Real-World Integration of a Sepsis Deep Learning Technology Into Routine Clinical Care: Implementation Study

Expectations of Intensive Care Physicians Regarding an AI-Based Decision Support System for Weaning From Continuous Renal Replacement Therapy: Predevelopment Survey Study

Background

Critically ill patients in intensive care units (ICUs) require continuous monitoring, generating vast amounts of data. Clinical decision support systems (CDSS) leveraging artificial intelligence (AI) technologies have shown promise in improving diagnostic, prognostic, and therapeutic decision-making. However, these models are rarely implemented in clinical practice.

Objective

The aim of this study was to survey ICU physicians to understand their expectations, opinions, and level of knowledge regarding a proposed AI-based CDSS for continuous renal replacement therapy (CRRT) weaning, a clinical decision-making process that is still complex and lacking in guidelines. This will be used to guide the development of an AI-based CDSS on which our team is working to ensure user-centered design and successful integration into clinical practice.

Methods

A prospective cross-sectional survey of French-speaking physicians with clinical activity in intensive care was conducted between December 2023 and April 2024. The questionnaire consisted of 20 questions structured around 4 axes: overview of the problem and current practices concerning weaning from CRRT, opinion on AI-based CDSS, implementation in daily clinical practice, real-life operation and willingness to adopt the CDSS in everyday practice. Statistical analyses included Wilcoxon rank sum tests for quantitative variables and χ2 or Fisher exact tests for qualitative variables, with multivariate analyses performed using ordinal logistic regression.

Results

A total of 171 complete responses were received. Physicians expressed an interest in a CDSS for CRRT weaning, with 70.2% (120/171) viewing AI-based CDSS favorably. Opinions were split regarding the difficulty of the weaning decision itself, with 46.2% (79/171) disagreeing that it is challenging, while 31.6% (54/171) agreed. However, 66.1% (113/171) of respondents supported the value of an AI-based CDSS to assist them in this decision, with younger physicians showing stronger support (81.8%, 27/33 vs 62.3%; 86/138; P=.01). Most respondents (163/171, 95.3%) emphasized the importance of understanding the criteria used by the model to make its predictions.

Conclusions

Our findings highlight an optimistic attitude among ICU physicians toward AI-based CDSS for CRRT weaning, emphasizing the need for transparency, integration into existing workflows, and alignment with clinicians' decision-making processes. Actionable recommendations include incorporating key variables such as urine output and biological parameters, defining probability thresholds for recommendations and ensuring model transparency to facilitate the successful adoption and integration into clinical practice. The methodology of this survey may help the development of further predevelopment studies accompanying AI-based CDSS projects.

Managing risk and resilience in autonomous and intelligent systems: Exploring safety in the development, deployment, and use of artificial intelligence in healthcare

Autonomous and intelligent systems (AIS) are being developed and deployed across a wide range of sectors and encompass a variety of technologies designed to engage in different forms of independent reasoning and self-directed behavior. These technologies may bring considerable benefits to society but also pose a range of risk management challenges, particularly when deployed in safety-critical sectors where complex interactions between human, social, and technical processes underpin safety and resilience. Healthcare is one safety-critical sector at the forefront of efforts to develop and deploy intelligent technologies, such as through artificial intelligence (AI) systems intended to automate key aspects of healthcare tasks such as reading medical images to identify signs of pathology. This article develops a qualitative analysis of the sociotechnical sources of risk and resilience associated with the development, deployment, and use of AI in healthcare, drawing on 40 in-depth interviews with participants involved in the development, management, and regulation of AI. Qualitative template analysis is used to examine sociotechnical sources of risk and resilience, drawing on and elaborating Macrae's (2022, Risk Analysis, 42(9), 1999-2025) SOTEC framework that integrates structural, organizational, technological, epistemic, and cultural sources of risk in AIS. This analysis explores an array of sociotechnical sources of risk associated with the development, deployment, and use of AI in healthcare and identifies an array of sociotechnical patterns of resilience that may counter those risks. In doing so, the SOTEC framework is elaborated and translated to define key sources of both risk and resilience in AIS.

Approaches to identify scenarios for data science implementations within healthcare settings: recommendations based on experiences at multiple academic institutions

Objectives

To describe successful and unsuccessful approaches to identify scenarios for data science implementations within healthcare settings and to provide recommendations for future scenario identification procedures.

Materials and methods

Representatives from seven Toronto academic healthcare institutions participated in a one-day workshop. Each institution was asked to provide an introduction to their clinical data science program and to provide an example of a successful and unsuccessful approach to scenario identification at their institution. Using content analysis, common observations were summarized.

Results

Observations were coalesced to idea generation and value proposition, prioritization, approval and champions. Successful experiences included promoting a portfolio of ideas, articulating value proposition, ensuring alignment with organization priorities, ensuring approvers can adjudicate feasibility and identifying champions willing to take ownership over the projects.

Conclusion

Based on academic healthcare data science program experiences, we provided recommendations for approaches to identify scenarios for data science implementations within healthcare settings.

Large Language Model-Based Critical Care Big Data Deployment and Extraction: Descriptive Analysis

Background

Publicly accessible critical care-related databases contain enormous clinical data, but their utilization often requires advanced programming skills. The growing complexity of large databases and unstructured data presents challenges for clinicians who need programming or data analysis expertise to utilize these systems directly.

Objective

This study aims to simplify critical care-related database deployment and extraction via large language models.

Methods

The development of this platform was a 2-step process. First, we enabled automated database deployment using Docker container technology, with incorporated web-based analytics interfaces Metabase and Superset. Second, we developed the intensive care unit-generative pretrained transformer (ICU-GPT), a large language model fine-tuned on intensive care unit (ICU) data that integrated LangChain and Microsoft AutoGen.

Results

The automated deployment platform was designed with user-friendliness in mind, enabling clinicians to deploy 1 or multiple databases in local, cloud, or remote environments without the need for manual setup. After successfully overcoming GPT's token limit and supporting multischema data, ICU-GPT could generate Structured Query Language (SQL) queries and extract insights from ICU datasets based on request input. A front-end user interface was developed for clinicians to achieve code-free SQL generation on the web-based client.

Conclusions

By harnessing the power of our automated deployment platform and ICU-GPT model, clinicians are empowered to easily visualize, extract, and arrange critical care-related databases more efficiently and flexibly than manual methods. Our research could decrease the time and effort spent on complex bioinformatics methods and advance clinical research.

Low responsiveness of machine learning models to critical or deteriorating health conditions

BACKGROUND

Machine learning (ML) based mortality prediction models can be immensely useful in intensive care units. Such a model should generate warnings to alert physicians when a patient's condition rapidly deteriorates, or their vitals are in highly abnormal ranges. Before clinical deployment, it is important to comprehensively assess a model's ability to recognize critical patient conditions.

METHODS

We develop multiple medical ML testing approaches, including a gradient ascent method and neural activation map. We systematically assess these machine learning models' ability to respond to serious medical conditions using additional test cases, some of which are time series. Guided by medical doctors, our evaluation involves multiple machine learning models, resampling techniques, and four datasets for two clinical prediction tasks.

RESULTS

We identify serious deficiencies in the models' responsiveness, with the models being unable to recognize severely impaired medical conditions or rapidly deteriorating health. For in-hospital mortality prediction, the models tested using our synthesized cases fail to recognize 66% of the injuries. In some instances, the models fail to generate adequate mortality risk scores for all test cases. Our study identifies similar kinds of deficiencies in the responsiveness of 5-year breast and lung cancer prediction models.

CONCLUSIONS

Using generated test cases, we find that statistical machine-learning models trained solely from patient data are grossly insufficient and have many dangerous blind spots. Most of the ML models tested fail to respond adequately to critically ill patients. How to incorporate medical knowledge into clinical machine learning models is an important future research direction.

Leveraging artificial intelligence to reduce diagnostic errors in emergency medicine: Challenges, opportunities, and future directions

Diagnostic errors in health care pose significant risks to patient safety and are disturbingly common. In the emergency department (ED), the chaotic and high-pressure environment increases the likelihood of these errors, as emergency clinicians must make rapid decisions with limited information, often under cognitive overload. Artificial intelligence (AI) offers promising solutions to improve diagnostic errors in three key areas: information gathering, clinical decision support (CDS), and feedback through quality improvement. AI can streamline the information-gathering process by automating data retrieval, reducing cognitive load, and providing clinicians with essential patient details quickly. AI-driven CDS systems enhance diagnostic decision making by offering real-time insights, reducing cognitive biases, and prioritizing differential diagnoses. Furthermore, AI-powered feedback loops can facilitate continuous learning and refinement of diagnostic processes by providing targeted education and outcome feedback to clinicians. By integrating AI into these areas, the potential for reducing diagnostic errors and improving patient safety in the ED is substantial. However, successfully implementing AI in the ED is challenging and complex. Developing, validating, and implementing AI as a safe, human-centered ED tool requires thoughtful design and meticulous attention to ethical and practical considerations. Clinicians and patients must be integrated as key stakeholders across these processes. Ultimately, AI should be seen as a tool that assists clinicians by supporting better, faster decisions and thus enhances patient outcomes.

