Artificial Intelligence Applications in Health Care Practice: A Scoping Review (Preprint)

Current Implementation of Digital Health in Chronic Disease Management: Scoping Review

BACKGROUND

Approximately 1 in 3 adults live with multiple chronic diseases. Digital health is being harnessed to improve continuity of care and management of chronic diseases. However, meaningful uptake of digital health for chronic disease management remains low. It is unclear how these innovations have been implemented and evaluated.

OBJECTIVE

This scoping review aims to identify how digital health innovations for chronic disease management have been implemented and evaluated: what implementation frameworks, methods, and strategies were used; how successful these strategies were; key barriers and enablers to implementation; and lessons learned and recommendations shared by study authors.

METHODS

We used the Joanna Briggs Institute methodology for scoping reviews. Five databases were searched for studies published between January 2015 and March 2023: PubMed, Scopus, CINAHL, PsycINFO, and IEEE Xplore. We included primary studies of any study design with any type of digital health innovations for chronic diseases that benefit patients, caregivers, or health care professionals. We extracted study characteristics; type of digital health innovation; implementation frameworks, strategies, and outcome measures used; barriers and enablers to implementation; lessons learned; and recommendations reported by study authors. We used established taxonomies to synthesize extracted data. Extracted barriers and enablers were grouped into categories for reporting. Descriptive statistics were used to consolidate extracted data.

RESULTS

A total of 252 studies were included, comprising mainly mobile health (107/252, 42.5%), eHealth (61/252, 24.2%), and telehealth (97/252, 38.5%), with some studies involving more than 1 innovation. Only 23 studies (23/252, 9.1%) reported using an implementation science theory, model, or framework; the most common were implementation theories, classic theories, and determinant frameworks, with 7 studies each. Of 252 studies, 144 (57.1%) used 2 to 5 implementation strategies. Frequently used strategies were "obtain and use patient or consumer feedback" (196/252, 77.8%); "audit and provide feedback" (106/252, 42.1%); and piloting before implementation or "stage implementation scale-up" (85/252, 33.7%). Commonly measured implementation outcomes were acceptability, feasibility, and adoption of the digital innovation. Of 252 studies, 247 studies (98%) did not measure service outcomes, while patient health outcomes were measured in 89 studies (35.3%). The main method used to assess outcomes was surveys (173/252, 68.7%), followed by interviews (95/252, 37.7%). Key barriers impacting implementation were data privacy concerns and patient preference for in-person consultations. Key enablers were training for health care workers and personalization of digital health features to patient needs.

CONCLUSIONS

This review generated a summary of how digital health in chronic disease management is currently implemented and evaluated and serves as a useful resource for clinicians, researchers, health system managers, and policy makers planning real-world implementation. Future studies should investigate whether using implementation science frameworks, including how well they are used, would yield better outcomes compared to not using them.

Assessing Visitor Expectations of AI Nursing Robots in Hospital Settings: Cross-Sectional Study Using the Kano Model

Background

Globally, the rates at which the aging population and the prevalence of chronic diseases are increasing are substantial. With declining birth rates and a growing percentage of older individuals, the demand for nursing staff is steadily rising. However, the shortage of nursing personnel has been a long-standing issue. In recent years, numerous researchers have advocated for the implementation of nursing robots as a substitute for traditional human labor.

Objective

This study analyzes hospital visitors' attitudes and priorities regarding the functional areas of artificial intelligence (AI) nursing robots based on the Kano model. Building on this analysis, recommendations are provided for the functional optimization of AI nursing robots, aiming to facilitate their adoption in the nursing field.

Methods

Using a random sampling method, 457 hospital visitors were surveyed between December 2023 and March 2024 to compare the differences in demand for AI nursing robot functionalities among the visitors.

Results

A comparative analysis of the Kano attribute quadrant diagrams showed that visitors seeking hospitalization prioritized functional aspects that enhance medical activities. In contrast, visitors attending outpatient examinations focused more on functional points that assist in medical treatment. Additionally, visitors whose purpose was companionship and care emphasized functional aspects that offer psychological and life support to patients.

Conclusions

AI nursing robots serve various functional areas and cater to diverse audience groups. In the future, it is essential to thoroughly consider users' functional needs and implement targeted functional developments to maximize the effectiveness of AI nursing robots.

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.

Exploring geriatric nurses' perspectives on the adoption of AI in elderly care a qualitative study

This phenomenological study explored the perspectives of geriatric nurses on the adoption of artificial intelligence (AI) in elderly care. Thematic analysis of semi-structured interviews with 17 nurses revealed perceived benefits, challenges, ethical considerations, and practical implications. Participants acknowledged AI's potential for improving diagnostic accuracy, personalized care, continuous monitoring, and data pattern insights. However, concerns were raised regarding workflow integration, cost barriers, resistance to change, data privacy, diminishment of human elements, and the need for ethical guidelines. A cautious optimism was expressed, emphasizing the importance of addressing practical challenges, maintaining the human touch, and fostering a collaborative approach. The findings highlight the need for comprehensive training, user-centered design, ethical frameworks, and strategies to overcome financial and implementation barriers. Future research should focus on evaluating the impact of AI implementation on patient outcomes and nursing experiences.

Definitions and Characteristics of Patient Digital Twins Being Developed for Clinical Use: Scoping Review

BACKGROUND

The concept of digital twins, widely adopted in industry, is entering health care. However, there is a lack of consensus on what constitutes the digital twin of a patient.

OBJECTIVE

The objective of this scoping review was to analyze definitions and characteristics of patient digital twins being developed for clinical use, as reported in the scientific literature.

METHODS

We searched PubMed, Scopus, Embase, IEEE, and Google Scholar for studies claiming digital twin development or evaluation until August 2023. Data on definitions, characteristics, and development phase were extracted. Unsupervised classification of claimed digital twins was performed.

RESULTS

We identified 86 papers representing 80 unique claimed digital twins, with 98% (78/80) in preclinical phases. Among the 55 papers defining "digital twin," 76% (42/55) described a digital replica, 42% (23/55) mentioned real-time updates, 24% (13/55) emphasized patient specificity, and 15% (8/55) included 2-way communication. Among claimed digital twins, 60% (48/80) represented specific organs (primarily heart: 15/48, 31%; bones or joints: 10/48, 21%; lung: 6/48, 12%; and arteries: 5/48, 10%); 14% (11/80) embodied biological systems such as the immune system; and 26% (21/80) corresponded to other products (prediction models, etc). The patient data used to develop and run the claimed digital twins encompassed medical imaging examinations (35/80, 44% of publications), clinical notes (15/80, 19% of publications), laboratory test results (13/80, 16% of publications), wearable device data (12/80, 15% of publications), and other modalities (32/80, 40% of publications). Regarding data flow between patients and their virtual counterparts, 16% (13/80) claimed that digital twins involved no flow from patient to digital twin, 73% (58/80) used 1-way flow from patient to digital twin, and 11% (9/80) enabled 2-way data flow between patient and digital twin. Based on these characteristics, unsupervised classification revealed 3 clusters: simulation patient digital twins in 54% (43/80) of publications, monitoring patient digital twins in 28% (22/80) of publications, and research-oriented models unlinked to specific patients in 19% (15/80) of publications. Simulation patient digital twins used computational modeling for personalized predictions and therapy evaluations, mostly for one-time assessments, and monitoring digital twins harnessed aggregated patient data for continuous risk or outcome forecasting and care optimization.

CONCLUSIONS

We propose defining a patient digital twin as "a viewable digital replica of a patient, organ, or biological system that contains multidimensional, patient-specific information and informs decisions" and to distinguish simulation and monitoring digital twins. These proposed definitions and subtypes offer a framework to guide research into realizing the potential of these personalized, integrative technologies to advance clinical care.

A Narrative Review of Ethical Issues in the Use of Artificial Intelligence Enabled Diagnostics for Diabetic Retinopathy

INTRODUCTION

Diabetic retinopathy is one of the leading causes of avoidable blindness among adults globally, and screening programmes can enable early diagnosis and prevention of progression. Artificial intelligence (AI) diagnostic solutions have been developed to diagnose diabetic retinopathy. The aim of this review is to identify ethical concerns related to AI-enabled diabetic retinopathy diagnostics and enable future research to explore these issues further.

METHODS

This is a narrative review that uses thematic analysis methods to develop key findings. We searched two databases, PubMed and Scopus, for papers focused on the intersection of AI, diagnostics, ethics, and diabetic retinopathy and conducted a citation search. Primary research articles published in English between 1 January 2013 and 14 June 2024 were included. From the 1878 papers that were screened, nine papers met inclusion and exclusion criteria and were selected for analysis.

RESULTS

We found that existing literature highlights ensuring patient data has appropriate protection and ownership, that bias in algorithm training data is minimised, informed patient decision-making is encouraged, and negative consequences in the context of clinical practice are mitigated.

CONCLUSIONS

While the technical developments in AI-enabled diabetic retinopathy diagnostics receive the bulk of the research focus, we found that insufficient attention is paid to how this technology is accessed equitably in different settings and which safeguards are needed against exploitative practices. Such ethical issues merit additional exploration and practical problem-solving through primary research. AI-enabled diabetic retinopathy screening has the potential to enable screening at a scale that was previously not possible and could contribute to reducing preventable blindness. It will only achieve this if ethical issues are emphasised, understood, and addressed throughout the translation of this technology to clinical practice.

