1 Helena Mayerhoffer
1 Boris Ilić
1 University of Applied Health Sciences, Zagreb, Croatia
https://doi.org/10.24141/2/9/2/13
Author for correspondence:
Helena Mayerhoffer
University of Applied Health Sciences, Zagreb, Croatia E-mail: mayerhoffer.helena@gmail.com
Keywords: artificial intelligence, triage, medical practice, nursing practice
Introduction. Artificial intelligence can analyze large-scale datasets to enhance decision-making and efficiency. Artificial intelligence has potential in trauma triage, yet remains underexplored. Effective triage is essential for optimizing patient outcomes and resource allocation, but current systems heavily rely on nurses’ judgment, which is subject to variabil- ity. AI-driven models could enhance accuracy, reduce bias, and support clinical decision-making in emer- gency care.
Aim. This literature review explores the role of arti- ficial intelligence in medical triage, assessing its im- pact on accuracy, efficiency, and decision-making in patient assessment and prioritization.
Methods. A systematic search was conducted in March 2025 using PubMed, Web of Science and Hrčak databases to identify studies published be- tween 2022 and 2025. Articles meeting predefined inclusion criteria were selected, resulting in 31 stud- ies being included in the final review. The review fol- lowed PRISMA guidelines. Inclusion criteria consisted of systematic reviews, review articles, original re- search papers, cross-sectional studies, clinical trials, randomized controlled trials, and meta-analyses pub- lished in Croatian or English. The search terms includ- ed “artificial intelligence”, “triage”, “medical practice” and “nursing practice”.
Results. The reviewed studies demonstrate that AI models can enhance triage accuracy and reliability, sometimes outperforming healthcare professionals
in specific tasks. They showed high specificity in identifying critical cases and improving triage con- sistency. However, limitations were noted, including reduced accuracy in complex cases, overestimation of urgency, and variability in performance across different triage systems. Key limitations identified include suboptimal reproducibility in disaster simula- tions, poor performance in complex triage scenarios, training data bias, and lack of algorithm transparency. These inconsistencies highlight the need for cautious interpretation and refinement before clinical imple- mentation. While AI supports triage decision-making, human oversight remains essential. The potential of artificial intelligence depends on model training, data quality, and clinical integration. While some models perform well in emergency triage, others show incon- sistencies in disaster scenarios. AI should be seen as a complement to human expertise rather than a re- placement. The implications of these limitations in- clude risks to patient safety, limited generalizability and challenges in regulatory validation. Addressing these issues is crucial to ensure safe and effective integration of AI into clinical workflows. Challenges such as data bias, transparency, and model variability must be addressed for successful AI implementation in emergency medicine.
Conclusion. AI-driven triage systems improve ac- curacy and efficiency but require further refinement for reliability in complex cases. They function best as supportive tools rather than independent decision- makers. Future research should focus on optimizing AI integration into clinical practice to enhance emer- gency care outcomes.
Technological advancements have played a crucial role in shaping global development, with artificial intelligence emerging as one of the most rapidly evolving fields, particularly in medicine (1). Artificial intelligence (AI) refers to the ability of non-human systems to analyze input data and make decisions accordingly. Within artificial intelligence, machine learning focuses on developing algorithms that con-
tinuously improve through data exposure, allow- ing them to identify key patterns linked to specific outcomes. These patterns are stored in model pa- rameters, which guide data processing and decision- making. Deep learning, a subset of machine learning, enables models to recognize and apply intricate data patterns (2). Artificial intelligence models are typi- cally developed by training on extensive datasets to generate meaningful outputs that address pre- defined objectives. In medicine, such objectives may include patient diagnosis or prognosis, drug discov- ery and note transcription (3). Despite its growing applications, one underexplored area where artificial intelligence could be transformative is trauma triage. Triage involves categorizing patients based on injury severity to ensure they receive appropriate care at the right time and location (4). Effective triage mini- mizes preventable disabilities and fatalities while preventing emergency departments from becoming overwhelmed (5).
Nurses play a highly specialized role in the triage process, particularly within emergency departments where clinical urgency and time pressures demand rapid, accurate decision-making. Triage nurses are responsible for conducting fast yet thorough clinical assessments, determining the urgency and severity of patient conditions and ensuring appropriate prior- itization for care. This complex task requires not only clinical experience but also critical thinking, decision- making skills, and the ability to perform under pres- sure. To perform triage effectively and minimize the risk of error, nurses must undergo structured educa- tion and continuous training tailored to the demands of this high-stakes environment (6).