Checklist Approach to Developing and Implementing AI in Clinical Settings: Instrument Development Study

Background

The integration of artificial intelligence (AI) in health care settings demands a nuanced approach that considers both technical performance and sociotechnical factors.

Objective

This study aimed to develop a checklist that addresses the sociotechnical aspects of AI deployment in health care and provides a structured, holistic guide for teams involved in the life cycle of AI systems.

Methods

A literature synthesis identified 20 relevant studies, forming the foundation for the Clinical AI Sociotechnical Framework checklist. A modified Delphi study was then conducted with 35 global health care professionals. Participants assessed the checklist's relevance across 4 stages: "Planning," "Design," "Development," and "Proposed Implementation." A consensus threshold of 80% was established for each item. IQRs and Cronbach α were calculated to assess agreement and reliability.

Results

The initial checklist had 45 questions. Following participant feedback, the checklist was refined to 34 items, and a final round saw 100% consensus on all items (mean score >0.8, IQR 0). Based on the outcome of the Delphi study, a final checklist was outlined, with 1 more question added to make 35 questions in total.

Conclusions

The Clinical AI Sociotechnical Framework checklist provides a comprehensive, structured approach to developing and implementing AI in clinical settings, addressing technical and social factors critical for adoption and success. This checklist is a practical tool that aligns AI development with real-world clinical needs, aiming to enhance patient outcomes and integrate smoothly into health care workflows.

Machine Learning in Optimising Nursing Care Delivery Models: An Empirical Analysis of Hospital Wards

OBJECTIVE

This study aims to assess the performance of machine learning (ML) techniques in optimising nurse staffing and evaluating the appropriateness of nursing care delivery models in hospital wards. The primary outcome measures include the adequacy of nurse staffing and the appropriateness of the nursing care delivery system.

BACKGROUND

Historical and current healthcare challenges, such as nurse shortages and increasing patient acuity, necessitate innovative approaches to nursing care delivery. For instance, the COVID-19 pandemic highlighted the need for flexible and scalable staffing models to manage surges in patient volume and acuity.

MATERIALS AND METHODS

A descriptive study was conducted in 39 inpatient wards across a university hospital and three state hospitals, involving 117 ward-level observations. Data were collected using the Rush Medicus Patient Classification Scale and analysed using k-Nearest Neighbour, Support Vector Machine, Random Forest, and Logistic Regression algorithms. Effectiveness was measured by the accuracy of machine learning predictions regarding nurse staffing adequacy, while suitability was determined by the congruence between observed nursing care models and patient needs.

REPORTING METHOD

STROBE checklist.

RESULTS

The Random Forest algorithm demonstrated the highest accuracy in predicting both nurse staffing adequacy and the appropriateness of nursing care delivery systems. The study found that 68.4% of wards had sufficient nurse staffing and 26.5% of wards used appropriate care delivery models, with functional nursing and total patient care models being the most commonly used.

DISCUSSION

The study highlights functional nursing and total patient care models, emphasising the need to consider nurse qualifications and patient needs in selecting care systems. Machine learning, particularly the Random Forest algorithm, proved effective in aligning staffing with patient requirements.

CONCLUSION

Machine learning, particularly the Random Forest algorithm, proves effective in optimising nursing care delivery models, suggesting significant potential for enhancing patient care and nurse satisfaction.

IMPLICATIONS

The research underscores machine learning's role in improving nursing care delivery, aligning nurse staffing with patient needs, and advancing healthcare outcomes.

IMPACT

The findings advocate for integrating machine learning in the planning of nursing care delivery models. This study sets a precedent for using data-driven approaches to improve nurse staffing and care delivery, potentially enhancing global clinical outcomes and operational efficiencies. The global clinical community can learn from this study the value of employing machine learning techniques to make informed, evidence-based decisions in healthcare management.

PATIENT OR PUBLIC CONTRIBUTION

While the study lacked direct patient involvement, its goal was to enhance patient care and healthcare efficiency. Future research will aim to incorporate patient and public insights more directly.

The Use of AI in Mental Health Services to Support Decision-Making: Scoping Review

BACKGROUND

Recent advancements in artificial intelligence (AI) have changed the care processes in mental health, particularly in decision-making support for health care professionals and individuals with mental health problems. AI systems provide support in several domains of mental health, including early detection, diagnostics, treatment, and self-care. The use of AI systems in care flows faces several challenges in relation to decision-making support, stemming from technology, end-user, and organizational perspectives with the AI disruption of care processes.

OBJECTIVE

This study aims to explore the use of AI systems in mental health to support decision-making, focusing on 3 key areas: the characteristics of research on AI systems in mental health; the current applications, decisions, end users, and user flow of AI systems to support decision-making; and the evaluation of AI systems for the implementation of decision-making support, including elements influencing the long-term use.

METHODS

A scoping review of empirical evidence was conducted across 5 databases: PubMed, Scopus, PsycINFO, Web of Science, and CINAHL. The searches were restricted to peer-reviewed articles published in English after 2011. The initial screening at the title and abstract level was conducted by 2 reviewers, followed by full-text screening based on the inclusion criteria. Data were then charted and prepared for data analysis.

RESULTS

Of a total of 1217 articles, 12 (0.99%) met the inclusion criteria. These studies predominantly originated from high-income countries. The AI systems were used in health care, self-care, and hybrid care contexts, addressing a variety of mental health problems. Three types of AI systems were identified in terms of decision-making support: diagnostic and predictive AI, treatment selection AI, and self-help AI. The dynamics of the type of end-user interaction and system design were diverse in complexity for the integration and use of the AI systems to support decision-making in care processes. The evaluation of the use of AI systems highlighted several challenges impacting the implementation and functionality of the AI systems in care processes, including factors affecting accuracy, increase of demand, trustworthiness, patient-physician communication, and engagement with the AI systems.

CONCLUSIONS

The design, development, and implementation of AI systems to support decision-making present substantial challenges for the sustainable use of this technology in care processes. The empirical evidence shows that the evaluation of the use of AI systems in mental health is still in its early stages, with need for more empirically focused research on real-world use. The key aspects requiring further investigation include the evaluation of the use of AI-supported decision-making from human-AI interaction and human-computer interaction perspectives, longitudinal implementation studies of AI systems in mental health to assess the use, and the integration of shared decision-making in AI systems.

Empowering nurses to champion Health equity & BE FAIR: Bias elimination for fair and responsible AI in healthcare

BACKGROUND

The concept of health equity by design encompasses a multifaceted approach that integrates actions aimed at eliminating biased, unjust, and correctable differences among groups of people as a fundamental element in the design of algorithms. As algorithmic tools are increasingly integrated into clinical practice at multiple levels, nurses are uniquely positioned to address challenges posed by the historical marginalization of minority groups and its intersections with the use of "big data" in healthcare settings; however, a coherent framework is needed to ensure that nurses receive appropriate training in these domains and are equipped to act effectively.

PURPOSE

We introduce the Bias Elimination for Fair AI in Healthcare (BE FAIR) framework, a comprehensive strategic approach that incorporates principles of health equity by design, for nurses to employ when seeking to mitigate bias and prevent discriminatory practices arising from the use of clinical algorithms in healthcare. By using examples from a "real-world" AI governance framework, we aim to initiate a wider discourse on equipping nurses with the skills needed to champion the BE FAIR initiative.

METHODS

Drawing on principles recently articulated by the Office of the National Coordinator for Health Information Technology, we conducted a critical examination of the concept of health equity by design. We also reviewed recent literature describing the risks of artificial intelligence (AI) technologies in healthcare as well as their potential for advancing health equity. Building on this context, we describe the BE FAIR framework, which has the potential to enable nurses to take a leadership role within health systems by implementing a governance structure to oversee the fairness and quality of clinical algorithms. We then examine leading frameworks for promoting health equity to inform the operationalization of BE FAIR within a local AI governance framework.

RESULTS

The application of the BE FAIR framework within the context of a working governance system for clinical AI technologies demonstrates how nurses can leverage their expertise to support the development and deployment of clinical algorithms, mitigating risks such as bias and promoting ethical, high-quality care powered by big data and AI technologies.

CONCLUSION AND RELEVANCE

As health systems learn how well-intentioned clinical algorithms can potentially perpetuate health disparities, we have an opportunity and an obligation to do better. New efforts empowering nurses to advocate for BE FAIR, involving them in AI governance, data collection methods, and the evaluation of tools intended to reduce bias, mark important steps in achieving equitable healthcare for all.