Studies of Artificial Intelligence/Machine Learning Registered on ClinicalTrials.gov: Cross-Sectional Study With Temporal Trends, 2010-2023

BACKGROUND

The rapid growth of research in artificial intelligence (AI) and machine learning (ML) continues. However, it is unclear whether this growth reflects an increase in desirable study attributes or merely perpetuates the same issues previously raised in the literature.

OBJECTIVE

This study aims to evaluate temporal trends in AI/ML studies over time and identify variations that are not apparent from aggregated totals at a single point in time.

METHODS

We identified AI/ML studies registered on ClinicalTrials.gov with start dates between January 1, 2010, and December 31, 2023. Studies were included if AI/ML-specific terms appeared in the official title, detailed description, brief summary, intervention, primary outcome, or sponsors' keywords. Studies registered as systematic reviews and meta-analyses were excluded. We reported trends in AI/ML studies over time, along with study characteristics that were fast-growing and those that remained unchanged during 2010-2023.

RESULTS

Of 3106 AI/ML studies, only 7.6% (n=235) were regulated by the US Food and Drug Administration. The most common study characteristics were randomized (56.2%; 670/1193; interventional) and prospective (58.9%; 1126/1913; observational) designs; a focus on diagnosis (28.2%; 335/1190) and treatment (24.4%; 290/1190); hospital/clinic (44.2%; 1373/3106) or academic (28%; 869/3106) sponsorship; and neoplasm (12.9%; 420/3245), nervous system (12.2%; 395/3245), cardiovascular (11.1%; 356/3245) or pathological conditions (10%; 325/3245; multiple counts per study possible). Enrollment data were skewed to the right: maximum 13,977,257; mean 16,962 (SD 288,155); median 255 (IQR 80-1000). The most common size category was 101-1000 (44.8%; 1372/3061; excluding withdrawn or missing), but large studies (n>1000) represented 24.1% (738/3061) of all studies: 29% (551/1898) of observational studies and 16.1% (187/1163) of trials. Study locations were predominantly in high-income countries (75.3%; 2340/3106), followed by upper-middle-income (21.7%; 675/3106), lower-middle-income (2.8%; 88/3106), and low-income countries (0.1%; 3/3106). The fastest-growing characteristics over time were high-income countries (location); Europe, Asia, and North America (location); diagnosis and treatment (primary purpose); hospital/clinic and academia (lead sponsor); randomized and prospective designs; and the 1-100 and 101-1000 size categories. Only 5.6% (47/842) of completed studies had results available on ClinicalTrials.gov, and this pattern persisted. Over time, there was an increase in not only the number of newly initiated studies, but also the number of completed studies without posted results.

CONCLUSIONS

Much of the rapid growth in AI/ML studies comes from high-income countries in high-resource settings, albeit with a modest increase in upper-middle-income countries (mostly China). Lower-middle-income or low-income countries remain poorly represented. The increase in randomized or prospective designs, along with 738 large studies (n>1000), mostly ongoing, may indicate that enough studies are shifting from an in silico evaluation stage toward a prospective comparative evaluation stage. However, the ongoing limited availability of basic results on ClinicalTrials.gov contrasts with this field's rapid advancements and the public registry's role in reducing publication and outcome reporting biases.

Mapping the Evolution of Digital Health Research: Bibliometric Overview of Research Hotspots, Trends, and Collaboration of Publications in JMIR (1999-2024)

BACKGROUND

While bibliometric studies of individual journals have been conducted, to the best of our knowledge, bibliometric mapping has not yet been utilized to analyze the literature published by the Journal of Medical Internet Research (JMIR).

OBJECTIVE

In celebration of the journal's 25th anniversary, this study aimed to review the entire collection of JMIR publications from 1999 to 2024 and provide a comprehensive overview of the main publication characteristics.

METHODS

This study included papers published in JMIR during the 25-year period from 1999 to 2024. The data were analyzed using CiteSpace, VOSviewer, and the "Bibliometrix" package in R. Through descriptive bibliometrics, we examined the dynamics and trend patterns of JMIR literature production and identified the most prolific authors, papers, institutions, and countries. Bibliometric maps were used to visualize the content of published articles and to identify the most prominent research terms and topics, along with their evolution. A bibliometric network map was constructed to determine the hot research topics over the past 25 years.

RESULTS

This study revealed positive trends in literature production, with both the total number of publications and the average number of citations increasing over the years. And the global COVID-19 pandemic induced an explosive rise in the number of publications in JMIR. The most productive institutions were predominantly from the United States, which ranked highest in successful publications within the journal. The editor-in-chief of JMIR was identified as a pioneer in this field. The thematic analysis indicated that the most prolific topics aligned with the primary aims and scope of the journal. Currently and in the foreseeable future, the main themes of JMIR include "artificial intelligence," "patient empowerment," and "victimization."

CONCLUSIONS

This bibliometric study highlighted significant contributions to digital health by identifying key research trends, themes, influential authors, and collaborations. The findings underscore the necessity to enhance publications from developing countries, improve gender diversity among authors, and expand the range of research topics explored in the journal.

Reconfiguration of uncertainty: Introducing AI for prediction of mortality at the emergency department

The promise behind many advanced digital technologies in healthcare is to provide novel and accurate information, aiding medical experts to navigate and, ultimately, decrease uncertainty in their clinical work. However, sociological studies have started to show that these technologies are not producing straightforward objective knowledge, but instead often become associated with new uncertainties arising in unanticipated places and situations. This study contributes to the body of work by presenting a qualitative study of an Artificial Intelligence (AI) algorithm designed to predict the risk of mortality in patients discharged to home from the emergency department (ED). Through in-depth interviews with physicians working at the ED of a Swedish hospital, we demonstrate that while the AI algorithm can reduce targeted uncertainty, it simultaneously introduces three new forms of uncertainty into clinical practice: epistemic uncertainty, actionable uncertainty and ethical uncertainty. These new uncertainties require deliberate management and control, marking a shift from the physicians' accustomed comfort with uncertainty in mortality prediction. Our study advances the understanding of the recursive nature and temporal dynamics of uncertainty in medical work, showing how new uncertainties emerge from attempts to manage existing ones. It also reveals that physicians' attitudes towards, and management of, uncertainty vary depending on its form and underscores the intertwined role of digital technology in this process. By examining AI in emergency care, we provide valuable insights into how this epistemic technology reconfigures clinical uncertainty, offering significant theoretical and practical implications for the integration of AI in healthcare.

Systematic Review of the Application of Artificial Intelligence in Healthcare and Nursing Care

This systematic review explores the complex relationship between artificial intelligence (AI) and healthcare, with an explicit focus on nursing care. Examining a range of studies from 2020, the research investigates the impact of AI on clinical decision-making, patient care and healthcare administration. Through a comprehensive literature review, the study highlights the potential benefits of AI integration in improving the efficiency and efficacy of healthcare. AI technologies offer opportunities for personalised patient care, predictive analytics and enhanced clinical processes, with the ultimate aim of transforming the healthcare system. However, ethical considerations and regulatory frameworks are crucial, emphasising patient privacy, autonomy and data security. The findings underscore the need for transparency, accountability and fairness in the application of AI in healthcare. While AI promises to improve patient outcomes and streamline healthcare delivery, careful consideration of ethical implications and regulatory compliance are essential for responsible implementation.

Early Prediction of Cardiac Arrest in the Intensive Care Unit Using Explainable Machine Learning: Retrospective Study

BACKGROUND

Cardiac arrest (CA) is one of the leading causes of death among patients in the intensive care unit (ICU). Although many CA prediction models with high sensitivity have been developed to anticipate CA, their practical application has been challenging due to a lack of generalization and validation. Additionally, the heterogeneity among patients in different ICU subtypes has not been adequately addressed.

OBJECTIVE

This study aims to propose a clinically interpretable ensemble approach for the timely and accurate prediction of CA within 24 hours, regardless of patient heterogeneity, including variations across different populations and ICU subtypes. Additionally, we conducted patient-independent evaluations to emphasize the model's generalization performance and analyzed interpretable results that can be readily adopted by clinicians in real-time.

METHODS

Patients were retrospectively analyzed using data from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) and the eICU-Collaborative Research Database (eICU-CRD). To address the problem of underperformance, we constructed our framework using feature sets based on vital signs, multiresolution statistical analysis, and the Gini index, with a 12-hour window to capture the unique characteristics of CA. We extracted 3 types of features from each database to compare the performance of CA prediction between high-risk patient groups from MIMIC-IV and patients without CA from eICU-CRD. After feature extraction, we developed a tabular network (TabNet) model using feature screening with cost-sensitive learning. To assess real-time CA prediction performance, we used 10-fold leave-one-patient-out cross-validation and a cross-data set method. We evaluated MIMIC-IV and eICU-CRD across different cohort populations and subtypes of ICU within each database. Finally, external validation using the eICU-CRD and MIMIC-IV databases was conducted to assess the model's generalization ability. The decision mask of the proposed method was used to capture the interpretability of the model.

RESULTS

The proposed method outperformed conventional approaches across different cohort populations in both MIMIC-IV and eICU-CRD. Additionally, it achieved higher accuracy than baseline models for various ICU subtypes within both databases. The interpretable prediction results can enhance clinicians' understanding of CA prediction by serving as a statistical comparison between non-CA and CA groups. Next, we tested the eICU-CRD and MIMIC-IV data sets using models trained on MIMIC-IV and eICU-CRD, respectively, to evaluate generalization ability. The results demonstrated superior performance compared with baseline models.