Errors in triage can lead to over-triage, where non- critical patients are sent to higher-level facilities un- necessarily, or under-triage, where critically injured patients do not receive specialized trauma care. Both scenarios contribute to poor patient outcomes and inefficient resource allocation (7). Currently, nurses use various conventional triage tools, such as the National Early Warning Score, Modified Early Warn- ing Score, Revised Trauma Score, and Trauma and In- jury Severity Score, depending on hospital protocols. These tools rely on basic physiological data, including respiratory rate, systolic blood pressure, heart rate, capillary refill time and Glasgow Coma Scale. Nurses integrate this information with diagnostic reasoning to determine a patient’s triage category. This typi- cally follows an analytical reasoning approach where
past experience and existing knowledge inform decision-making (8). However, the effectiveness of these tools depends on nurses’ judgment, which can be influenced by stress levels, variability in physical examinations, and differences in clinical experience
(9). Artificial intelligence, machine learning and deep learning offer a potential solution to these limitations by leveraging predictive analytics and large trauma databases such as the Trauma Audit & Research Network (9). Although several studies have explored the application of artificial intelligence in various do- mains of medicine, there is a notable lack of focused reviews that critically assess its role specifically in emergency and trauma triage. Existing literature often discusses artificial intelligence applications broadly, without addressing the unique challenges and opportunities related to triage settings. There- fore, this review aims to fill this gap by synthesizing recent findings on the integration of artificial intel- ligence into triage systems, evaluating its impact on decision-making accuracy, efficiency and potential clinical implications.
This literature review aims to explore the impact of artificial intelligence on triage processes in medi- cal practice, examining how technological advance- ments enhance efficiency, accuracy, and decision- making in patient assessment and prioritization.
A systematic literature search was performed in March 2025 utilizing PubMed, Web of Science and Hrčak databases, covering the period from January 2022 to January 2025. The selection process was
guided by predefined inclusion and exclusion criteria to ensure that only high-quality and relevant studies were included in the review (Table 1). The rationale for selecting this time frame was to focus on the most recent developments in AI applications in triage, as rapid advancements have occurred in the field in the last three years. The literature search was conducted independently by a single researcher under the guid- ance of a mentor, as part of a manuscript prepared for publication. We selected PubMed, Web of Science and Hrčak for our systematic literature review to en- sure comprehensive coverage across international biomedical and multidisciplinary research, alongside regionally specific Croatian scientific output. While this combination maximizes relevance for our topic, it may inherently limit the inclusion of studies from other specialized databases.
Eligible studies included systematic reviews, review articles, original research papers, cross-sectional studies, clinical trials, randomized controlled trials and meta-analyses. These study types were cho- sen due to their methodological rigor and ability to provide comprehensive, evidence-based insights relevant to the research question. To maintain focus on the most recent advancements, only articles pub- lished between 2022 and 2025 were considered. Ad- ditionally, studies had to be published in either Croa- tian or English to be included in the analysis. These language criteria were selected to ensure accurate interpretation of the content by the researchers.
During the search, the authors used the following keywords for the inclusion criteria: “artificial intel- ligence”, “triage”, “medical practice” and “nursing practice”. These terms were selected based on their relevance to the intersection of artificial intelligence and emergency care, as well as their frequent ap- pearance in the literature. Boolean operators such as “AND” and “OR” were used to combine keywords effectively and broaden the search scope. For exam- ple, the query in PubMed was structured as: “artifi- cial intelligence” OR “machine learning” AND “triage” AND “medical practice” AND “nursing practice”. No additional filters were applied regarding article type or study design, to avoid prematurely excluding rel- evant literature.
The screening process followed the PRISMA guide- lines to ensure the selection of relevant studies (Fig- ure 1). Initially, records were identified from three databases, two international and one Croatian data- base: PubMed (n=289), Web of Science (n=15) Hrčak (n=1), resulting in a total of 305 records. Duplicate records (n=13) were removed, along with records excluded for other reasons (n=246), such as being outside the scope of the review, not being peer-re- viewed articles or lacking relevance to the research question. Following these exclusions, 46 records were screened based on their titles and abstracts. During this phase, 15 records were excluded, with the primary reason being the unavailability of a full-text version. The remaining 31 studies were assessed for eligibility, all of which met the inclusion criteria and were subsequently included in the final review. The data from the included studies were synthesized narratively. Key themes, findings, and methodologi- cal characteristics were extracted and summarized in a descriptive manner to identify common patterns, trends, and gaps in the literature. No quantitative synthesis (meta-analysis) was performed due to the heterogeneity of the included studies.
