vaccine misinformation and other medical inaccuracies by analyzing large data sets and identifying false or misleading content prior to its widespread dissemination . AI systems could also be deployed to analyze data from social media and other sources in real time , enabling early detection of health threats and improving response times . At the patient care level , AI solutions integrated within EHRs , incorporating natural language processing , enable the efficient triage of patients reporting positive results from SARS-CoV-2 tests taken at home . This integration leads to reduced time required to respond to a positive test result and increases the probability of receiving an antiviral prescription within the critical five-day treatment timeframe .”
Conquering entire biological systems seems to be the desire of some experts , as Wong , et al . ( 2023 ) point out , “ Artificial intelligence , which focuses on developing machines capable of reasoning with data , has recently matured as an exciting field that draws on both these features to accelerate scientific discovery . Because AI-based approaches can integrate large amounts of quantitative and omics data , they are particularly adept at dealing with biological complexity , extending our knowledge and facilitating our efforts to reverse-engineer and control biology . AI-based approaches are particularly useful in addressing the problem of infectious diseases , which are complex across different scales , ranging from cells to communities , and for which advances in medicine and biotechnology are essential drivers of progress .”
The researchers say that AI is essential for anti-infective drug discovery , as existing pharmaceuticals become less effective because of the spread of multi-drug resistance . “ There is therefore an urgent need for new anti-infective treatments , particularly ones that represent unprecedented chemical spaces or therapeutic modalities .” They explain that AI , and in particular machine learning , a subfield of AI which uses data to train machines to make predictions , has been helpful in facilitating training models to identify new drugs or new uses of existing drugs . ML can virtually screen compound libraries at a scale that would be impossible to screen empirically , they say , adding that AI can also predict anti-infective drug activity , drug-target interactions , and therapeutic design .
Additionally , AI ’ s role in facilitating vaccine design and informing treatment strategies is ongoing . As Wong , et al . ( 2023 ) observe , “ Sequence-based ML approaches to mRNA and nucleic acid vaccines can accelerate design , and the turnaround times for the synthesis and experimental validation of these vaccines are short .” They add , “ In general , ML has made an outsized
By analyzing patient data and considering factors such as prior antimicrobial use and culture and susceptibility data , AI algorithms further guide clinicians in determining the likelihood of infection , selecting the most appropriate empiric and targeted regimens , provide dose optimization , and minimize the risk of resistance development .”
contribution to analyzing large and often convoluted datasets in infectious diseases research . While these examples illustrate the promise of using ML to elucidate key factors underlying infections and how infections progress within hosts , understanding host-pathogen interactions and immune responses remains a challenging biological problem . This problem can be addressed by integrating high-throughput datasets — including sequencing , structural , and microscopy data — with detailed mechanistic studies , experimentation , and infection models .”
AI ’ s Role in Antimicrobial Stewardship
Marra , et al . ( 2023 ) say that AI can revolutionize antimicrobial stewardship through personalized treatment and improved outpatient practices .
“ AI enhances treatment decisions by providing individualized , real-time recommendations to healthcare providers on optimal antimicrobial treatment ,” they observe . “ By analyzing patient data and considering factors such as prior antimicrobial use and culture and susceptibility data , AI algorithms further guide clinicians in determining the likelihood of infection , selecting the most appropriate empiric and targeted regimens , provide dose optimization , and minimize the risk of resistance development . The integration of standard operating procedures , analytic tools , data types , and quality control into a laboratory data warehouse accessed by a large language model will create new possibilities for improving clinical microbiology laboratory practices . Additionally , AI can aid in the prediction of antimicrobial resistance patterns directly from mass spectra profiles compared to traditional laboratory-based susceptibility testing . Collaboration between healthcare personnel and AI systems requires a mutual understanding of roles and responsibilities . Efforts to ensure that microbiology technicians , infection prevention and control practitioners , and antimicrobial stewardship clinicians are equipped to work alongside AI technologies , leveraging their expertise in tandem with AI insights , can optimize the potential benefits while minimizing potential disruptions .”
The Rise of the Machines
The benefits of AI must be tempered with a thoughtful consideration of the impact on other aspects of healthcare delivery .
Marra , et al . ( 2023 ) ponder the hesitance around embracing AI too quickly and without qualifiers , noting that “ If technology executives fear AI , so should we .” They point to the May 2023 statement issued by more than 350 tech executives summarizing the imminent public health threat of AI : “ Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war .” This followed another high-profile letter signed by executives of Apple and Tesla calling for a six-month moratorium on the development of advanced AI systems until we have more robust processes to keep them in check .
While the world being taken over by robots may sound like it ’ s a plot from a science fiction novel or movie , there are legitimate concerns around data privacy and ethical considerations . Healthcare professionals are concerned about any potential patient harm caused by biases or a data breach .
As Marra , et al . ( 2023 ) observe , “ All applications of AI rely on individual patient-level data , which ought to be safe and protected . When patients agree to receive healthcare within our institutions , they are not necessarily consenting to use of this data for purposes outside of individualized patient care . Informed consent is mandatory for research involving human subjects , but somehow use of patient data for AI applications is a workaround for the acquisition and use of protected health information . Moreover , large data sets utilized by AI depend on the information fed into the system which can be inaccurate and can contain harmful biases . Of greatest concern is EHR biases based on race , ethnicity , gender , socioeconomic status , education level , and other social determinants of health which can be input into AI data sets and may serve to perpetuate and amplify biases , causing significant patient harm … Concerns also exist about transparency and trust in AI tools . Understanding the sources of training and validation data is fundamental for confidence in large language model capabilities . The ‘ black box ’ nature of these models further exacerbates these concerns , as users are unaware of AI system biases , thereby eroding public trust . This is particularly relevant when AI predictions are incorrect .”
20 • www . healthcarehygienemagazine . com • june 2024