The Role of Machine Learning in the Detection of Cardiac Fibrosis in Electrocardiograms: Scoping Review

BACKGROUND

Cardiovascular disease remains the leading cause of mortality worldwide. Cardiac fibrosis impacts the underlying pathophysiology of many cardiovascular diseases by altering structural integrity and impairing electrical conduction. Identifying cardiac fibrosis is essential for the prognosis and management of cardiovascular disease; however, current diagnostic methods face challenges due to invasiveness, cost, and inaccessibility. Electrocardiograms (ECGs) are widely available and cost-effective for monitoring cardiac electrical activity. While ECG-based methods for inferring fibrosis exist, they are not commonly used due to accuracy limitations and the need for cardiac expertise. However, the ECG shows promise as a target for machine learning (ML) applications in fibrosis detection.

OBJECTIVE

This study aims to synthesize and critically evaluate the current state of ECG-based ML approaches for cardiac fibrosis detection.

METHODS

We conducted a scoping review of research in ECG-based ML applications to identify cardiac fibrosis. Comprehensive searches were performed in PubMed, IEEE Xplore, Scopus, Web of Science, and DBLP databases, including publications up to October 2024. Studies were included if they applied ML techniques to detect cardiac fibrosis using ECG or vectorcardiogram data and provided sufficient methodological details and outcome metrics. Two reviewers independently assessed eligibility and extracted data on the ML models used, their performance metrics, study designs, and limitations.

RESULTS

We identified 11 studies evaluating ML approaches for detecting cardiac fibrosis using ECG data. These studies used various ML techniques, including classical (8/11, 73%), ensemble (3/11, 27%), and deep learning models (4/11, 36%). Support vector machines were the most used classical model (6/11, 55%), with the best-performing models of each study achieving accuracies of 77% to 93%. Among deep learning approaches, convolutional neural networks showed promising results, with one study reporting an area under the receiver operating characteristic curve (AUC) of 0.89 when combined with clinical features. Notably, a large-scale convolutional neural network study (n=14,052) achieved an AUC of 0.84 for detecting cardiac fibrosis, outperforming cardiologists (AUC 0.63-0.66). However, many studies had limited sample sizes and lacked external validation, potentially impacting the generalizability of the findings. Variability in reporting methods may affect the reproducibility and applicability of these ML-based approaches.

CONCLUSIONS

ML-augmented ECG analysis shows promise for accessible and cost-effective detection of cardiac fibrosis. However, there are common limitations with respect to study design and insufficient external validation, raising concerns about the generalizability and clinical applicability of the findings. Inconsistencies in methodologies and incomplete reporting further impede cross-study comparisons. Future work may benefit from using prospective study designs, larger and more clinically and demographically diverse datasets, advanced ML models, and rigorous external validation. Addressing these challenges could pave the way for the clinical implementation of ML-based ECG detection of cardiac fibrosis to improve patient outcomes and health care resource allocation.

Artificial Intelligence in Nursing: Technological Benefits to Nurse's Mental Health and Patient Care Quality

Nurses are frontline caregivers who handle heavy workloads and high-stakes activities. They face several mental health issues, including stress, burnout, anxiety, and depression. The welfare of nurses and the standard of patient treatment depends on resolving this problem. Artificial intelligence is revolutionising healthcare, and its integration provides many possibilities in addressing these concerns. This review examines literature published over the past 40 years, concentrating on AI integration in nursing for mental health support, improved patient care, and ethical issues. Using databases such as PubMed and Google Scholar, a thorough search was conducted with Boolean operators, narrowing results for relevance. Critically examined were publications on artificial intelligence applications in patient care ethics, mental health, and nursing and mental health. The literature examination revealed that, by automating repetitive chores and improving workload management, artificial intelligence (AI) can relieve mental health challenges faced by nurses and improve patient care. Practical implications highlight the requirement of using rigorous implementation strategies that address ethical issues, data privacy, and human-centred decision-making. All changes must direct the integration of artificial intelligence in nursing to guarantee its sustained and significant influence on healthcare.

Implementation of Machine Learning Applications in Health Care Organizations: Systematic Review of Empirical Studies

BACKGROUND

There is a growing enthusiasm for machine learning (ML) among academics and health care practitioners. Despite the transformative potential of ML-based applications for patient care, their uptake and implementation in health care organizations are sporadic. Numerous challenges currently impede or delay the widespread implementation of ML in clinical practice, and limited knowledge is available regarding how these challenges have been addressed.

OBJECTIVE

This work aimed to (1) examine the characteristics of ML-based applications and the implementation process in clinical practice, using the Consolidated Framework for Implementation Research (CFIR) for theoretical guidance and (2) synthesize the strategies adopted by health care organizations to foster successful implementation of ML.

METHODS

A systematic literature review was conducted based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. The search was conducted in PubMed, Scopus, and Web of Science over a 10-year period (2013-2023). The search strategy was built around 4 blocks of keywords (artificial intelligence, implementation, health care, and study type). Only empirical studies documenting the implementation of ML applications in clinical settings were considered. The implementation process was investigated using a thematic analysis and coding procedure.

RESULTS

Thirty-four studies were selected for data synthesis. Selected papers were relatively recent, with only 9% (3/34) of records published before 2019. ML-based applications were implemented mostly within hospitals (29/34, 85%). In terms of clinical workflow, ML-based applications supported mostly prognosis (20/34, 59%) and diagnosis (10/34, 29%). The implementation efforts were analyzed using CFIR domains. As for the inner setting domain, access to knowledge and information (12/34, 35%), information technology infrastructure (11/34, 32%), and organizational culture (9/34, 26%) were among the most observed dimensions influencing the success of implementation. As for the ML innovation itself, factors deemed relevant were its design (15/34, 44%), the relative advantage with respect to existing clinical practice (14/34, 41%), and perceived complexity (14/34, 41%). As for the other domains (ie, processes, roles, and outer setting), stakeholder engagement (12/34, 35%), reflecting and evaluating practices (11/34, 32%), and the presence of implementation leaders (9/34, 26%) were the main factors identified as important.

CONCLUSIONS

This review sheds some light on the factors that are relevant and that should be accounted for in the implementation process of ML-based applications in health care. While the relevance of ML-specific dimensions, like trust, emerges clearly across several implementation domains, the evidence from this review highlighted that relevant implementation factors are not necessarily specific for ML but rather transversal for digital health technologies. More research is needed to further clarify the factors that are relevant to implementing ML-based applications at the organizational level and to support their uptake within health care organizations.

TRIAL REGISTRATION

PROSPERO 403873; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=403873.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

RR2-10.2196/47971.

Artificial intelligence in healthcare: a scoping review of perceived threats to patient rights and safety

BACKGROUND

The global health system remains determined to leverage on every workable opportunity, including artificial intelligence (AI) to provide care that is consistent with patients' needs. Unfortunately, while AI models generally return high accuracy within the trials in which they are trained, their ability to predict and recommend the best course of care for prospective patients is left to chance.

PURPOSE

This review maps evidence between January 1, 2010 to December 31, 2023, on the perceived threats posed by the usage of AI tools in healthcare on patients' rights and safety.

METHODS

We deployed the guidelines of Tricco et al. to conduct a comprehensive search of current literature from Nature, PubMed, Scopus, ScienceDirect, Dimensions AI, Web of Science, Ebsco Host, ProQuest, JStore, Semantic Scholar, Taylor & Francis, Emeralds, World Health Organisation, and Google Scholar. In all, 80 peer reviewed articles qualified and were included in this study.

RESULTS

We report that there is a real chance of unpredictable errors, inadequate policy and regulatory regime in the use of AI technologies in healthcare. Moreover, medical paternalism, increased healthcare cost and disparities in insurance coverage, data security and privacy concerns, and bias and discriminatory services are imminent in the use of AI tools in healthcare.

CONCLUSIONS

Our findings have some critical implications for achieving the Sustainable Development Goals (SDGs) 3.8, 11.7, and 16. We recommend that national governments should lead in the roll-out of AI tools in their healthcare systems. Also, other key actors in the healthcare industry should contribute to developing policies on the use of AI in healthcare systems.

Machine Learning Operations in Health Care: A Scoping Review

The use of machine learning tools in health care is rapidly expanding. However, the processes that support these tools in deployment, that is, machine learning operations, are still emerging. The purpose of this work was not only to provide a comprehensive synthesis of existing literature in the field but also to identify gaps and offer insights for adoption in clinical practice. A scoping review was conducted using the MEDLINE, PubMed, Google Scholar, Embase, and Scopus databases. We used MeSH and non-MeSH search terms to identify pertinent articles, with the authors performing 2 screening phases and assigning relevance scores: 148 English language articles most salient to the review were eligible for inclusion; 98 offered the most unique information and these were supplemented by 50 additional sources, yielding 148 references. From the 148 references, we distilled 7 key topic areas, based on a synthesis of the available literature and how that aligned with practitioner needs. The 7 topic areas were machine learning model monitoring; automated retraining systems; ethics, equity, and bias; clinical workflow integration; infrastructure, human resources, and technology stack; regulatory considerations; and financial considerations. This review provides an overview of best practices and knowledge gaps of this domain in health care and identifies the strengths and weaknesses of the literature, which may be useful to health care machine learning practitioners and consumers.