CONCLUSIONS

Our novel framework for learning unique features provides stable predictive power across different ICU environments. Most of the interpretable global information reveals statistical differences between CA and non-CA groups, demonstrating its utility as an indicator for clinical decisions. Consequently, the proposed CA prediction system is a clinically validated algorithm that enables clinicians to intervene early based on CA prediction information and can be applied to clinical trials in digital health.

The Diagnostic Ability of GPT-3.5 and GPT-4.0 in Surgery: Comparative Analysis

BACKGROUND

ChatGPT (OpenAI) has shown great potential in clinical diagnosis and could become an excellent auxiliary tool in clinical practice. This study investigates and evaluates ChatGPT in diagnostic capabilities by comparing the performance of GPT-3.5 and GPT-4.0 across model iterations.

OBJECTIVE

This study aims to evaluate the precise diagnostic ability of GPT-3.5 and GPT-4.0 for colon cancer and its potential as an auxiliary diagnostic tool for surgeons and compare the diagnostic accuracy rates between GTP-3.5 and GPT-4.0. We precisely assess the accuracy of primary and secondary diagnoses and analyze the causes of misdiagnoses in GPT-3.5 and GPT-4.0 according to 7 categories: patient histories, symptoms, physical signs, laboratory examinations, imaging examinations, pathological examinations, and intraoperative findings.

METHODS

We retrieved 316 case reports for intestinal cancer from the Chinese Medical Association Publishing House database, of which 286 cases were deemed valid after data cleansing. The cases were translated from Mandarin to English and then input into GPT-3.5 and GPT-4.0 using a simple, direct prompt to elicit primary and secondary diagnoses. We conducted a comparative study to evaluate the diagnostic accuracy of GPT-4.0 and GPT-3.5. Three senior surgeons from the General Surgery Department, specializing in Colorectal Surgery, assessed the diagnostic information at the Chinese PLA (People's Liberation Army) General Hospital. The accuracy of primary and secondary diagnoses was scored based on predefined criteria. Additionally, we analyzed and compared the causes of misdiagnoses in both models according to 7 categories: patient histories, symptoms, physical signs, laboratory examinations, imaging examinations, pathological examinations, and intraoperative findings.

RESULTS

Out of 286 cases, GPT-4.0 and GPT-3.5 both demonstrated high diagnostic accuracy for primary diagnoses, but the accuracy rates of GPT-4.0 were significantly higher than GPT-3.5 (mean 0.972, SD 0.137 vs mean 0.855, SD 0.335; t285=5.753; P<.001). For secondary diagnoses, the accuracy rates of GPT-4.0 were also significantly higher than GPT-3.5 (mean 0.908, SD 0.159 vs mean 0.617, SD 0.349; t285=-7.727; P<.001). GPT-3.5 showed limitations in processing patient history, symptom presentation, laboratory tests, and imaging data. While GPT-4.0 improved upon GPT-3.5, it still has limitations in identifying symptoms and laboratory test data. For both primary and secondary diagnoses, there was no significant difference in accuracy related to age, gender, or system group between GPT-4.0 and GPT-3.5.

CONCLUSIONS

This study demonstrates that ChatGPT, particularly GPT-4.0, possesses significant diagnostic potential, with GPT-4.0 exhibiting higher accuracy than GPT-3.5. However, GPT-4.0 still has limitations, particularly in recognizing patient symptoms and laboratory data, indicating a need for more research in real-world clinical settings to enhance its diagnostic capabilities.

Assessing accuracy of ChatGPT in response to questions from day to day pharmaceutical care in hospitals

Background

The advent of Large Language Models (LLMs) such as ChatGPT introduces opportunities within the medical field. Nonetheless, use of LLM poses a risk when healthcare practitioners and patients present clinical questions to these programs without a comprehensive understanding of its suitability for clinical contexts.

Objective

The objective of this study was to assess ChatGPT's ability to generate appropriate responses to clinical questions that hospital pharmacists could encounter during routine patient care.

Methods

Thirty questions from 10 different domains within clinical pharmacy were collected during routine care. Questions were presented to ChatGPT in a standardized format, including patients' age, sex, drug name, dose, and indication. Subsequently, relevant information regarding specific cases were provided, and the prompt was concluded with the query "what would a hospital pharmacist do?". The impact on accuracy was assessed for each domain by modifying personification to "what would you do?", presenting the question in Dutch, and regenerating the primary question. All responses were independently evaluated by two senior hospital pharmacists, focusing on the availability of an advice, accuracy and concordance.

Results

In 77% of questions, ChatGPT provided an advice in response to the question. For these responses, accuracy and concordance were determined. Accuracy was correct and complete for 26% of responses, correct but incomplete for 22% of responses, partially correct and partially incorrect for 30% of responses and completely incorrect for 22% of responses. The reproducibility was poor, with merely 10% of responses remaining consistent upon regeneration of the primary question.

Conclusions

While concordance of responses was excellent, the accuracy and reproducibility were poor. With the described method, ChatGPT should not be used to address questions encountered by hospital pharmacists during their shifts. However, it is important to acknowledge the limitations of our methodology, including potential biases, which may have influenced the findings.

Leadership for AI Transformation in Health Care Organization: Scoping Review

BACKGROUND

The leaders of health care organizations are grappling with rising expenses and surging demands for health services. In response, they are increasingly embracing artificial intelligence (AI) technologies to improve patient care delivery, alleviate operational burdens, and efficiently improve health care safety and quality.

OBJECTIVE

In this paper, we map the current literature and synthesize insights on the role of leadership in driving AI transformation within health care organizations.

METHODS

We conducted a comprehensive search across several databases, including MEDLINE (via Ovid), PsycINFO (via Ovid), CINAHL (via EBSCO), Business Source Premier (via EBSCO), and Canadian Business & Current Affairs (via ProQuest), spanning articles published from 2015 to June 2023 discussing AI transformation within the health care sector. Specifically, we focused on empirical studies with a particular emphasis on leadership. We used an inductive, thematic analysis approach to qualitatively map the evidence. The findings were reported in accordance with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analysis extension for Scoping Reviews) guidelines.

RESULTS

A comprehensive review of 2813 unique abstracts led to the retrieval of 97 full-text articles, with 22 included for detailed assessment. Our literature mapping reveals that successful AI integration within healthcare organizations requires leadership engagement across technological, strategic, operational, and organizational domains. Leaders must demonstrate a blend of technical expertise, adaptive strategies, and strong interpersonal skills to navigate the dynamic healthcare landscape shaped by complex regulatory, technological, and organizational factors.

CONCLUSIONS

In conclusion, leading AI transformation in healthcare requires a multidimensional approach, with leadership across technological, strategic, operational, and organizational domains. Organizations should implement a comprehensive leadership development strategy, including targeted training and cross-functional collaboration, to equip leaders with the skills needed for AI integration. Additionally, when upskilling or recruiting AI talent, priority should be given to individuals with a strong mix of technical expertise, adaptive capacity, and interpersonal acumen, enabling them to navigate the unique complexities of the healthcare environment.

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 Impact of Information Relevancy and Interactivity on Intensivists' Trust in a Machine Learning-Based Bacteremia Prediction System: Simulation Study

Background

The exponential growth in computing power and the increasing digitization of information have substantially advanced the machine learning (ML) research field. However, ML algorithms are often considered "black boxes," and this fosters distrust. In medical domains, in which mistakes can result in fatal outcomes, practitioners may be especially reluctant to trust ML algorithms.

Objective

The aim of this study is to explore the effect of user-interface design features on intensivists' trust in an ML-based clinical decision support system.

Methods

A total of 47 physicians from critical care specialties were presented with 3 patient cases of bacteremia in the setting of an ML-based simulation system. Three conditions of the simulation were tested according to combinations of information relevancy and interactivity. Participants' trust in the system was assessed by their agreement with the system's prediction and a postexperiment questionnaire. Linear regression models were applied to measure the effects.

Results

Participants' agreement with the system's prediction did not differ according to the experimental conditions. However, in the postexperiment questionnaire, higher information relevancy ratings and interactivity ratings were associated with higher perceived trust in the system (P<.001 for both). The explicit visual presentation of the features of the ML algorithm on the user interface resulted in lower trust among the participants (P=.05).

Conclusions

Information relevancy and interactivity features should be considered in the design of the user interface of ML-based clinical decision support systems to enhance intensivists' trust. This study sheds light on the connection between information relevancy, interactivity, and trust in human-ML interaction, specifically in the intensive care unit environment.

Human-AI Teaming in Critical Care: A Comparative Analysis of Data Scientists' and Clinicians' Perspectives on AI Augmentation and Automation

BACKGROUND

Artificial intelligence (AI) holds immense potential for enhancing clinical and administrative health care tasks. However, slow adoption and implementation challenges highlight the need to consider how humans can effectively collaborate with AI within broader socio-technical systems in health care.

OBJECTIVE

In the example of intensive care units (ICUs), we compare data scientists' and clinicians' assessments of the optimal utilization of human and AI capabilities by determining suitable levels of human-AI teaming for safely and meaningfully augmenting or automating 6 core tasks. The goal is to provide actionable recommendations for policy makers and health care practitioners regarding AI design and implementation.

METHODS

In this multimethod study, we combine a systematic task analysis across 6 ICUs with an international Delphi survey involving 19 health data scientists from the industry and academia and 61 ICU clinicians (25 physicians and 36 nurses) to define and assess optimal levels of human-AI teaming (level 1=no performance benefits; level 2=AI augments human performance; level 3=humans augment AI performance; level 4=AI performs without human input). Stakeholder groups also considered ethical and social implications.