Table 1. Inclusion and exclusion criteria | ||
Inclusion criteria | Exclusion criteria | |
Type/category of the article | Systematic review Review article Original research paper Cross-sectional study Clinical trial Randomized controlled trials Meta-analyses | Letters Editorials Book chapters |
Content (keywords) | Artificial intelligence Triage Medical practice Nursing practice | Other |
Publication date | 2022-2025 | Articles published before 2022 |
Language | Croatian English | Other |
This systematic review included 31 articles pub- lished in the last three years (2022-2025). Due to the heterogeneity in study designs, populations, and outcome measures, a narrative synthesis was con- ducted. These articles were selected to provide an overview of the role of artificial intelligence models, particularly large language models, in emergency department triage and decision-making processes. The results were thematically grouped into key areas relevant to triage accuracy, model reliability, compari- son of AI models with healthcare professionals and clinical implications.
Table 2 includes a detailed compilation of the results from these articles. We included the information about the authors, the year the paper was written, the aim, the type of study and the population of the research.
A total of 12 studies demonstrated that AI mod- els, particularly fine-tuned large language models achieved high levels of triage accuracy, often out- performing or closely matching healthcare profes- sionals in specific scenarios (10,12-15,20,22,25- 27,33,36). For example, GPT-4.0 and Claude-3 Opus models showed sensitivity greater than 77% and specificity greater than 91% in pediatric emergency cases (10). ChatGPT also demonstrated strong agree- ment with expert assessments, achieving accuracy greater than 94% in identifying high-acuity patients (14,20,23,25,27,34,36). In contrast, the AI-powered tool SMASS showed worse performance compared to nurses (11).
While AI models showed high initial performance, their consistency varied depending on context and case complexity. Five studies identified reproducibil- ity issues in simulated disaster scenarios and com- plex patient cases. For example, ChatGPT showed suboptimal repeatability in mass casualty triage (17,18,28,29,31), with performance heavily depend- ent on prompt design and prior training. In contrast,
Figure 1. PRISMA flow diagram
four studies reported improved reliability metrics for fine-tuned models, suggesting that targeted training can enhance consistency (10,20,23,30).
Although artificial intelligence demonstrated strong capabilities, eight studies consistently highlighted the superior accuracy and contextual judgment of trained healthcare professionals, especially in com- plex or high-stakes triage scenarios (16,18,19,23,28- 30,36). In several cases, artificial intelligence models underperformed or showed tendencies toward over- triage or under-triage, underscoring the continued need for human oversight. For instance, a real-time voice AI system for medical record input demon- strated mixed results for completeness and accuracy compared to manual nurse input, despite improving efficiency (40).
Seven studies addressed the broader implications of artificial intelligence implementation. Benefits include reduced administrative workload, earlier identification of critical cases, and improved decision support (15,21-23,26,33). However, significant chal- lenges were also identified, such as data bias, lack of transparency, ethical concerns, and variability in per- formance depending on the model and specific use case (15,30,35).
This literature review has identified a collection of studies that provide substantial evidence supporting the advantageous impact of artificial intelligence in triage across various emergency care settings. The findings highlight the ability of artificial intelligence to enhance accuracy, reduce human bias, improve consistency, and support clinical decision-making in both routine and mass casualty incidents. While arti- ficial intelligence demonstrates significant potential in optimizing emergency triage, its effectiveness var-
ies depending on the model, training data, and the complexity of cases.
However, these benefits must be interpreted cautiously in light of significant limitations. Key limitations include model inconsistency, lack of transparency and poten- tial algorithmic biases. These factors may compromise decision-making accuracy and raise concerns regarding patient safety. Inadequate transparency makes it dif- ficult for healthcare providers to understand and trust AI-generated decisions, potentially leading to resist- ance in adoption or delayed critical interventions.
In addressing the research question, “How does the application of artificial intelligence influence the ac- curacy, consistency, and efficiency of triage in vari- ous emergency care settings?”, the findings indicate a multifaceted impact, demonstrating both significant promise and areas requiring careful consideration.