Recommendations to promote fairness and inclusion in biomedical AI research and clinical use

OBJECTIVE

Understanding and quantifying biases when designing and implementing actionable approaches to increase fairness and inclusion is critical for artificial intelligence (AI) in biomedical applications.

METHODS

In this Special Communication, we discuss how bias is introduced at different stages of the development and use of AI applications in biomedical sciences and health care. We describe various AI applications and their implications for fairness and inclusion in sections on 1) Bias in Data Source Landscapes, 2) Algorithmic Fairness, 3) Uncertainty in AI Predictions, 4) Explainable AI for Fairness and Equity, and 5) Sociological/Ethnographic Issues in Data and Results Representation.

RESULTS

We provide recommendations to address biases when developing and using AI in clinical applications.

CONCLUSION

These recommendations can be applied to informatics research and practice to foster more equitable and inclusive health care systems and research discoveries.

A comprehensive overview of barriers and strategies for AI implementation in healthcare: Mixed-method design

Implementation of artificial intelligence systems for healthcare is challenging. Understanding the barriers and implementation strategies can impact their adoption and allows for better anticipation and planning. This study's objective was to create a detailed inventory of barriers to and strategies for AI implementation in healthcare to support advancements in methods and implementation processes in healthcare. A sequential explanatory mixed method design was used. Firstly, scoping reviews and systematic literature reviews were identified using PubMed. Selected studies included empirical cases of AI implementation and use in clinical practice. As the reviews were deemed insufficient to fulfil the aim of the study, data collection shifted to the primary studies included in those reviews. The primary studies were screened by title and abstract, and thereafter read in full text. Then, data on barriers to and strategies for AI implementation were extracted from the included articles, thematically coded by inductive analysis, and summarized. Subsequently, a direct qualitative content analysis of 69 interviews with healthcare leaders and healthcare professionals confirmed and added results from the literature review. Thirty-eight empirical cases from the six identified scoping and literature reviews met the inclusion and exclusion criteria. Barriers to and strategies for AI implementation were grouped under three phases of implementation (planning, implementing, and sustaining the use) and were categorized into eleven concepts; Leadership, Buy-in, Change management, Engagement, Workflow, Finance and human resources, Legal, Training, Data, Evaluation and monitoring, Maintenance. Ethics emerged as a twelfth concept through qualitative analysis of the interviews. This study illustrates the inherent challenges and useful strategies in implementing AI in healthcare practice. Future research should explore various aspects of leadership, collaboration and contracts among key stakeholders, legal strategies surrounding clinicians' liability, solutions to ethical dilemmas, infrastructure for efficient integration of AI in workflows, and define decision points in the implementation process.

Implementing AI in Hospitals to Achieve a Learning Health System: Systematic Review of Current Enablers and Barriers

BACKGROUND

Efforts are underway to capitalize on the computational power of the data collected in electronic medical records (EMRs) to achieve a learning health system (LHS). Artificial intelligence (AI) in health care has promised to improve clinical outcomes, and many researchers are developing AI algorithms on retrospective data sets. Integrating these algorithms with real-time EMR data is rare. There is a poor understanding of the current enablers and barriers to empower this shift from data set-based use to real-time implementation of AI in health systems. Exploring these factors holds promise for uncovering actionable insights toward the successful integration of AI into clinical workflows.

OBJECTIVE

The first objective was to conduct a systematic literature review to identify the evidence of enablers and barriers regarding the real-world implementation of AI in hospital settings. The second objective was to map the identified enablers and barriers to a 3-horizon framework to enable the successful digital health transformation of hospitals to achieve an LHS.

METHODS

The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines were adhered to. PubMed, Scopus, Web of Science, and IEEE Xplore were searched for studies published between January 2010 and January 2022. Articles with case studies and guidelines on the implementation of AI analytics in hospital settings using EMR data were included. We excluded studies conducted in primary and community care settings. Quality assessment of the identified papers was conducted using the Mixed Methods Appraisal Tool and ADAPTE frameworks. We coded evidence from the included studies that related to enablers of and barriers to AI implementation. The findings were mapped to the 3-horizon framework to provide a road map for hospitals to integrate AI analytics.

RESULTS

Of the 1247 studies screened, 26 (2.09%) met the inclusion criteria. In total, 65% (17/26) of the studies implemented AI analytics for enhancing the care of hospitalized patients, whereas the remaining 35% (9/26) provided implementation guidelines. Of the final 26 papers, the quality of 21 (81%) was assessed as poor. A total of 28 enablers was identified; 8 (29%) were new in this study. A total of 18 barriers was identified; 5 (28%) were newly found. Most of these newly identified factors were related to information and technology. Actionable recommendations for the implementation of AI toward achieving an LHS were provided by mapping the findings to a 3-horizon framework.

CONCLUSIONS

Significant issues exist in implementing AI in health care. Shifting from validating data sets to working with live data is challenging. This review incorporated the identified enablers and barriers into a 3-horizon framework, offering actionable recommendations for implementing AI analytics to achieve an LHS. The findings of this study can assist hospitals in steering their strategic planning toward successful adoption of AI.

The AI Future of Emergency Medicine

In the coming years, artificial intelligence (AI) and machine learning will likely give rise to profound changes in the field of emergency medicine, and medicine more broadly. This article discusses these anticipated changes in terms of 3 overlapping yet distinct stages of AI development. It reviews some fundamental concepts in AI and explores their relation to clinical practice, with a focus on emergency medicine. In addition, it describes some of the applications of AI in disease diagnosis, prognosis, and treatment, as well as some of the practical issues that they raise, the barriers to their implementation, and some of the legal and regulatory challenges they create.

Implementation considerations for the adoption of artificial intelligence in the emergency department

OBJECTIVE

Artificial intelligence (AI) has emerged as a potentially transformative force, particularly in the realm of emergency medicine (EM). The implementation of AI in emergency departments (ED) has the potential to improve patient care through various modalities. However, the implementation of AI in the ED presents unique challenges that influence its clinical adoption. This scoping review summarizes the current literature exploring the barriers and facilitators of the clinical implementation of AI in the ED.

METHODS

We systematically searched Embase (Ovid), MEDLINE (Ovid), Web of Science, and Engineering Village. All articles were published in English through November 20th, 2023. Two reviewers screened the search results, with disagreements resolved through third-party adjudication.

RESULTS

A total of 8172 studies were included in the preliminary search, with 22 selected for the final data extraction. 10 studies were reviews and the remaining 12 were primary quantitative, qualitative, and mixed-methods studies. Out of the 22, 13 studies investigated a specific AI tool or application. Common barriers to implementation included a lack of model interpretability and explainability, encroachment on physician autonomy, and medicolegal considerations. Common facilitators to implementation included educating staff on the model, efficient integration into existing workflows, and sound external validation.

CONCLUSION

There is increasing literature on AI implementation in the ED. Our research suggests that the most common barrier facing AI implementation in the ED is model interpretability and explainability. More primary research investigating the implementation of specific AI tools should be undertaken to help facilitate their successful clinical adoption in the ED.

Leveraging Artificial Intelligence for Expediting Implementation Efforts

Expedited implementation of evidence into practice and policymaking is critical to ensure the delivery of effective care and improve health-care outcomes. Implementation science deals with the designing of methods and strategies for increasing and facilitating the uptake of evidence into practice and policymaking. Nevertheless, the process of designing and selecting methods and strategies for implementing evidence is complicated because of the complexity of health-care settings where implementation is desired. Artificial intelligence (AI) has revolutionized a range of fields, including genomics, education, drug trials, research, and health care. This commentary discusses how AI can be leveraged to expedite implementation science efforts for transforming health-care practice. Four key aspects of AI use in implementation science are highlighted: (a) AI for implementation planning (e.g., needs assessment, predictive analytics, and data management), (b) AI for developing implementation tools and guidelines, (c) AI for designing and applying implementation strategies, and (d) AI for monitoring and evaluating implementation outcomes. Use of AI along the implementation continuum from planning to delivery and evaluation can enable more precise and accurate implementation of evidence into practice.

