How artificial intelligence could improve speed and accuracy of response to infectious disease outbreaks in hospitals, and even prevent them
Reports and ProceedingsPlease mention the European Congress of Clinical Microbiology and Infectious Diseases (ECCMID 2024, Barcelona, 27-30 April) if using this material
A new research review to be given at a pre-congress day for this year’s European Congress of Clinical Microbiology and Infectious Diseases (ECCMID 2024) will highlight the potential artificial intelligence (AI) has to improve the speed and accuracy of investigations into infectious disease outbreaks in hospitals, and potentially provide real time information to stop or prevent them. The talk will be by Dr Jonas Marschall, Division of Infectious Diseases, Washington University School of Medicine, St. Louis, MO, USA.
Dr Marschall uses the example of an outbreak of vancomycin-resistant Enterococcus faecium (VRE) that began in late 2017, in Bern University Hospital, Switzerland and went on until July 2020, receiving substantial media attention and becoming Switzerland’s largest-ever multidrug-resistant organism outbreak.
The investigations into the outbreak revealed most VRE affected patients were colonised and not infected (a “silent outbreak”); that isolation of VRE patients was costly (requiring isolation rooms, personal protective equipment and bed closures); and that screening for VRE required a substantial logistical and financial effort (screening were proximity-based, captured entire floors, or were even hospital-wide).
Dr Marschall and colleagues then re-analysed the medical records generated during the first two years of the outbreak period (1/2018-12/2019) and identified (and mostly confirmed) risk factors for VRE colonisation by using various statistical methods and then moved to a framework called network graph theory and graph neural networks (a type of AI).
Briefly, network graphs inspect the connections between discrete “nodes” in a network (which could be patients, rooms, devices) and determines, for example, which nodes are most connected or which nodes have the shortest path to another node (and thus play a larger role in the outbreak).
Dr Marschall explains: “The more traditional methods for analysing outbreaks might yield signals that tend to confirm previously known and often generic risk factors, without adding the detail needed to make specific interventions”.
Extrapolating from their preliminary work in the field, Dr Marschall thinks AI-based analyses could ultimately provide answers to key questions in an outbreak – for example, their study (click on link above for full article) showed that the “electrocardiography service” and the “examination room ZZ” were at the centre of many interactions, and thus could likely have served as a place of transmission – in consequence, that room/device/person could be the target for disinfection/interventions.
However, in this study, Dr Marschall explains that not all employee interactions were captured because not every profession logs their time with a patient as well as, for example, nursing personnel do (which leaves gaps in our understanding of the many interactions happening in an acute care hospital). To extract the maximum of information, all interactions between patients, employees, visitors, rooms, and devices would have to be captured. Also, Dr Marschall and colleagues based their work on the local data infrastructure in Bern – while the general approach they describe can be used for any setting, a given hospital would need to ensure its data is in an analysable format and labeled so as to facilitate interpretation.
To make AI input into infectious disease outbreaks a success, Dr Marschall explains that hospital teams must pivot from research to operations, meaning implementing and refining AI tools as an outbreak happens in real time (which is not easy to predict), He explains “this approach could even help with individual patients infected with a multidrug-resistant organism or small clusters of such patients because it could identify surrounding patients/employees and rooms/devices that would need to addressed, either by screening or by targeted disinfection. The beauty of AI in outbreak management (and where its greatest power lies) is to make real time or near real time operational decisions easier, quicker and more precise.”
In his talk, Dr Marschall highlights that novel approaches to medical data (such as network graphs and temporal graph neural networks) can provide us with the framework to elevate outbreak investigations to the next stage. It can identify specific "hot spots" of an outbreak, where transmissions are likely to happen, and give us the tools to target these hot spots in order to fight an outbreak. He concludes: “If this help comes to infection prevention experts in near real time, it will dramatically improve their ability to respond to an outbreak.”
COI STATEMENT
Dr Marschall lists the following disclosures: Precision Medicine Approach for Innovative Outbreak Investigation using Machine Learning Methods”. Swiss National Science Foundation, spark grant #190977. PI Atkinson, co-applicant Marschall. 31.01.2020 – 27.02.2022. Part of this work has been presented at ECCMID Lisbon 2022: Ellenberger et al, Spread forecasting of nosocomial vancomycin-resistant Enterococci from hospital interactions using a novel temporal graph neural network. Session B0002. Member, National Center for Infection Control, Swissnoso. 2013-current.
Artificial intelligence has huge potential in infection control, as long as the right questions are asked and safeguards are in place
*Please mention the European Congress of Clinical Microbiology and Infectious Diseases (ECCMID 2024, Barcelona, 27-30 April) if using this material*
A new research review to be given at a pre-congress day for this year’s European Congress of Clinical Microbiology and Infectious Diseases (ECCMID 2024) will look at the many ways artificial intelligence can help prevent infectious disease outbreaks including ensuring staff wear personal protective equipment correctly and managing day-to-day hospital activities such as medication prescription and cleaning. The presentation will be given by Prof Richard Drew, Rotunda Hospital and CHI at Temple St, Irish Meningitis and Sepsis Reference Laboratory and the Royal College of Surgeons in Ireland, Dublin, Ireland.
“Artificial intelligence is a rapidly developing area with huge potential for cost savings, but also wasting money,” explains Prof Drew. “The key is to identify problems in your own institution that AI can help analyse and then fix. For example, can we ensure staff are wearing face masks properly? How do we keep the air/environment clean? When should we switch from intravenous to oral antibiotic therapy for individual patients?”
For the face masks example, Prof Drew will refer to a review paper by Alturki et al, Frontiers in Public Health, 2022, where researchers reviewed how AI was used to both identify firstly if a mask was being worn at all, and secondly if it had been fitted properly. This review paper analysed over 30 papers on the use of facial recognition AI technology to assess if staff were wearing masks correctly, concluding AI performs very well in identifying correct mask wearing in general. “However, even though AI technology successfully identified correct mask wearing, we must be careful that staff do not find such monitoring too intrusive,” says Prof Drew.
He will also look how AI has evolved cleaning in hospitals from traditional manual scrubbing of all corners of the hospital to intelligent robots that know where to focus their cleaning. Robots are, with the assistance of AI, able to monitor the environment and air quality in real time, and then target cleaning where needed.
Recent advances in big data analytics have allowed for research groups from the UK (Bolton et al. Nature Communications, 2024) to analyse data from thousands of admissions to help identify when it is optimal to switch from IV antibiotics to oral antibiotics. Prof Drew explains: “Although this technology will not replace medical experience, it is a tool that could streamline antimicrobial stewardship rounds to focus in on patients who are suitable for oral switch, thus saving staff time and improving patient care.”
In summary, Professor Drew will say the key to successful AI use in infection control is to first identify what problems your institution has and then see if AI can provide a solution. He says: “We should look to offload repetitive tasks to AI systems such as environmental cleaning and mask compliance auditing. AI can also offer significant opportunities in terms of big data analytics of certain patient groups. However, we have to ensure that staff engage with AI developments, and do not feel overwhelmed with the data outputs or consider AI monitoring systems as too intrusive on their personal freedom. It is important too that health systems still appreciate that infection prevention and control (IPC) practitioners are always needed to spot new or emerging problems, identify cultural aspects of IPC, and ensure appropriate communication with other staff.”
COI STATEMENT
The author declares no conflicts of interest
No comments:
Post a Comment