Artificial intelligence models have demonstrated re- markable accuracy in triage classification, often per- forming at levels comparable to, or exceeding, human clinicians. Fine-tuned GPT-4.0, for example, achieved a sensitivity of 77.1% and specificity of 92.5% in predicting Emergency Severity Index (ESI) levels, while Claude-3 Opus exhibited the highest reliabil- ity among tested AI models, with a Fleiss κ of 0.85 in pediatric triage (10). Similarly, ChatGPT showed strong agreement with human experts in emergency department triage (Kappa = 0.659) and a high speci- ficity of 99.86% in identifying critical cases (11). This addresses the part of the research question by show- ing that artificial intelligence generally enhances tri- age accuracy, particularly in identifying high-acuity patients and predicting ESI levels. However, AI’s ac- curacy is not uniform across all triage systems. When tested on the Canadian Triage and Acuity Scale (CTAS), ChatGPT exhibited only 47.5% accuracy, with a substantial rate of over-triage (38.7%) and under- triage (13.7%), raising concerns about its reliability in certain frameworks (17). Similarly, ChatGPT’s tri- age performance in simulated disaster scenarios us- ing the START protocol was suboptimal (63.9%) due to inconsistencies in repeatability and reproducibility
(16). These variations highlight that while artificial intelligence holds promise, its effectiveness is highly dependent on the triage system used and the com- plexity of cases it encounters.
Table 2. Overview of the research findings on technological advancements in triage | ||||||
Authors, year | Type of study | Population | AI Models Evaluated | Comparators | Key findings | AI vs. Human performance |
Ho et al. (2025). (10) | Original research paper | 70 pediatric vignettes (ESI Handbook v4) | Claude-3 Opus, GPT-4.0 (fine- tuned), Mistral- Large | N/A | Claude-3 Opus: Sensitivity 80.6%, Specificity 91.3%, F1 73.9%; Fine- tuned GPT-4.0 improved: F1 74.6%, P < 0.04 | AI models (especially fine-tuned) show strong accuracy and reliability (κ: 0.85) |
Lindner, Ravioli (2025). (11) | Original research paper (retrospective) | 1021 adult ED patients | SMASS (AI- powered assessment tool) | MTS (Manchester Triage System) | SMASS showed significant over-/ under-triage vs. MTS (Kappa 0.167). Sensitivity 62%, specificity 73% for acute/ non-acute. | SMASS performed worse than human- applied MTS, requires significant training on real-world ED data to improve accuracy and consistency. |
Porcellato et al. (2025). (12) | Systematic review | 24 studies on critical care patients | Diverse AI techniques (machine learning, deep learning, LLMs | Varies by included study (traditional methods, other AI models) | Predictive models show varying performance (e.g., one AI-ECG model showed 76% accuracy, 73% sensitivity= | AI significantly enhances human decision-making in trauma triage, outperforming conventional tools (AUC-ROC 0.09), though study variations prevent universally firm conclusions. |
Arslan et al. (2024). (13) | Observational study | 468 adult ED patients | ChatGPT, Copilot | Triage nurses | ChatGPT: 66.5%, Copilot: 61.8%, Nurses: 65.2%; AI better at identifying high- risk patients (87.8% vs. 32.7%) | AI outperformed nurses in high-risk identification, more consistent across ages |
Colakca et al. (2024). (14) | Cross-sectional study | 745 adult ED patients | ChatGPT-4 | Expert triage | High agreement (Kappa = 0.659); Specificity: ESI-1: 99.86%, ESI-2: 95.38% | AI highly effective at identifying critical cases |
Di Sarno et al. (2024). (15) | Literature review | Pediatric patients in emergency medicine settings | AI-driven Clinical Decision Support Systems (CDS), Socially Assistive Robots (SARs) | Traditional clinical assessment methods | AI improves triage accuracy, early sepsis detection, and traumatic brain injury evaluation; SARs reduce pediatric stress | AI models outperform traditional methods in accuracy and efficiency, but issues with data bias, transparency, and clinical integration remain |
Table 2. Overview of the research findings on technological advancements in triage | ||||||
Authors, year | Type of study | Population | AI Models Evaluated | Comparators | Key findings | AI vs. Human performance |
Eraybar et al. (2024). (16) | Observational study | 86 clinicians (ED professionals) | ChatGPT, Google Bard | Human professionals | Professionals: 30.7 correct, AI: 25.5; No significant AI difference (p=0.821) | Humans outperformed AI; AI not yet as accurate |
Franc et al. (2024). (17) | Original research paper | 391 disaster triage vignettes | ChatGPT-4 | START protocol (no human comparator) | Accuracy: 63.9%; Poor reproducibility; caution advised | AI performance suboptimal for disaster triage |