The algorithm journey map: a tangible approach to implementing AI solutions in healthcare

When integrating AI tools in healthcare settings, complex interactions between technologies and primary users are not always fully understood or visible. This deficient and ambiguous understanding hampers attempts by healthcare organizations to adopt AI/ML, and it also creates new challenges for researchers to identify opportunities for simplifying adoption and developing best practices for the use of AI-based solutions. Our study fills this gap by documenting the process of designing, building, and maintaining an AI solution called SepsisWatch at Duke University Health System. We conducted 20 interviews with the team of engineers and scientists that led the multi-year effort to build the tool, integrate it into practice, and maintain the solution. This "Algorithm Journey Map" enumerates all social and technical activities throughout the AI solution's procurement, development, integration, and full lifecycle management. In addition to mapping the "who?" and "what?" of the adoption of the AI tool, we also show several 'lessons learned' throughout the algorithm journey maps including modeling assumptions, stakeholder inclusion, and organizational structure. In doing so, we identify generalizable insights about how to recognize and navigate barriers to AI/ML adoption in healthcare settings. We expect that this effort will further the development of best practices for operationalizing and sustaining ethical principles-in algorithmic systems.

Machine learning for healthcare that matters: Reorienting from technical novelty to equitable impact

Despite significant technical advances in machine learning (ML) over the past several years, the tangible impact of this technology in healthcare has been limited. This is due not only to the particular complexities of healthcare, but also due to structural issues in the machine learning for healthcare (MLHC) community which broadly reward technical novelty over tangible, equitable impact. We structure our work as a healthcare-focused echo of the 2012 paper "Machine Learning that Matters", which highlighted such structural issues in the ML community at large, and offered a series of clearly defined "Impact Challenges" to which the field should orient itself. Drawing on the expertise of a diverse and international group of authors, we engage in a narrative review and examine issues in the research background environment, training processes, evaluation metrics, and deployment protocols which act to limit the real-world applicability of MLHC. Broadly, we seek to distinguish between machine learning ON healthcare data and machine learning FOR healthcare-the former of which sees healthcare as merely a source of interesting technical challenges, and the latter of which regards ML as a tool in service of meeting tangible clinical needs. We offer specific recommendations for a series of stakeholders in the field, from ML researchers and clinicians, to the institutions in which they work, and the governments which regulate their data access.

Impact of a deep learning sepsis prediction model on quality of care and survival

Sepsis remains a major cause of mortality and morbidity worldwide. Algorithms that assist with the early recognition of sepsis may improve outcomes, but relatively few studies have examined their impact on real-world patient outcomes. Our objective was to assess the impact of a deep-learning model (COMPOSER) for the early prediction of sepsis on patient outcomes. We completed a before-and-after quasi-experimental study at two distinct Emergency Departments (EDs) within the UC San Diego Health System. We included 6217 adult septic patients from 1/1/2021 through 4/30/2023. The exposure tested was a nurse-facing Best Practice Advisory (BPA) triggered by COMPOSER. In-hospital mortality, sepsis bundle compliance, 72-h change in sequential organ failure assessment (SOFA) score following sepsis onset, ICU-free days, and the number of ICU encounters were evaluated in the pre-intervention period (705 days) and the post-intervention period (145 days). The causal impact analysis was performed using a Bayesian structural time-series approach with confounder adjustments to assess the significance of the exposure at the 95% confidence level. The deployment of COMPOSER was significantly associated with a 1.9% absolute reduction (17% relative decrease) in in-hospital sepsis mortality (95% CI, 0.3%-3.5%), a 5.0% absolute increase (10% relative increase) in sepsis bundle compliance (95% CI, 2.4%-8.0%), and a 4% (95% CI, 1.1%-7.1%) reduction in 72-h SOFA change after sepsis onset in causal inference analysis. This study suggests that the deployment of COMPOSER for early prediction of sepsis was associated with a significant reduction in mortality and a significant increase in sepsis bundle compliance.

Grand rounds in methodology: key considerations for implementing machine learning solutions in quality improvement initiatives

Machine learning (ML) solutions are increasingly entering healthcare. They are complex, sociotechnical systems that include data inputs, ML models, technical infrastructure and human interactions. They have promise for improving care across a wide range of clinical applications but if poorly implemented, they may disrupt clinical workflows, exacerbate inequities in care and harm patients. Many aspects of ML solutions are similar to other digital technologies, which have well-established approaches to implementation. However, ML applications present distinct implementation challenges, given that their predictions are often complex and difficult to understand, they can be influenced by biases in the data sets used to develop them, and their impacts on human behaviour are poorly understood. This manuscript summarises the current state of knowledge about implementing ML solutions in clinical care and offers practical guidance for implementation. We propose three overarching questions for potential users to consider when deploying ML solutions in clinical care: (1) Is a clinical or operational problem likely to be addressed by an ML solution? (2) How can an ML solution be evaluated to determine its readiness for deployment? (3) How can an ML solution be deployed and maintained optimally? The Quality Improvement community has an essential role to play in ensuring that ML solutions are translated into clinical practice safely, effectively, and ethically.

AI and professional liability assessment in healthcare. A revolution in legal medicine?

The adoption of advanced artificial intelligence (AI) systems in healthcare is transforming the healthcare-delivery landscape. Artificial intelligence may enhance patient safety and improve healthcare outcomes, but it presents notable ethical and legal dilemmas. Moreover, as AI streamlines the analysis of the multitude of factors relevant to malpractice claims, including informed consent, adherence to standards of care, and causation, the evaluation of professional liability might also benefit from its use. Beginning with an analysis of the basic steps in assessing professional liability, this article examines the potential new medical-legal issues that an expert witness may encounter when analyzing malpractice cases and the potential integration of AI in this context. These changes related to the use of integrated AI, will necessitate efforts on the part of judges, experts, and clinicians, and may require new legislative regulations. A new expert witness will be likely necessary in the evaluation of professional liability cases. On the one hand, artificial intelligence will support the expert witness; however, on the other hand, it will introduce specific elements into the activities of healthcare workers. These elements will necessitate an expert witness with a specialized cultural background. Examining the steps of professional liability assessment indicates that the likely path for AI in legal medicine involves its role as a collaborative and integrated tool. The combination of AI with human judgment in these assessments can enhance comprehensiveness and fairness. However, it is imperative to adopt a cautious and balanced approach to prevent complete automation in this field.

[Use of AI-based applications by hospital staff: task profiles and qualification requirements]

BACKGROUND

Artificial intelligence (AI) is becoming increasingly important for the future development of hospitals. To unlock the large potential of AI, job profiles of hospital staff members need to be further developed in the direction of AI and digitization skills through targeted qualification measures. This affects both medical and non-medical processes along the entire value chain in hospitals. The aim of this paper is to provide an overview of the skills required to deal with smart technologies in a clinical context and to present measures for training employees.

METHODS

As part of the "SmartHospital.NRW" project in 2022, we conducted a literature review as well as interviews and workshops with experts. AI technologies and fields of application were identified.

RESULTS

Key findings include adapted and new task profiles, synergies and dependencies between individual task profiles, and the need for a comprehensive interdisciplinary and interprofessional exchange when using AI-based applications in hospitals.

DISCUSSION

Our article shows that hospitals need to promote digital health literacy skills for hospital staff members at an early stage and at the same time recruit technology- and AI-savvy staff. Interprofessional exchange formats and accompanying change management are essential for the use of AI in hospitals.

A framework for understanding label leakage in machine learning for health care

INTRODUCTION

The pitfalls of label leakage, contamination of model input features with outcome information, are well established. Unfortunately, avoiding label leakage in clinical prediction models requires more nuance than the common advice of applying "no time machine rule."

FRAMEWORK

We provide a framework for contemplating whether and when model features pose leakage concerns by considering the cadence, perspective, and applicability of predictions. To ground these concepts, we use real-world clinical models to highlight examples of appropriate and inappropriate label leakage in practice.

RECOMMENDATIONS

Finally, we provide recommendations to support clinical and technical stakeholders as they evaluate the leakage tradeoffs associated with model design, development, and implementation decisions. By providing common language and dimensions to consider when designing models, we hope the clinical prediction community will be better prepared to develop statistically valid and clinically useful machine learning models.

21st century (clinical) decision support in nursing and allied healthcare. Developing a learning health system: a reasoned design of a theoretical framework

In this paper, we present a framework for developing a Learning Health System (LHS) to provide means to a computerized clinical decision support system for allied healthcare and/or nursing professionals. LHSs are well suited to transform healthcare systems in a mission-oriented approach, and is being adopted by an increasing number of countries. Our theoretical framework provides a blueprint for organizing such a transformation with help of evidence based state of the art methodologies and techniques to eventually optimize personalized health and healthcare. Learning via health information technologies using LHS enables users to learn both individually and collectively, and independent of their location. These developments demand healthcare innovations beyond a disease focused orientation since clinical decision making in allied healthcare and nursing is mainly based on aspects of individuals' functioning, wellbeing and (dis)abilities. Developing LHSs depends heavily on intertwined social and technological innovation, and research and development. Crucial factors may be the transformation of the Internet of Things into the Internet of FAIR data & services. However, Electronic Health Record (EHR) data is in up to 80% unstructured including free text narratives and stored in various inaccessible data warehouses. Enabling the use of data as a driver for learning is challenged by interoperability and reusability.To address technical needs, key enabling technologies are suitable to convert relevant health data into machine actionable data and to develop algorithms for computerized decision support. To enable data conversions, existing classification and terminology systems serve as definition providers for natural language processing through (un)supervised learning.To facilitate clinical reasoning and personalized healthcare using LHSs, the development of personomics and functionomics are useful in allied healthcare and nursing. Developing these omics will be determined via text and data mining. This will focus on the relationships between social, psychological, cultural, behavioral and economic determinants, and human functioning.Furthermore, multiparty collaboration is crucial to develop LHSs, and man-machine interaction studies are required to develop a functional design and prototype. During development, validation and maintenance of the LHS continuous attention for challenges like data-drift, ethical, technical and practical implementation difficulties is required.