RESULTS

Both stakeholder groups chose level 2 and 3 human-AI teaming for 4 out of 6 core tasks in the ICU. For one task (monitoring), level 4 was the preferred design choice. For the task of patient interactions, both data scientists and clinicians agreed that AI should not be used regardless of technological feasibility due to the importance of the physician-patient and nurse-patient relationship and ethical concerns. Human-AI design choices rely on interpretability, predictability, and control over AI systems. If these conditions are not met and AI performs below human-level reliability, a reduction to level 1 or shifting accountability away from human end users is advised. If AI performs at or beyond human-level reliability and these conditions are not met, shifting to level 4 automation should be considered to ensure safe and efficient human-AI teaming.

CONCLUSIONS

By considering the sociotechnical system and determining appropriate levels of human-AI teaming, our study showcases the potential for improving the safety and effectiveness of AI usage in ICUs and broader health care settings. Regulatory measures should prioritize interpretability, predictability, and control if clinicians hold full accountability. Ethical and social implications must be carefully evaluated to ensure effective collaboration between humans and AI, particularly considering the most recent advancements in generative AI.

Knowledge, Attitude and Practice of Radiologists Regarding Artificial Intelligence in Medical Imaging

Purpose

This study aimed to investigate the knowledge, attitudes, and practice (KAP) of radiologists regarding artificial intelligence (AI) in medical imaging in the southeast of China.

Methods

This cross-sectional study was conducted among radiologists in the Jiangsu, Zhejiang, and Fujian regions from October to December 2022. A self-administered questionnaire was used to collect demographic data and assess the KAP of participants towards AI in medical imaging. A structural equation model (SEM) was used to analyze the relationships between KAP.

Results

The study included 452 valid questionnaires. The mean knowledge score was 9.01±4.87, the attitude score was 48.96±4.90, and 75.22% of participants actively engaged in AI-related practices. Having a master's degree or above (OR=1.877, P=0.024), 5-10 years of radiology experience (OR=3.481, P=0.010), AI diagnosis-related training (OR=2.915, P<0.001), and engaging in AI diagnosis-related research (OR=3.178, P<0.001) were associated with sufficient knowledge. Participants with a junior college degree (OR=2.139, P=0.028), 5-10 years of radiology experience (OR=2.462, P=0.047), and AI diagnosis-related training (OR=2.264, P<0.001) were associated with a positive attitude. Higher knowledge scores (OR=5.240, P<0.001), an associate senior professional title (OR=4.267, P=0.026), 5-10 years of radiology experience (OR=0.344, P=0.044), utilizing AI diagnosis (OR=3.643, P=0.001), and engaging in AI diagnosis-related research (OR=6.382, P<0.001) were associated with proactive practice. The SEM showed that knowledge had a direct effect on attitude (β=0.481, P<0.001) and practice (β=0.412, P<0.001), and attitude had a direct effect on practice (β=0.135, P<0.001).

Conclusion

Radiologists in southeastern China hold a favorable outlook on AI-assisted medical imaging, showing solid understanding and enthusiasm for its adoption, despite half lacking relevant training. There is a need for more AI diagnosis-related training, an efficient standardized AI database for medical imaging, and active promotion of AI-assisted imaging in clinical practice. Further research with larger sample sizes and more regions is necessary.

The use of artificial intelligence to optimize medication alerts generated by clinical decision support systems: a scoping review

OBJECTIVE

Current Clinical Decision Support Systems (CDSSs) generate medication alerts that are of limited clinical value, causing alert fatigue. Artificial Intelligence (AI)-based methods may help in optimizing medication alerts. Therefore, we conducted a scoping review on the current state of the use of AI to optimize medication alerts in a hospital setting. Specifically, we aimed to identify the applied AI methods used together with their performance measures and main outcome measures.

MATERIALS AND METHODS

We searched Medline, Embase, and Cochrane Library database on May 25, 2023 for studies of any quantitative design, in which the use of AI-based methods was investigated to optimize medication alerts generated by CDSSs in a hospital setting. The screening process was supported by ASReview software.

RESULTS

Out of 5625 citations screened for eligibility, 10 studies were included. Three studies (30%) reported on both statistical performance and clinical outcomes. The most often reported performance measure was positive predictive value ranging from 9% to 100%. Regarding main outcome measures, alerts optimized using AI-based methods resulted in a decreased alert burden, increased identification of inappropriate or atypical prescriptions, and enabled prediction of user responses. In only 2 studies the AI-based alerts were implemented in hospital practice, and none of the studies conducted external validation.

DISCUSSION AND CONCLUSION

AI-based methods can be used to optimize medication alerts in a hospital setting. However, reporting on models' development and validation should be improved, and external validation and implementation in hospital practice should be encouraged.

Energy landscape analysis and time-series clustering analysis of patient state multistability related to rheumatoid arthritis drug treatment: The KURAMA cohort study

Rheumatoid arthritis causes joint inflammation due to immune abnormalities, resulting in joint pain and swelling. In recent years, there have been considerable advancements in the treatment of this disease. However, only approximately 60% of patients achieve remission. Patients with multifactorial diseases shift between states from day to day. Patients may remain in a good or poor state with few or no transitions, or they may switch between states frequently. The visualization of time-dependent state transitions, based on the evaluation axis of stable/unstable states, may provide useful information for achieving rheumatoid arthritis treatment goals. Energy landscape analysis can be used to quantitatively determine the stability/instability of each state in terms of energy. Time-series clustering is another method used to classify transitions into different groups to identify potential patterns within a time-series dataset. The objective of this study was to utilize energy landscape analysis and time-series clustering to evaluate multidimensional time-series data in terms of multistability. We profiled each patient's state transitions during treatment using energy landscape analysis and time-series clustering. Energy landscape analysis divided state transitions into two patterns: "good stability leading to remission" and "poor stability leading to treatment dead-end." The number of patients whose disease status improved increased markedly until approximately 6 months after treatment initiation and then plateaued after 1 year. Time-series clustering grouped patients into three clusters: "toward good stability," "toward poor stability," and "unstable." Patients in the "unstable" cluster are considered to have clinical courses that are difficult to predict; therefore, these patients should be treated with more care. Early disease detection and treatment initiation are important. The evaluation of state multistability enables us to understand a patient's current state in the context of overall state transitions related to rheumatoid arthritis drug treatment and to predict future state transitions.

Artificial intelligence performance in clinical neurology queries: the ChatGPT model

INTRODUCTION

The use of artificial intelligence technology is progressively expanding and advancing in the health and biomedical literature. Since its launch, ChatGPT has rapidly gained popularity and become one of the fastest-growing artificial intelligence applications in history. This study evaluated the accuracy and comprehensiveness of ChatGPT-generated responses to medical queries in clinical neurology.

METHODS

We directed 216 questions from different subspecialties to ChatGPT. The questions were classified into three categories: multiple-choice, descriptive, and binary (yes/no answers). Each question in all categories was subjectively rated as easy, medium, or hard according to its difficulty level. Questions that also tested for intuitive clinical thinking and reasoning ability were evaluated in a separate category.

RESULTS

ChatGPT correctly answered 141 questions (65.3%). No significant difference was detected in the accuracy and comprehensiveness scale scores or correct answer rates in comparisons made according to the question style or difficulty level. However, a comparative analysis assessing question characteristics revealed significantly lower accuracy and comprehensiveness scale scores and correct answer rates for questions based on interpretations that required critical thinking (p = 0.007, 0.007, and 0.001, respectively).

CONCLUSION

ChatGPT had a moderate overall performance in clinical neurology and demonstrated inadequate performance in answering questions that required interpretation and critical thinking. It also displayed limited performance in specific subspecialties. It is essential to acknowledge the limitations of artificial intelligence and diligently verify medical information produced by such models using reliable sources.

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.

Artificial Intelligence and Occupational Health and Safety, Benefits and Drawbacks

This paper discusses the impact of artificial intelligence (AI) on occupational health and safety. Although the integration of AI into the field of occupational health and safety is still in its early stages, it has numerous applications in the workplace. Some of these applications offer numerous benefits for the health and safety of workers, such as continuous monitoring of workers' health and safety and the workplace environment through wearable devices and sensors. However, AI might have negative impacts in the workplace, such as ethical worries and data privacy concerns. To maximize the benefits and minimize the drawbacks of AI in the workplace, certain measures should be applied, such as training for both employers and employees and setting policies and guidelines regulating the integration of AI in the workplace.

Exploring the Use and Implications of AI in Sexual and Reproductive Health and Rights: Protocol for a Scoping Review

BACKGROUND

Artificial intelligence (AI) has emerged as a transformative force across the health sector and has garnered significant attention within sexual and reproductive health and rights (SRHR) due to polarizing views on its opportunities to advance care and the heightened risks and implications it brings to people's well-being and bodily autonomy. As the fields of AI and SRHR evolve, clarity is needed to bridge our understanding of how AI is being used within this historically politicized health area and raise visibility on the critical issues that can facilitate its responsible and meaningful use.

OBJECTIVE

This paper presents the protocol for a scoping review to synthesize empirical studies that focus on the intersection of AI and SRHR. The review aims to identify the characteristics of AI systems and tools applied within SRHR, regarding health domains, intended purpose, target users, AI data life cycle, and evidence on benefits and harms.

