingly employed to automate laboratory result analysis , predict infection patterns , or recommend treatment strategies , it becomes crucial to address the ethical and workforce implications associated with these changes . Although the displacement of technical duties is inevitable , this may free up time for tasks involving strategic planning ( such as identification and evaluation of novel antimicrobial stewardship and infection prevention initiatives ) and human interaction ( such as participating in handshake stewardship ). Further , all these applications are in full alignment with the goals of infection prevention , that is , to improve patient outcomes by preventing the spread of infections and optimize healthcare practices .”
Experts acknowledge that HAI surveillance is a complex skill , demanding specialized knowledge and resources , and two IP & C experts say that AI can potentially transform this task that is an integral part of infection preventionists ’ workday .
In a recent research project , Timothy Wiemken , PhD , MPH , an associate professor in the division of infectious diseases , allergy , and immunology at Saint Louis University and lead author of the paper , and Ruth Carrico , PhD , DNP , from the University of Louisville School of Medicine , investigated the use of AI , particularly generative large language models , to improve HAI surveillance . They assessed two AI agents , OpenAI ’ s chatGPT plus ( GPT-4 ) and a Mixtral 8 × 7b-based local model , for their ability to identify central line-associated bloodstream infections ( CLABSIs ) and catheter-associated urinary tract infections ( CAUTIs ) from six National Health Care Safety Network ( NHSN ) training scenarios . The complexity of these scenarios was analyzed , and responses were matched against expert opinions . Specifically , descriptions of six fictional patient scenarios with varying levels of complexity were presented to the AI tools , which were then asked whether the descriptions represented a CLABSI or a
14 • www . healthcarehygienemagazine . com • june 2024