The use of probabilistic graphical models in pediatric sepsis: a feasibility and scoping review

Background

Recent research has demonstrated that machine learning (ML) has the potential to improve several aspects of medical application for critical illness, including sepsis. This scoping review aims to evaluate the feasibility of probabilistic graphical model (PGM) methods in pediatric sepsis application and describe the use of pediatric sepsis definition in these studies.

Methods

Literature searches were conducted in PubMed, Scopus, Cumulative Index to Nursing and Allied Health Literature (CINAHL+), and Web of Sciences from 2000-2023. Keywords included "pediatric", "neonates", "infants", "machine learning", "probabilistic graphical model", and "sepsis".

Results

A total of 3,244 studies were screened, and 72 were included in this scoping review. Sepsis was defined using positive microbiology cultures in 19 studies (26.4%), followed by the 2005's international pediatric sepsis consensus definition in 11 studies (15.3%), and Sepsis-3 definition in seven studies (9.7%). Other sepsis definitions included: bacterial infection, the international classification of diseases, clinicians' assessment, and antibiotic administration time. Among the most common ML approaches used were logistic regression (n=27), random forest (n=24), and Neural Network (n=18). PGMs were used in 13 studies (18.1%), including Bayesian classifiers (n=10), and the Markov Model (n=3). When applied on the same dataset, PGMs show a relatively inferior performance to other ML models in most cases. Other aspects of explainability and transparency were not examined in these studies.

Conclusions

Current studies suggest that the performance of probabilistic graphic models is relatively inferior to other ML methods. However, its explainability and transparency advantages make it a potentially viable method for several pediatric sepsis studies and applications.

Effects of machine learning-based clinical decision support systems on decision-making, care delivery, and patient outcomes: a scoping review

OBJECTIVE

This study aims to summarize the research literature evaluating machine learning (ML)-based clinical decision support (CDS) systems in healthcare settings.

MATERIALS AND METHODS

We conducted a review in accordance with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta Analyses extension for Scoping Review). Four databases, including PubMed, Medline, Embase, and Scopus were searched for studies published from January 2016 to April 2021 evaluating the use of ML-based CDS in clinical settings. We extracted the study design, care setting, clinical task, CDS task, and ML method. The level of CDS autonomy was examined using a previously published 3-level classification based on the division of clinical tasks between the clinician and CDS; effects on decision-making, care delivery, and patient outcomes were summarized.

RESULTS

Thirty-two studies evaluating the use of ML-based CDS in clinical settings were identified. All were undertaken in developed countries and largely in secondary and tertiary care settings. The most common clinical tasks supported by ML-based CDS were image recognition and interpretation (n = 12) and risk assessment (n = 9). The majority of studies examined assistive CDS (n = 23) which required clinicians to confirm or approve CDS recommendations for risk assessment in sepsis and for interpreting cancerous lesions in colonoscopy. Effects on decision-making, care delivery, and patient outcomes were mixed.

CONCLUSION

ML-based CDS are being evaluated in many clinical areas. There remain many opportunities to apply and evaluate effects of ML-based CDS on decision-making, care delivery, and patient outcomes, particularly in resource-constrained settings.

A Systematic Review of the Barriers to the Implementation of Artificial Intelligence in Healthcare

Artificial intelligence (AI) is expected to improve healthcare outcomes by facilitating early diagnosis, reducing the medical administrative burden, aiding drug development, personalising medical and oncological management, monitoring healthcare parameters on an individual basis, and allowing clinicians to spend more time with their patients. In the post-pandemic world where there is a drive for efficient delivery of healthcare and manage long waiting times for patients to access care, AI has an important role in supporting clinicians and healthcare systems to streamline the care pathways and provide timely and high-quality care for the patients. Despite AI technologies being used in healthcare for some decades, and all the theoretical potential of AI, the uptake in healthcare has been uneven and slower than anticipated and there remain a number of barriers, both overt and covert, which have limited its incorporation. This literature review highlighted barriers in six key areas: ethical, technological, liability and regulatory, workforce, social, and patient safety barriers. Defining and understanding the barriers preventing the acceptance and implementation of AI in the setting of healthcare will enable clinical staff and healthcare leaders to overcome the identified hurdles and incorporate AI technologies for the benefit of patients and clinical staff.

Barriers and Enablers for Implementation of an Artificial Intelligence-Based Decision Support Tool to Reduce the Risk of Readmission of Patients With Heart Failure: Stakeholder Interviews

BACKGROUND

Artificial intelligence (AI) applications in health care are expected to provide value for health care organizations, professionals, and patients. However, the implementation of such systems should be carefully planned and organized in order to ensure quality, safety, and acceptance. The gathered view of different stakeholders is a great source of information to understand the barriers and enablers for implementation in a specific context.

OBJECTIVE

This study aimed to understand the context and stakeholder perspectives related to the future implementation of a clinical decision support system for predicting readmissions of patients with heart failure. The study was part of a larger project involving model development, interface design, and implementation planning of the system.

METHODS

Interviews were held with 12 stakeholders from the regional and municipal health care organizations to gather their views on the potential effects implementation of such a decision support system could have as well as barriers and enablers for implementation. Data were analyzed based on the categories defined in the nonadoption, abandonment, scale-up, spread, sustainability (NASSS) framework.

RESULTS

Stakeholders had in general a positive attitude and curiosity toward AI-based decision support systems, and mentioned several barriers and enablers based on the experiences of previous implementations of information technology systems. Central aspects to consider for the proposed clinical decision support system were design aspects, access to information throughout the care process, and integration into the clinical workflow. The implementation of such a system could lead to a number of effects related to both clinical outcomes as well as resource allocation, which are all important to address in the planning of implementation. Stakeholders saw, however, value in several aspects of implementing such system, emphasizing the increased quality of life for those patients who can avoid being hospitalized.

CONCLUSIONS

Several ideas were put forward on how the proposed AI system would potentially affect and provide value for patients, professionals, and the organization, and implementation aspects were important parts of that. A successful system can help clinicians to prioritize the need for different types of treatments but also be used for planning purposes within the hospital. However, the system needs not only technological and clinical precision but also a carefully planned implementation process. Such a process should take into consideration the aspects related to all the categories in the NASSS framework. This study further highlighted the importance to study stakeholder needs early in the process of development, design, and implementation of decision support systems, as the data revealed new information on the potential use of the system and the placement of the application in the care process.

Implementation frameworks for end-to-end clinical AI: derivation of the SALIENT framework

OBJECTIVE

To derive a comprehensive implementation framework for clinical AI models within hospitals informed by existing AI frameworks and integrated with reporting standards for clinical AI research.

MATERIALS AND METHODS

(1) Derive a provisional implementation framework based on the taxonomy of Stead et al and integrated with current reporting standards for AI research: TRIPOD, DECIDE-AI, CONSORT-AI. (2) Undertake a scoping review of published clinical AI implementation frameworks and identify key themes and stages. (3) Perform a gap analysis and refine the framework by incorporating missing items.

RESULTS

The provisional AI implementation framework, called SALIENT, was mapped to 5 stages common to both the taxonomy and the reporting standards. A scoping review retrieved 20 studies and 247 themes, stages, and subelements were identified. A gap analysis identified 5 new cross-stage themes and 16 new tasks. The final framework comprised 5 stages, 7 elements, and 4 components, including the AI system, data pipeline, human-computer interface, and clinical workflow.

DISCUSSION

This pragmatic framework resolves gaps in existing stage- and theme-based clinical AI implementation guidance by comprehensively addressing the what (components), when (stages), and how (tasks) of AI implementation, as well as the who (organization) and why (policy domains). By integrating research reporting standards into SALIENT, the framework is grounded in rigorous evaluation methodologies. The framework requires validation as being applicable to real-world studies of deployed AI models.

CONCLUSIONS

A novel end-to-end framework has been developed for implementing AI within hospital clinical practice that builds on previous AI implementation frameworks and research reporting standards.