METHODS

The scoping review follows the standard methodology developed by Arksey and O'Malley. We will search the following electronic databases: MEDLINE (PubMed), Scopus, Web of Science, and CINAHL. Inclusion criteria comprise the use of AI systems and tools in sexual and reproductive health and clear methodology describing either quantitative or qualitative approaches, including program descriptions. Studies will be excluded if they focus entirely on digital interventions that do not explicitly use AI systems and tools, are about robotics or nonhuman subjects, or are commentaries. We will not exclude articles based on geographic location, language, or publication date. The study will present the uses of AI across sexual and reproductive health domains, the intended purpose of the AI system and tools, and maturity within the AI life cycle. Outcome measures will be reported on the effect, accuracy, acceptability, resource use, and feasibility of studies that have deployed and evaluated AI systems and tools. Ethical and legal considerations, as well as findings from qualitative studies, will be synthesized through a narrative thematic analysis. We will use the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) format for the publication of the findings.

RESULTS

The database searches resulted in 12,793 records when the searches were conducted in October 2023. Screening is underway, and the analysis is expected to be completed by July 2024.

CONCLUSIONS

The findings will provide key insights on usage patterns and evidence on the use of AI in SRHR, as well as convey key ethical, safety, and legal considerations. The outcomes of this scoping review are contributing to a technical brief developed by the World Health Organization and will guide future research and practice in this highly charged area of work.

TRIAL REGISTRATION

OSF Registries osf.io/ma4d9; https://osf.io/ma4d9.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

PRR1-10.2196/53888.

Barriers and facilitators to guideline for the management of pediatric off-label use of drugs in China: a qualitative descriptive study

BACKGROUND

Despite being a global public health concern, there is a research gap in analyzing implementation strategies for managing off-label drug use in children. This study aims to understand professional health managers' perspectives on implementing the Guideline in hospitals and determine the Guideline's implementation facilitators and barriers.

METHODS

Pediatric directors, pharmacy directors, and medical department directors from secondary and tertiary hospitals across the country were recruited for online interviews. The interviews were performed between June 27 and August 25, 2022. The Consolidated Framework for Implementation Research (CFIR) was adopted for data collection, data analysis, and findings interpretation to implement interventions across healthcare settings.

RESULTS

Individual interviews were conducted with 28 healthcare professionals from all over the Chinese mainland. Key stakeholders in implementing the Guideline for the Management of Pediatric Off-Label Use of Drugs in China (2021) were interviewed to identify 57 influencing factors, including 27 facilitators, 29 barriers, and one neutral factor, based on the CFIR framework. The study revealed the complexity of the factors influencing managing children's off-label medication use. A lack of policy incentives was the key obstacle in external settings. The communication barrier between pharmacists and physicians was the most critical internal barrier.

CONCLUSION

To our knowledge, this study significantly reduces the implementation gap in managing children's off-label drug use. We provided a reference for the standardized management of children's off-label use of drugs.

Artificial intelligence in wound care: diagnosis, assessment and treatment of hard-to-heal wounds: a narrative review

OBJECTIVE

The effective assessment of wounds, both acute and hard-to-heal, is an important component in the delivery by wound care practitioners of efficacious wound care for patients. Improved wound diagnosis, optimising wound treatment regimens, and enhanced prevention of wounds aid in providing patients with a better quality of life (QoL). There is significant potential for the use of artificial intelligence (AI) in health-related areas such as wound care. However, AI-based systems remain to be developed to a point where they can be used clinically to deliver high-quality wound care. We have carried out a narrative review of the development and use of AI in the diagnosis, assessment and treatment of hard-to-heal wounds. We retrieved 145 articles from several online databases and other online resources, and 81 of them were included in this narrative review. Our review shows that AI application in wound care offers benefits in the assessment/diagnosis, monitoring and treatment of acute and hard-to-heal wounds. As well as offering patients the potential of improved QoL, AI may also enable better use of healthcare resources.

Utilizing Artificial Intelligence Application for Diagnosis of Oral Lesions and Assisting Young Oral Histopathologist in Deriving Diagnosis from Provided Features - A Pilot study

Background

AI in healthcare services is advancing every day, with a focus on uprising cognitive capabilities. Higher cognitive functions in AI entail performing intricate processes like decision-making, problem-solving, perception, and reasoning. This advanced cognition surpasses basic data handling, encompassing skills to grasp ideas, understand and apply information contextually, and derive novel insights from previous experiences and acquired knowledge. ChatGPT, a natural language processing model, exemplifies this evolution by engaging in conversations with humans, furnishing responses to inquiries.

Objective

We aimed to understand the capability of ChatGPT in solving doubts pertaining to symptoms and histological features related to subject of oral pathology. The study's objective is to evaluate ChatGPT's effectiveness in answering questions pertaining to diagnoses.

Methods

This cross-sectional study was done using an AI-based ChatGPT application that provides free service for research and learning purposes. The current version of ChatGPT3.5 was used to obtain responses for a total of 25 queries. These randomly asked questions were based on basic queries from patient aspect and early oral histopathologists. These responses were obtained and stored for further processing. The responses were evaluated by five experienced pathologists on a four point liekart scale. The score were further subjected for deducing kappa values for reliability.

Result & Statistical Analysis

A total of 25 queries were solved by the program in the shortest possible time for an answer. The sensitivity and specificity of the methods and the responses were represented using frequency and percentages. Both the responses were analysed and were statistically significant based on the measurement of kappa values.

Conclusion

The proficiency of ChatGPT in handling intricate reasoning queries within pathology demonstrated a noteworthy level of relational accuracy. Consequently, its text output created coherent links between elements, producing meaningful responses. This suggests that scholars or students can rely on this program to address reasoning-based inquiries. Nevertheless, considering the continual advancements in the program's development, further research is essential to determine its accuracy levels in future versions.

A multinational study on artificial intelligence adoption: Clinical implementers' perspectives

BACKGROUND

Despite substantial progress in AI research for healthcare, translating research achievements to AI systems in clinical settings is challenging and, in many cases, unsatisfactory. As a result, many AI investments have stalled at the prototype level, never reaching clinical settings.

OBJECTIVE

To improve the chances of future AI implementation projects succeeding, we analyzed the experiences of clinical AI system implementers to better understand the challenges and success factors in their implementations.

METHODS

Thirty-seven implementers of clinical AI from European and North and South American countries were interviewed. Semi-structured interviews were transcribed and analyzed qualitatively with the framework method, identifying the success factors and the reasons for challenges as well as documenting proposals from implementers to improve AI adoption in clinical settings.

RESULTS

We gathered the implementers' requirements for facilitating AI adoption in the clinical setting. The main findings include 1) the lesser importance of AI explainability in favor of proper clinical validation studies, 2) the need to actively involve clinical practitioners, and not only clinical researchers, in the inception of AI research projects, 3) the need for better information structures and processes to manage data access and the ethical approval of AI projects, 4) the need for better support for regulatory compliance and avoidance of duplications in data management approval bodies, 5) the need to increase both clinicians' and citizens' literacy as respects the benefits and limitations of AI, and 6) the need for better funding schemes to support the implementation, embedding, and validation of AI in the clinical workflow, beyond pilots.

CONCLUSION

Participants in the interviews are positive about the future of AI in clinical settings. At the same time, they proposenumerous measures to transfer research advancesinto implementations that will benefit healthcare personnel. Transferring AI research into benefits for healthcare workers and patients requires adjustments in regulations, data access procedures, education, funding schemes, and validation of AI systems.

From Pixels to Prognosis: A Narrative Review on Artificial Intelligence's Pioneering Role in Colorectal Carcinoma Histopathology

Colorectal carcinoma, a prevalent and deadly malignancy, necessitates precise histopathological assessment for effective diagnosis and prognosis. Artificial intelligence (AI) emerges as a transformative force in this realm, offering innovative solutions to enhance traditional histopathological methods. This narrative review explores AI's pioneering role in colorectal carcinoma histopathology, encompassing its evolution, techniques, and advancements. AI algorithms, notably machine learning and deep learning, have revolutionized image analysis, facilitating accurate diagnosis and prognosis prediction. Furthermore, AI-driven histopathological analysis unveils potential biomarkers and therapeutic targets, heralding personalized treatment approaches. Despite its promise, challenges persist, including data quality, interpretability, and integration. Collaborative efforts among researchers, clinicians, and AI developers are imperative to surmount these hurdles and realize AI's full potential in colorectal carcinoma care. This review underscores AI's transformative impact and implications for future oncology research, clinical practice, and interdisciplinary collaboration.

Artificial Intelligence: A Game-Changer for Mental Health Care

Starting from the escalating global burden of mental health disorders, exacerbated by the COVID-19 pandemic, the article examines the potential of artificial intelligence (AI) to revolutionize mental health care. With nearly one in five adults facing mental health issues and suicide ranking as a leading cause of death among the young, the strained mental health system seeks innovative solutions. The text discusses the rapid evolution of AI, particularly in image analysis for early physical health diagnoses, and its promising applications in mental health, including predictive analytics for various disorders. AI's ability to analyze written language, speech characteristics, and physiological signals from wearables offers avenues for remote monitoring and early prognosis. Despite the need to address ethical considerations, particularly biases in data sets and concerns about potential patient detachment, the article advocates for AI as a complementary tool rather than a replacement for human involvement in mental health services. Overall, the article emphasizes the transformative potential of AI in enhancing diagnostics, monitoring, and treatment strategies for mental health disorders.