Understanding Mental Health Clinicians' Perceptions and Concerns Regarding Using Passive Patient-Generated Health Data for Clinical Decision-Making: Qualitative Semistructured Interview Study

BACKGROUND

Digital health-tracking tools are changing mental health care by giving patients the ability to collect passively measured patient-generated health data (PGHD; ie, data collected from connected devices with little to no patient effort). Although there are existing clinical guidelines for how mental health clinicians should use more traditional, active forms of PGHD for clinical decision-making, there is less clarity on how passive PGHD can be used.

OBJECTIVE

We conducted a qualitative study to understand mental health clinicians' perceptions and concerns regarding the use of technology-enabled, passively collected PGHD for clinical decision-making. Our interviews sought to understand participants' current experiences with and visions for using passive PGHD.

METHODS

Mental health clinicians providing outpatient services were recruited to participate in semistructured interviews. Interview recordings were deidentified, transcribed, and qualitatively coded to identify overarching themes.

RESULTS

Overall, 12 mental health clinicians (n=11, 92% psychiatrists and n=1, 8% clinical psychologist) were interviewed. We identified 4 overarching themes. First, passive PGHD are patient driven-we found that current passive PGHD use was patient driven, not clinician driven; participating clinicians only considered passive PGHD for clinical decision-making when patients brought passive data to clinical encounters. The second theme was active versus passive data as subjective versus objective data-participants viewed the contrast between active and passive PGHD as a contrast between interpretive data on patients' mental health and objective information on behavior. Participants believed that prioritizing passive over self-reported, active PGHD would reduce opportunities for patients to reflect upon their mental health, reducing treatment engagement and raising questions about how passive data can best complement active data for clinical decision-making. Third, passive PGHD must be delivered at appropriate times for action-participants were concerned with the real-time nature of passive PGHD; they believed that it would be infeasible to use passive PGHD for real-time patient monitoring outside clinical encounters and more feasible to use passive PGHD during clinical encounters when clinicians can make treatment decisions. The fourth theme was protecting patient privacy-participating clinicians wanted to protect patient privacy within passive PGHD-sharing programs and discussed opportunities to refine data sharing consent to improve transparency surrounding passive PGHD collection and use.

CONCLUSIONS

Although passive PGHD has the potential to enable more contextualized measurement, this study highlights the need for building and disseminating an evidence base describing how and when passive measures should be used for clinical decision-making. This evidence base should clarify how to use passive data alongside more traditional forms of active PGHD, when clinicians should view passive PGHD to make treatment decisions, and how to protect patient privacy within passive data-sharing programs. Clear evidence would more effectively support the uptake and effective use of these novel tools for both patients and their clinicians.

Implementation of Prediction Models in the Emergency Department from an Implementation Science Perspective-Determinants, Outcomes, and Real-World Impact: A Scoping Review

STUDY OBJECTIVE

Prediction models offer a promising form of clinical decision support in the complex and fast-paced environment of the emergency department (ED). Despite significant advancements in model development and validation, implementation of such models in routine clinical practice remains elusive. This scoping review aims to survey the current state of prediction model implementation in the ED and to provide insights on contributing factors and outcomes from an implementation science perspective.

METHODS

We searched 4 databases from their inception to May 20, 2022: MEDLINE (through PubMed), Embase, Scopus, and CINAHL. Articles that reported implementation outcomes and/or contextual determinants under the Reach, Effectiveness, Adoption, Implementation Maintenance (RE-AIM)/Practical, Robust, Implementation, and Sustainability Model (PRISM) framework were included. Characteristics of studies, models, and results of the RE-AIM/PRISM domains were summarized narratively.

RESULTS

Thirty-six reports on 31 implementations were included. The most common prediction models implemented were early warning scores. The most common implementation strategies used were training stakeholders, infrastructural changes, and using evaluative or iterative strategies. Only one report examined ED patients' perspectives, whereas the rest were focused on the experience of health care workers or organizational stakeholders. Key determinants of successful implementation include strong stakeholder engagement, codevelopment of workflows and implementation strategies, education, and usability.

CONCLUSION

Examining ED prediction models from an implementation science perspective can provide valuable insights and help guide future implementations.

Deployment of machine learning algorithms to predict sepsis: systematic review and application of the SALIENT clinical AI implementation framework

OBJECTIVE

To retrieve and appraise studies of deployed artificial intelligence (AI)-based sepsis prediction algorithms using systematic methods, identify implementation barriers, enablers, and key decisions and then map these to a novel end-to-end clinical AI implementation framework.

MATERIALS AND METHODS

Systematically review studies of clinically applied AI-based sepsis prediction algorithms in regard to methodological quality, deployment and evaluation methods, and outcomes. Identify contextual factors that influence implementation and map these factors to the SALIENT implementation framework.

RESULTS

The review identified 30 articles of algorithms applied in adult hospital settings, with 5 studies reporting significantly decreased mortality post-implementation. Eight groups of algorithms were identified, each sharing a common algorithm. We identified 14 barriers, 26 enablers, and 22 decision points which were able to be mapped to the 5 stages of the SALIENT implementation framework.

DISCUSSION

Empirical studies of deployed sepsis prediction algorithms demonstrate their potential for improving care and reducing mortality but reveal persisting gaps in existing implementation guidance. In the examined publications, key decision points reflecting real-word implementation experience could be mapped to the SALIENT framework and, as these decision points appear to be AI-task agnostic, this framework may also be applicable to non-sepsis algorithms. The mapping clarified where and when barriers, enablers, and key decisions arise within the end-to-end AI implementation process.

CONCLUSIONS

A systematic review of real-world implementation studies of sepsis prediction algorithms was used to validate an end-to-end staged implementation framework that has the ability to account for key factors that warrant attention in ensuring successful deployment, and which extends on previous AI implementation frameworks.

Developing and Validating a Prediction Model For Death or Critical Illness in Hospitalized Adults, an Opportunity for Human-Computer Collaboration

Hospital early warning systems that use machine learning (ML) to predict clinical deterioration are increasingly being used to aid clinical decision-making. However, it is not known how ML predictions complement physician and nurse judgment. Our objective was to train and validate a ML model to predict patient deterioration and compare model predictions with real-world physician and nurse predictions.

DESIGN

Retrospective and prospective cohort study.

SETTING

Academic tertiary care hospital.

PATIENTS

Adult general internal medicine hospitalizations.

MEASUREMENTS AND MAIN RESULTS

We developed and validated a neural network model to predict in-hospital death and ICU admission in 23,528 hospitalizations between April 2011 and April 2019. We then compared model predictions with 3,374 prospectively collected predictions from nurses, residents, and attending physicians about their own patients in 960 hospitalizations between April 30, and August 28, 2019. ML model predictions achieved clinician-level accuracy for predicting ICU admission or death (ML median F1 score 0.32 [interquartile range (IQR) 0.30-0.34], AUC 0.77 [IQ 0.76-0.78]; clinicians median F1-score 0.33 [IQR 0.30-0.35], AUC 0.64 [IQR 0.63-0.66]). ML predictions were more accurate than clinicians for ICU admission. Of all ICU admissions and deaths, 36% occurred in hospitalizations where the model and clinicians disagreed. Combining human and model predictions detected 49% of clinical deterioration events, improving sensitivity by 16% compared with clinicians alone and 24% compared with the model alone while maintaining a positive predictive value of 33%, thus keeping false alarms at a clinically acceptable level.

CONCLUSIONS

ML models can complement clinician judgment to predict clinical deterioration in hospital. These findings demonstrate important opportunities for human-computer collaboration to improve prognostication and personalized medicine in hospital.

Accelerating health system innovation: principles and practices from the Duke Institute for Health Innovation

The Duke Institute for Health Innovation (DIHI) was launched in 2013. Frontline staff members submit proposals for innovation projects that align with strategic priorities set by organizational leadership. Funded projects receive operational and technical support from institute staff members and a transdisciplinary network of collaborators to develop and implement solutions as part of routine clinical care, ranging from machine learning algorithms to mobile applications. DIHI's operations are shaped by four guiding principles: build to show value, build to integrate, build to scale, and build responsibly. Between 2013 and 2021, more than 600 project proposals have been submitted to DIHI. More than 85 innovation projects, both through the application process and other strategic partnerships, have been supported and implemented. DIHI's funding has incubated 12 companies, engaged more than 300 faculty members, staff members, and students, and contributed to more than 50 peer-reviewed publications. DIHI's practices can serve as a model for other health systems to systematically source, develop, implement, and scale innovations.

Usability and Clinician Acceptance of a Deep Learning-Based Clinical Decision Support Tool for Predicting Glaucomatous Visual Field Progression

PRCIS

We updated a clinical decision support tool integrating predicted visual field (VF) metrics from an artificial intelligence model and assessed clinician perceptions of the predicted VF metric in this usability study.

PURPOSE

To evaluate clinician perceptions of a prototyped clinical decision support (CDS) tool that integrates visual field (VF) metric predictions from artificial intelligence (AI) models.

METHODS

Ten ophthalmologists and optometrists from the University of California San Diego participated in 6 cases from 6 patients, consisting of 11 eyes, uploaded to a CDS tool ("GLANCE", designed to help clinicians "at a glance"). For each case, clinicians answered questions about management recommendations and attitudes towards GLANCE, particularly regarding the utility and trustworthiness of the AI-predicted VF metrics and willingness to decrease VF testing frequency.

MAIN OUTCOMES AND MEASURES

Mean counts of management recommendations and mean Likert scale scores were calculated to assess overall management trends and attitudes towards the CDS tool for each case. In addition, system usability scale scores were calculated.

RESULTS

The mean Likert scores for trust in and utility of the predicted VF metric and clinician willingness to decrease VF testing frequency were 3.27, 3.42, and 2.64, respectively (1=strongly disagree, 5=strongly agree). When stratified by glaucoma severity, all mean Likert scores decreased as severity increased. The system usability scale score across all responders was 66.1±16.0 (43rd percentile).

CONCLUSIONS

A CDS tool can be designed to present AI model outputs in a useful, trustworthy manner that clinicians are generally willing to integrate into their clinical decision-making. Future work is needed to understand how to best develop explainable and trustworthy CDS tools integrating AI before clinical deployment.

Clinical Artificial Intelligence: Design Principles and Fallacies

Clinical artificial intelligence (AI)/machine learning (ML) is anticipated to offer new abilities in clinical decision support, diagnostic reasoning, precision medicine, clinical operational support, and clinical research, but careful concern is needed to ensure these technologies work effectively in the clinic. Here, we detail the clinical ML/AI design process, identifying several key questions and detailing several common forms of issues that arise with ML tools, as motivated by real-world examples, such that clinicians and researchers can better anticipate and correct for such issues in their own use of ML/AI techniques.

Finding undiagnosed patients with hepatitis C virus: an application of machine learning to US ambulatory electronic medical records

AIMS

To develop and validate a machine learning (ML) algorithm to identify undiagnosed hepatitis C virus (HCV) patients, in order to facilitate prioritisation of patients for targeted HCV screening.

METHODS

This retrospective study used ambulatory electronic medical records (EMR) from January 2015 to February 2020. A Gradient Boosting Trees algorithm was trained using patient records to predict initial HCV diagnosis and was validated on a temporally independent held-out cross-section of the data. The fold improvement in precision (proportion of patients identified by the algorithm who are HCV positive) over universal screening was examined and compared with risk-based screening.

RESULTS

21 508 positive (HCV diagnosed) and 28.2M unlabelled (lacking evidence of HCV diagnosis) patients met the inclusion criteria for the study. After down-sampling unlabelled patients to aid the algorithm's learning process, 16.2M unlabelled patients entered the analysis. Performance of the algorithm was compared with universal screening on the held-out cross-section, which had an incidence of HCV diagnoses of 0.02%. The algorithm achieved a 101.0 ×, 18.0 × and 5.1 × fold improvement in precision over universal screening at 5%, 20% and 50% levels of recall. When compared with risk-based screening, the algorithm required fewer patients to be screened and improved precision.

CONCLUSIONS

This study presents strong evidence towards the use of ML on EMR data for the prioritisation of patients for targeted HCV testing with potential to improve efficiency of resource utilisation, thereby reducing the workload for clinicians and saving healthcare costs. A prospective interventional study would allow for further validation before use in a clinical setting.

A smarter perspective: Learning with and from AI-cases

Artificial intelligence (AI) has only partially (or not at all) been integrated into medical education, leading to growing concerns regarding how to train healthcare practitioners to handle the changes brought about by the introduction of AI. Programming lessons and other technical information into healthcare curricula has been proposed as a solution to support healthcare personnel in using AI or other future technology. However, integrating these core elements of computer science knowledge might not meet the observed need that students will benefit from gaining practical experience with AI in the direct application area. Therefore, this paper proposes a dynamic approach to case-based learning that utilizes the scenarios where AI is currently used in clinical practice as examples. This approach will support students' understanding of technical aspects. Case-based learning with AI as an example provides additional benefits: (1) it allows doctors to compare their thought processes to the AI suggestions and critically reflect on the assumptions and biases of AI and clinical practice; (2) it incentivizes doctors to discuss and address ethical issues inherent to technology and those already existing in current clinical practice; (3) it serves as a foundation for fostering interdisciplinary collaboration via discussion of different views between technologists, multidisciplinary experts, and healthcare professionals. The proposed knowledge shift from AI as a technical focus to AI as an example for case-based learning aims to encourage a different perspective on educational needs. Technical education does not need to compete with other essential clinical skills as it could serve as a basis for supporting them, which leads to better medical education and practice, ultimately benefiting patients.

Computer clinical decision support that automates personalized clinical care: a challenging but needed healthcare delivery strategy

How to deliver best care in various clinical settings remains a vexing problem. All pertinent healthcare-related questions have not, cannot, and will not be addressable with costly time- and resource-consuming controlled clinical trials. At present, evidence-based guidelines can address only a small fraction of the types of care that clinicians deliver. Furthermore, underserved areas rarely can access state-of-the-art evidence-based guidelines in real-time, and often lack the wherewithal to implement advanced guidelines. Care providers in such settings frequently do not have sufficient training to undertake advanced guideline implementation. Nevertheless, in advanced modern healthcare delivery environments, use of eActions (validated clinical decision support systems) could help overcome the cognitive limitations of overburdened clinicians. Widespread use of eActions will require surmounting current healthcare technical and cultural barriers and installing clinical evidence/data curation systems. The authors expect that increased numbers of evidence-based guidelines will result from future comparative effectiveness clinical research carried out during routine healthcare delivery within learning healthcare systems.

Artificial Intelligence Implementation in Healthcare: A Theory-Based Scoping Review of Barriers and Facilitators

There is a large proliferation of complex data-driven artificial intelligence (AI) applications in many aspects of our daily lives, but their implementation in healthcare is still limited. This scoping review takes a theoretical approach to examine the barriers and facilitators based on empirical data from existing implementations. We searched the major databases of relevant scientific publications for articles related to AI in clinical settings, published between 2015 and 2021. Based on the theoretical constructs of the Consolidated Framework for Implementation Research (CFIR), we used a deductive, followed by an inductive, approach to extract facilitators and barriers. After screening 2784 studies, 19 studies were included in this review. Most of the cited facilitators were related to engagement with and management of the implementation process, while the most cited barriers dealt with the intervention's generalizability and interoperability with existing systems, as well as the inner settings' data quality and availability. We noted per-study imbalances related to the reporting of the theoretic domains. Our findings suggest a greater need for implementation science expertise in AI implementation projects, to improve both the implementation process and the quality of scientific reporting.

Invisible clinical labor driving the successful integration of AI in healthcare

Artificial Intelligence and Machine Learning (AI/ML) tools are changing the landscape of healthcare decision-making. Vast amounts of data can lead to efficient triage and diagnosis of patients with the assistance of ML methodologies. However, more research has focused on the technological challenges of developing AI, rather than the system integration. As a result, clinical teams' role in developing and deploying these tools has been overlooked. We look to three case studies from our research to describe the often invisible work that clinical teams do in driving the successful integration of clinical AI tools. Namely, clinical teams support data labeling, identifying algorithmic errors and accounting for workflow exceptions, translating algorithmic output to clinical next steps in care, and developing team awareness of how the tool is used once deployed. We call for detailed and extensive documentation strategies (of clinical labor, workflows, and team structures) to ensure this labor is valued and to promote sharing of sociotechnical implementation strategies.

Advanced interventions in the pre-hospital resuscitation of patients with non-compressible haemorrhage after penetrating injuries

Abstract

Early haemorrhage control and minimizing the time to definitive care have long been the cornerstones of therapy for patients exsanguinating from non-compressible haemorrhage (NCH) after penetrating injuries, as only basic treatment could be provided on scene. However, more recently, advanced on-scene treatments such as the transfusion of blood products, resuscitative thoracotomy (RT) and resuscitative endovascular balloon occlusion of the aorta (REBOA) have become available in a small number of pre-hospital critical care teams. Although these advanced techniques are included in the current traumatic cardiac arrest algorithm of the European Resuscitation Council (ERC), published in 2021, clear guidance on the practical application of these techniques in the pre-hospital setting is scarce. This paper provides a scoping review on how these advanced techniques can be incorporated into practice for the resuscitation of patients exsanguinating from NCH after penetrating injuries, based on available literature and the collective experience of several helicopter emergency medical services (HEMS) across Europe who have introduced these advanced resuscitation interventions into routine practice.

Graphical Abstract