Assessing the role of GPT-4 in thyroid ultrasound diagnosis and treatment recommendations: enhancing interpretability with a chain of thought approach

Background

As artificial intelligence (AI) becomes increasingly prevalent in the medical field, the effectiveness of AI-generated medical reports in disease diagnosis remains to be evaluated. ChatGPT is a large language model developed by open AI with a notable capacity for text abstraction and comprehension. This study aimed to explore the capabilities, limitations, and potential of Generative Pre-trained Transformer (GPT)-4 in analyzing thyroid cancer ultrasound reports, providing diagnoses, and recommending treatment plans.

Methods

Using 109 diverse thyroid cancer cases, we evaluated GPT-4's performance by comparing its generated reports to those from doctors with various levels of experience. We also conducted a Turing Test and a consistency analysis. To enhance the interpretability of the model, we applied the Chain of Thought (CoT) method to deconstruct the decision-making chain of the GPT model.

Results

GPT-4 demonstrated proficiency in report structuring, professional terminology, and clarity of expression, but showed limitations in diagnostic accuracy. In addition, our consistency analysis highlighted certain discrepancies in the AI's performance. The CoT method effectively enhanced the interpretability of the AI's decision-making process.

Conclusions

GPT-4 exhibits potential as a supplementary tool in healthcare, especially for generating thyroid gland diagnostic reports. Our proposed online platform, "ThyroAIGuide", alongside the CoT method, underscores the potential of AI to augment diagnostic processes, elevate healthcare accessibility, and advance patient education. However, the journey towards fully integrating AI into healthcare is ongoing, requiring continuous research, development, and careful monitoring by medical professionals to ensure patient safety and quality of care.

A Machine Learning Approach with Human-AI Collaboration for Automated Classification of Patient Safety Event Reports: Algorithm Development and Validation Study

BACKGROUND

Adverse events refer to incidents with potential or actual harm to patients in hospitals. These events are typically documented through patient safety event (PSE) reports, which consist of detailed narratives providing contextual information on the occurrences. Accurate classification of PSE reports is crucial for patient safety monitoring. However, this process faces challenges due to inconsistencies in classifications and the sheer volume of reports. Recent advancements in text representation, particularly contextual text representation derived from transformer-based language models, offer a promising solution for more precise PSE report classification. Integrating the machine learning (ML) classifier necessitates a balance between human expertise and artificial intelligence (AI). Central to this integration is the concept of explainability, which is crucial for building trust and ensuring effective human-AI collaboration.

OBJECTIVE

This study aims to investigate the efficacy of ML classifiers trained using contextual text representation in automatically classifying PSE reports. Furthermore, the study presents an interface that integrates the ML classifier with the explainability technique to facilitate human-AI collaboration for PSE report classification.

METHODS

This study used a data set of 861 PSE reports from a large academic hospital's maternity units in the Southeastern United States. Various ML classifiers were trained with both static and contextual text representations of PSE reports. The trained ML classifiers were evaluated with multiclass classification metrics and the confusion matrix. The local interpretable model-agnostic explanations (LIME) technique was used to provide the rationale for the ML classifier's predictions. An interface that integrates the ML classifier with the LIME technique was designed for incident reporting systems.

RESULTS

The top-performing classifier using contextual representation was able to obtain an accuracy of 75.4% (95/126) compared to an accuracy of 66.7% (84/126) by the top-performing classifier trained using static text representation. A PSE reporting interface has been designed to facilitate human-AI collaboration in PSE report classification. In this design, the ML classifier recommends the top 2 most probable event types, along with the explanations for the prediction, enabling PSE reporters and patient safety analysts to choose the most suitable one. The LIME technique showed that the classifier occasionally relies on arbitrary words for classification, emphasizing the necessity of human oversight.

CONCLUSIONS

This study demonstrates that training ML classifiers with contextual text representations can significantly enhance the accuracy of PSE report classification. The interface designed in this study lays the foundation for human-AI collaboration in the classification of PSE reports. The insights gained from this research enhance the decision-making process in PSE report classification, enabling hospitals to more efficiently identify potential risks and hazards and enabling patient safety analysts to take timely actions to prevent patient harm.

Comparing Artificial Intelligence and Senior Residents in Oral Lesion Diagnosis: A Comparative Study

INTRODUCTION

Artificial intelligence (AI) is a field of computer science that seeks to build intelligent machines that can carry out tasks that usually necessitate human intelligence. AI may help dentists with a variety of dental tasks, including clinical diagnosis and treatment planning. This study aims to compare the performance of AI and oral medicine residents in diagnosing different cases, providing treatment, and determining if it is reliable to assist them in their field of work.

METHODS

The study conducted a comparative analysis of the responses from third- and fourth-year residents trained in Oral Medicine and Pathology at King Saud University, College of Dentistry. The residents were given a closed multiple-choice test consisting of 19 questions with four response options labeled A-D and one question with five response options labeled A-E. The test was administered via Google Forms, and each resident's response was stored electronically in an Excel sheet (Microsoft® Corp., Redmond, WA). The residents' answers were then compared to the responses generated by three major language models: OpenAI, Stablediffusion, and PopAI. The questions were inputted into the language models in the same format as the original test, and prior to each question, an artificial intelligence chat session was created to eliminate memory retention bias. The input was done on November 19, 2023, the same day the official multiple-choice test was administered. The study had a sample size of 20 residents trained in Oral Medicine and Pathology at King Saud University, College of Dentistry, consisting of both third-year and fourth-year residents.

RESULT

The responses of three large language models (LLM), including OpenAI, Stablediffusion, and PopAI, as well as the responses of 20 senior residents for 20 clinical cases about oral lesion diagnosis. There were no significant variations observed for the remaining questions in the responses to only two questions (10%). For the remaining questions, there were no significant differences. The median (IQR) score of LLMs was 50.0 (45.0 to 60.0), with a minimum of 40 (for stable diffusion) and a maximum of 70 (for OpenAI). The median (IQR) score of senior residents was 65.0 (55.0-75.0). The highest and lowest scores of residents were 40 and 90, respectively. There was no significant difference in the percent scores of residents and LLMs (p = 0.211). The agreement level was measured using the Kappa value. The agreement among senior dental residents was observed to be weak, with a Kappa value of 0.396. In contrast, the agreement among LLMs demonstrated a moderate level, with a Kappa value of 0.622, suggesting a more cohesive alignment in responses among the artificial intelligence models. When comparing residents' responses with those generated by different OpenAI models, including OpenAI, Stablediffusion, and PopAI, the agreement levels were consistently categorized as weak, with Kappa values of 0.402, 0.381, and 0.392, respectively.

CONCLUSION

What the current study reveals is that when comparing the response score, there is no significant difference, in contrast to the agreement analysis among the residents, which was low compared to the LLMs, in which it was high. Dentists should consider that AI is very beneficial in providing diagnosis and treatment and use it to assist them.

Reimagining Core Entrustable Professional Activities for Undergraduate Medical Education in the Era of Artificial Intelligence

The proliferation of generative artificial intelligence (AI) and its extensive potential for integration into many aspects of health care signal a transformational shift within the health care environment. In this context, medical education must evolve to ensure that medical trainees are adequately prepared to navigate the rapidly changing health care landscape. Medical education has moved toward a competency-based education paradigm, leading the Association of American Medical Colleges (AAMC) to define a set of Entrustable Professional Activities (EPAs) as its practical operational framework in undergraduate medical education. The AAMC's 13 core EPAs for entering residencies have been implemented with varying levels of success across medical schools. In this paper, we critically assess the existing core EPAs in the context of rapid AI integration in medicine. We identify EPAs that require refinement, redefinition, or comprehensive change to align with the emerging trends in health care. Moreover, this perspective proposes a set of "emerging" EPAs, informed by the changing landscape and capabilities presented by generative AI technologies. We provide a practical evaluation of the EPAs, alongside actionable recommendations on how medical education, viewed through the lens of the AAMC EPAs, can adapt and remain relevant amid rapid technological advancements. By leveraging the transformative potential of AI, we can reshape medical education to align with an AI-integrated future of medicine. This approach will help equip future health care professionals with technological competence and adaptive skills to meet the dynamic and evolving demands in health care.

Existing Barriers Faced by and Future Design Recommendations for Direct-to-Consumer Health Care Artificial Intelligence Apps: Scoping Review

BACKGROUND

Direct-to-consumer (DTC) health care artificial intelligence (AI) apps hold the potential to bridge the spatial and temporal disparities in health care resources, but they also come with individual and societal risks due to AI errors. Furthermore, the manner in which consumers interact directly with health care AI is reshaping traditional physician-patient relationships. However, the academic community lacks a systematic comprehension of the research overview for such apps.

OBJECTIVE

This paper systematically delineated and analyzed the characteristics of included studies, identified existing barriers and design recommendations for DTC health care AI apps mentioned in the literature and also provided a reference for future design and development.

METHODS

This scoping review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews guidelines and was conducted according to Arksey and O'Malley's 5-stage framework. Peer-reviewed papers on DTC health care AI apps published until March 27, 2023, in Web of Science, Scopus, the ACM Digital Library, IEEE Xplore, PubMed, and Google Scholar were included. The papers were analyzed using Braun and Clarke's reflective thematic analysis approach.

RESULTS

Of the 2898 papers retrieved, 32 (1.1%) covering this emerging field were included. The included papers were recently published (2018-2023), and most (23/32, 72%) were from developed countries. The medical field was mostly general practice (8/32, 25%). In terms of users and functionalities, some apps were designed solely for single-consumer groups (24/32, 75%), offering disease diagnosis (14/32, 44%), health self-management (8/32, 25%), and health care information inquiry (4/32, 13%). Other apps connected to physicians (5/32, 16%), family members (1/32, 3%), nursing staff (1/32, 3%), and health care departments (2/32, 6%), generally to alert these groups to abnormal conditions of consumer users. In addition, 8 barriers and 6 design recommendations related to DTC health care AI apps were identified. Some more subtle obstacles that are particularly worth noting and corresponding design recommendations in consumer-facing health care AI systems, including enhancing human-centered explainability, establishing calibrated trust and addressing overtrust, demonstrating empathy in AI, improving the specialization of consumer-grade products, and expanding the diversity of the test population, were further discussed.

CONCLUSIONS

The booming DTC health care AI apps present both risks and opportunities, which highlights the need to explore their current status. This paper systematically summarized and sorted the characteristics of the included studies, identified existing barriers faced by, and made future design recommendations for such apps. To the best of our knowledge, this is the first study to systematically summarize and categorize academic research on these apps. Future studies conducting the design and development of such systems could refer to the results of this study, which is crucial to improve the health care services provided by DTC health care AI apps.

Road map for personalized exercise medicine in T2DM

The number of patients with type 2 diabetes mellitus (T2DM) is rising at an alarming rate. Regular physical activity and exercise are cornerstones in the therapy of T2DM. While a one-size-fits-all approach fails to account for many between-subject differences, the use of personalized exercise medicine has the potential of optimizing health outcomes. Here, a road map for personalized exercise therapy targeted at patients with T2DM is presented. It considers secondary complications, glucose management, response heterogeneity, and other relevant factors that might influence the effectiveness of exercise as medicine, taking exercise-medication-diet interactions, as well as feasibility and acceptance into account. Furthermore, the potential of artificial intelligence and machine learning-based applications in assisting sports therapists to find appropriate exercise programs is outlined.

Older Adults' Perceptions About Using Intelligent Toilet Seats Beyond Traditional Care: Web-Based Interview Survey

BACKGROUND

In contemporary society, age tech (age technology) represents a significant advancement in health care aimed at enhancing patient engagement, ensuring sustained independence, and promoting quality of life for older people. One innovative form of age tech is the intelligent toilet seat, which is designed to collect, analyze, and provide insights based on toileting logs and excreta data. Understanding how older people perceive and interact with such technology can offer invaluable insights to researchers, technology developers, and vendors.

OBJECTIVE

This study examined older adults' perspectives regarding the use of intelligent toilet seats. Through a qualitative methodology, this research aims to unearth the nuances of older people's opinions, shedding light on their preferences, concerns, and potential barriers to adoption.

METHODS

Data were collected using a web-based interview survey distributed on Amazon Mechanical Turk. The analyzed data set comprised 174 US-based individuals aged ≥65 years who voluntarily participated in this study. The qualitative data were carefully analyzed using NVivo (Lumivero) based on detailed content analysis, ensuring that emerging themes were coded and classified based on the conceptual similarities in the respondents' narratives.

RESULTS

The analysis revealed 5 dominant themes encompassing the opinions of aging adults. The perceived benefits and advantages of using the intelligent toilet seat were grouped into 3 primary themes: health-related benefits including the potential for early disease detection, continuous health monitoring, and seamless connection to health care insights. Technology-related advantages include the noninvasive nature of smart toilet seats and leveraging unique and innovative data collection and analysis technology. Use-related benefits include ease of use, potential for multiple users, and cost reduction owing to the reduced need for frequent clinical visits. Conversely, the concerns and perceived risks were classified into 2 significant themes: psychological concerns, which included concerns about embarrassment and aging-related stereotypes, and the potential emotional impact of constant health monitoring. Technical performance risks include concerns centered on privacy and security, device reliability, data accuracy, potential malfunctions, and the implications of false positives or negatives.

CONCLUSIONS

The decision of older adults to incorporate intelligent toilet seats into their daily lives depends on myriad factors. Although the potential health and technological benefits are evident, valid concerns that need to be addressed remain. To foster widespread adoption, it is imperative to enhance the advantages while simultaneously addressing and mitigating the identified risks. This balanced approach will pave the way for a more holistic integration of smart health care devices into the routines of the older population, ensuring that they reap the full benefits of age tech advancements.

A Framework to Guide Implementation of AI in Health Care: Protocol for a Cocreation Research Project

BACKGROUND

Artificial intelligence (AI) has the potential in health care to transform patient care and administrative processes, yet health care has been slow to adopt AI due to many types of barriers. Implementation science has shown the importance of structured implementation processes to overcome implementation barriers. However, there is a lack of knowledge and tools to guide such processes when implementing AI-based applications in health care.

OBJECTIVE

The aim of this protocol is to describe the development, testing, and evaluation of a framework, "Artificial Intelligence-Quality Implementation Framework" (AI-QIF), intended to guide decisions and activities related to the implementation of various AI-based applications in health care.

METHODS

The paper outlines the development of an AI implementation framework for broad use in health care based on the Quality Implementation Framework (QIF). QIF is a process model developed in implementation science. The model guides the user to consider implementation-related issues in a step-by-step design and plan and perform activities that support implementation. This framework was chosen for its adaptability, usability, broad scope, and detailed guidance concerning important activities and considerations for successful implementation. The development will proceed in 5 phases with primarily qualitative methods being used. The process starts with phase I, in which an AI-adapted version of QIF is created (AI-QIF). Phase II will produce a digital mockup of the AI-QIF. Phase III will involve the development of a prototype of the AI-QIF with an intuitive user interface. Phase IV is dedicated to usability testing of the prototype in health care environments. Phase V will focus on evaluating the usability and effectiveness of the AI-QIF. Cocreation is a guiding principle for the project and is an important aspect in 4 of the 5 development phases. The cocreation process will enable the use of both on research-based and practice-based knowledge.

RESULTS

The project is being conducted within the frame of a larger research program, with the overall objective of developing theoretically and empirically informed frameworks to support AI implementation in routine health care. The program was launched in 2021 and has carried out numerous research activities. The development of AI-QIF as a tool to guide the implementation of AI-based applications in health care will draw on knowledge and experience acquired from these activities. The framework is being developed over 2 years, from January 2023 to December 2024. It is under continuous development and refinement.

CONCLUSIONS

The development of the AI implementation framework, AI-QIF, described in this study protocol aims to facilitate the implementation of AI-based applications in health care based on the premise that implementation processes benefit from being well-prepared and structured. The framework will be coproduced to enhance its relevance, validity, usefulness, and potential value for application in practice.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/50216.

Development, Reliability, and Structural Validity of the Scale for Knowledge, Attitude, and Practice in Ethics Implementation Among AI Researchers: Cross-Sectional Study

BACKGROUND

Medical artificial intelligence (AI) has significantly contributed to decision support for disease screening, diagnosis, and management. With the growing number of medical AI developments and applications, incorporating ethics is considered essential to avoiding harm and ensuring broad benefits in the lifecycle of medical AI. One of the premises for effectively implementing ethics in Medical AI research necessitates researchers' comprehensive knowledge, enthusiastic attitude, and practical experience. However, there is currently a lack of an available instrument to measure these aspects.

OBJECTIVE

The aim of this study was to develop a comprehensive scale for measuring the knowledge, attitude, and practice of ethics implementation among medical AI researchers, and to evaluate its measurement properties.

METHODS

The construct of the Knowledge-Attitude-Practice in Ethics Implementation (KAP-EI) scale was based on the Knowledge-Attitude-Practice (KAP) model, and the evaluation of its measurement properties was in compliance with the COnsensus-based Standards for the selection of health status Measurement INstruments (COSMIN) reporting guidelines for studies on measurement instruments. The study was conducted in 2 phases. The first phase involved scale development through a systematic literature review, qualitative interviews, and item analysis based on a cross-sectional survey. The second phase involved evaluation of structural validity and reliability through another cross-sectional study.

RESULTS

The KAP-EI scale had 3 dimensions including knowledge (10 items), attitude (6 items), and practice (7 items). The Cronbach α for the whole scale reached .934. Confirmatory factor analysis showed that the goodness-of-fit indices of the scale were satisfactory (χ2/df ratio:=2.338, comparative fit index=0.949, Tucker Lewis index=0.941, root-mean-square error of approximation=0.064, and standardized root-mean-square residual=0.052).

CONCLUSIONS

The results show that the scale has good reliability and structural validity; hence, it could be considered an effective instrument. This is the first instrument developed for this purpose.

Perceptions and concerns of emergency medicine practitioners about artificial intelligence in emergency triage management during the pandemic: a national survey-based study

Objective

There have been continuous discussions over the ethics of using AI in healthcare. We sought to identify the ethical issues and viewpoints of Turkish emergency care doctors about the use of AI during epidemic triage.

Materials and methods

Ten emergency specialists were initially enlisted for this project, and their responses to open-ended questions about the ethical issues surrounding AI in the emergency room provided valuable information. A 15-question survey was created based on their input and was refined through a pilot test with 15 emergency specialty doctors. Following that, the updated survey was sent to emergency specialists via email, social media, and private email distribution.

Results

167 emergency medicine specialists participated in the study, with an average age of 38.22 years and 6.79 years of professional experience. The majority agreed that AI could benefit patients (54.50%) and healthcare professionals (70.06%) in emergency department triage during pandemics. Regarding responsibility, 63.47% believed in shared responsibility between emergency medicine specialists and AI manufacturers/programmers for complications. Additionally, 79.04% of participants agreed that the responsibility for complications in AI applications varies depending on the nature of the complication. Concerns about privacy were expressed by 20.36% regarding deep learning-based applications, while 61.68% believed that anonymity protected privacy. Additionally, 70.66% of participants believed that AI systems would be as sensitive as humans in terms of non-discrimination.

Conclusion

The potential advantages of deploying AI programs in emergency department triage during pandemics for patients and healthcare providers were acknowledged by emergency medicine doctors in Turkey. Nevertheless, they expressed notable ethical concerns related to the responsibility and accountability aspects of utilizing AI systems in this context.

Scoping Meta-Review of Methods Used to Assess Artificial Intelligence-Based Medical Devices for Heart Failure

Artificial intelligence and machine learning (AI/ML) are playing increasingly important roles, permeating the field of medical devices (MDs). This rapid progress has not yet been matched by the Health Technology Assessment (HTA) process, which still needs to define a common methodology for assessing AI/ML-based MDs. To collect existing evidence from the literature about the methods used to assess AI-based MDs, with a specific focus on those used for the management of heart failure (HF), the International Federation of Medical and Biological Engineering (IFMBE) conducted a scoping meta-review. This manuscript presents the results of this search, which covered the period from January 1974 to October 2022. After careful independent screening, 21 reviews, mainly conducted in North America and Europe, were retained and included. Among the findings were that deep learning is the most commonly utilised method and that electronic health records and registries are among the most prevalent sources of data for AI/ML algorithms. Out of the 21 included reviews, 19 focused on risk prediction and/or the early diagnosis of HF. Furthermore, 10 reviews provided evidence of the impact on the incidence/progression of HF, and 13 on the length of stay. From an HTA perspective, the main areas requiring improvement are the quality assessment of studies on AI/ML (included in 11 out of 21 reviews) and their data sources, as well as the definition of the criteria used to assess the selection of the most appropriate AI/ML algorithm.

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

BACKGROUND

An increasing interest in machine learning (ML) has been observed among scholars and health care professionals. However, while ML-based applications have been shown to be effective and have the potential to change the delivery of patient care, their implementation in health care organizations is complex. There are several challenges that currently hamper the uptake of ML in daily practice, and there is currently limited knowledge on how these challenges have been addressed in empirical studies on implemented ML-based applications.

OBJECTIVE

The aim of this systematic literature review is twofold: (1) to map the ML-based applications implemented in health care organizations, with a focus on investigating the organizational dimensions that are relevant in the implementation process; and (2) to analyze the processes and strategies adopted to foster a successful uptake of ML.

METHODS

We developed this protocol following the PRISMA-P (Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols) guidelines. The search was conducted on 3 databases (PubMed, Scopus, and Web of Science), considering a 10-year time frame (2013-2023). The search strategy was built around 4 blocks of keywords (artificial intelligence, implementation, health care, and study type). Based on the detailed inclusion criteria defined, only empirical studies documenting the implementation of ML-based applications used by health care professionals in clinical settings will be considered. The study protocol was registered in PROSPERO (International Prospective Register of Systematic Reviews).

RESULTS

The review is ongoing and is expected to be completed by September 2023. Data analysis is currently underway, and the first results are expected to be submitted for publication in November 2023. The study was funded by the European Union within the Multilayered Urban Sustainability Action (MUSA) project.

CONCLUSIONS

ML-based applications involving clinical decision support and automation of clinical tasks present unique traits that add several layers of complexity compared with earlier health technologies. Our review aims at contributing to the existing literature by investigating the implementation of ML from an organizational perspective and by systematizing a conspicuous amount of information on factors influencing implementation.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/47971.

Implementing AI in healthcare-the relevance of trust: a scoping review

Background

The process of translation of AI and its potential benefits into practice in healthcare services has been slow in spite of its rapid development. Trust in AI in relation to implementation processes is an important aspect. Without a clear understanding, the development of effective implementation strategies will not be possible, nor will AI advance despite the significant investments and possibilities.

Objective

This study aimed to explore the scientific literature regarding how trust in AI in relation to implementation in healthcare is conceptualized and what influences trust in AI in relation to implementation in healthcare.

Methods

This scoping review included five scientific databases. These were searched to identify publications related to the study aims. Articles were included if they were published in English, after 2012, and peer-reviewed. Two independent reviewers conducted an abstract and full-text review, as well as carrying out a thematic analysis with an inductive approach to address the study aims. The review was reported in accordance with the PRISMA-ScR guidelines.

Results

A total of eight studies were included in the final review. We found that trust was conceptualized in different ways. Most empirical studies had an individual perspective where trust was directed toward the technology's capability. Two studies focused on trust as relational between people in the context of the AI application rather than as having trust in the technology itself. Trust was also understood by its determinants and as having a mediating role, positioned between characteristics and AI use. The thematic analysis yielded three themes: individual characteristics, AI characteristics and contextual characteristics, which influence trust in AI in relation to implementation in healthcare.

Conclusions

Findings showed that the conceptualization of trust in AI differed between the studies, as well as which determinants they accounted for as influencing trust. Few studies looked beyond individual characteristics and AI characteristics. Future empirical research addressing trust in AI in relation to implementation in healthcare should have a more holistic view of the concept to be able to manage the many challenges, uncertainties, and perceived risks.

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.

Innovation in healthcare: leadership perceptions about the innovation characteristics of artificial intelligence-a qualitative interview study with healthcare leaders in Sweden

BACKGROUND

Despite the extensive hopes and expectations for value creation resulting from the implementation of artificial intelligence (AI) applications in healthcare, research has predominantly been technology-centric rather than focused on the many changes that are required in clinical practice for the technology to be successfully implemented. The importance of leaders in the successful implementation of innovations in healthcare is well recognised, yet their perspectives on the specific innovation characteristics of AI are still unknown. The aim of this study was therefore to explore the perceptions of leaders in healthcare concerning the innovation characteristics of AI intended to be implemented into their organisation.

METHODS

The study had a deductive qualitative design, using constructs from the innovation domain in the Consolidated Framework for Implementation Research (CFIR). Interviews were conducted with 26 leaders in healthcare.

RESULTS

Participants perceived that AI could provide relative advantages when it came to care management, supporting clinical decisions, and the early detection of disease and risk of disease. The development of AI in the organisation itself was perceived as the main current innovation source. The evidence base behind AI technology was questioned, in relation to its transparency, potential quality improvement, and safety risks. Although the participants acknowledged AI to be superior to human action in terms of effectiveness and precision in some situations, they also expressed uncertainty about the adaptability and trialability of AI. Complexities such as the characteristics of the technology, the lack of conceptual consensus about AI, and the need for a variety of implementation strategies to accomplish transformative change in practice were identified, as were uncertainties about the costs involved in AI implementation.

CONCLUSION

Healthcare leaders not only saw potential in the technology and its use in practice, but also felt that AI's opacity limits its evidence strength and that complexities in relation to AI itself and its implementation influence its current use in healthcare practice. More research is needed based on actual experiences using AI applications in real-world situations and their impact on clinical practice. New theories, models, and frameworks may need to be developed to meet challenges related to the implementation of AI in healthcare.

Analysing the Applicability of ChatGPT, Bard, and Bing to Generate Reasoning-Based Multiple-Choice Questions in Medical Physiology

Background Artificial intelligence (AI) is evolving in the medical education system. ChatGPT, Google Bard, and Microsoft Bing are AI-based models that can solve problems in medical education. However, the applicability of AI to create reasoning-based multiple-choice questions (MCQs) in the field of medical physiology is yet to be explored. Objective We aimed to assess and compare the applicability of ChatGPT, Bard, and Bing in generating reasoning-based MCQs for MBBS (Bachelor of Medicine, Bachelor of Surgery) undergraduate students on the subject of physiology. Methods The National Medical Commission of India has developed an 11-module physiology curriculum with various competencies. Two physiologists independently chose a competency from each module. The third physiologist prompted all three AIs to generate five MCQs for each chosen competency. The two physiologists who provided the competencies rated the MCQs generated by the AIs on a scale of 0-3 for validity, difficulty, and reasoning ability required to answer them. We analyzed the average of the two scores using the Kruskal-Wallis test to compare the distribution across the total and module-wise responses, followed by a post-hoc test for pairwise comparisons. We used Cohen's Kappa (Κ) to assess the agreement in scores between the two raters. We expressed the data as a median with an interquartile range. We determined their statistical significance by a p-value <0.05. Results ChatGPT and Bard generated 110 MCQs for the chosen competencies. However, Bing provided only 100 MCQs as it failed to generate them for two competencies. The validity of the MCQs was rated as 3 (3-3) for ChatGPT, 3 (1.5-3) for Bard, and 3 (1.5-3) for Bing, showing a significant difference (p<0.001) among the models. The difficulty of the MCQs was rated as 1 (0-1) for ChatGPT, 1 (1-2) for Bard, and 1 (1-2) for Bing, with a significant difference (p=0.006). The required reasoning ability to answer the MCQs was rated as 1 (1-2) for ChatGPT, 1 (1-2) for Bard, and 1 (1-2) for Bing, with no significant difference (p=0.235). K was ≥ 0.8 for all three parameters across all three AI models. Conclusion AI still needs to evolve to generate reasoning-based MCQs in medical physiology. ChatGPT, Bard, and Bing showed certain limitations. Bing generated significantly least valid MCQs, while ChatGPT generated significantly least difficult MCQs.