Franc, Cheng et al. (2024). (18) | Original research paper | 61 CTAS vignettes | ChatGPT | Canadian Triage Scale | Accuracy: 47.5%; Over-triage 38.7%, Under-triage 13.7% | ChatGPT showed high variability and low reliability |
Kim et al. (2024). (19) | Original research paper | 202 virtual patient cases | ChatGPT-3.5, ChatGPT-4.0 | Human paramedics | GPT-4.0 κ=0.523 vs. 3.5 κ=0.320; Human κ=0.646 | AI less reliable than humans, but GPT-4.0 better than 3.5 |
Liu et al. (2024). (20) | Original research paper – retrospective and prospective cohort study | Retrospective - 30 outpatient medical records Prospective - manual vs. ChatGPT triage for 45 outpatients based on age, gender, and symptoms | ChatGPT | Manual triage | Prospective: 93.3– 100% agreement; Retrospective: 17/30 rated 9.5–10 | High consistency with manual triage |
Mani et al. (2024). (21) | Review article | Patients, healthcare providers, and AI systems in emergency departments | AI tools for triage, patient monitoring, diagnosis, treatment planning, and decision support | Conventional nursing workflows | AI applications improve triage accuracy, monitoring, diagnosis, treatment planning, and decision- making, enhancing patient outcomes and workflow efficiency. | AI enhances performance but requires addressing data security, ethics, algorithm reliability, and staff training to achieve effective implementation. |
Ventura et al. (2024). (22) | Literature review | Patients that needed triage assessment | Deep learning models for injury diagnosis and outcome prediction | Traditional triage methods | Deep learning achieved high accuracy in diagnosing traumatic injuries and predicting hospitalization, mortality, and ICU admission. | AI outperformed traditional triage methods in accuracy and predictive performance. |
Masanneck et al. (2024). (23) | Original research paper | 124 anonymized ED case vignettes | ChatGPT (GPT- 3.5, GPT-4) | Untrained doctors | GPT-4 ≈ untrained doctors; GPT-3.5 worse; Over-triage common | LLMs showed potential, but didn’t match experts |
Table 2. Overview of the research findings on technological advancements in triage | ||||||
Authors, year | Type of study | Population | AI Models Evaluated | Comparators | Key findings | AI vs. Human performance |
Preiksaitis et al. (2024). (24) | Review article | Patients needing triage assessment | Large Language Models (LLMs) | Traditional triage and administrative methods | LLMs improve emergency care by enhancing real-time triage, recognizing patient urgency earlier, reducing administrative workload, and supporting patient-centered care. | LLMs show potential to support clinicians by increasing efficiency and triage accuracy, but direct performance comparisons with humans were not detailed. |
Sorich et al. (2024). (25) | Original research paper | 48 case vignettes | GPT-4o, Claude 3.5, Gemini 1.5 Pro | N/A | Triage accuracy ~92% across models; Claude-3.5 best overall | AI shows strong diagnostic and triage performance |
Tyler et al. (2024). (26) | Review article | Patients admitted to emergency departments in the USA | AI and Machine Learning models for triage, specific models not mentioned | Traditional triage systems (e.g., Emergency Severity Index) | AI and ML models improved triage by reducing mis- triage, enhancing prediction of critical outcomes, and outperforming conventional systems in forecasting admissions, disease identification, and deterioration. | AI models outperformed human-based systems in triage accuracy, efficiency, and resource allocation. |
Yi et al. (2024). (27) | Systematic review | Patients needing triage assessment | AI-based triage models, specific models not mentioned | Manual triage methods | AI demonstrated high accuracy (80.5%–99.1%), improved triage speed, reduced mis-triage, and enabled urgency classification and prognosis prediction more effectively. | AI triage outperformed manual methods in both accuracy and time efficiency. |
Mayerhoffer H. (2024). (28) | Original research paper | AI triage categorization | ChatGPT | Traditional triage | Correct in 43.33%; Tendency to over- triage for safety | AI errs conservatively but lacks high precision |
Table 2. Overview of the research findings on technological advancements in triage | ||||||
Authors, year | Type of study | Population | AI Models Evaluated | Comparators | Key findings | AI vs. Human performance |
Fraser et al. (2023). (29) | Original research paper | Original research paper / 40 emergency department patients | ChatGPT 3.5, ChatGPT 4.0, WebMD, Ada Health | ED diagnoses and physician reviews | ChatGPT 3.5 had the highest diagnostic accuracy (40% top-1, 63% top-3), but the highest unsafe triage rate (41%). ChatGPT 4.0 had better triage agreement (76%) and lower unsafe rate (22%), making it more reliable. | ChatGPT models showed moderate diagnostic accuracy but varied in triage safety—ChatGPT 4.0 performed better than 3.5 in triage alignment with physicians. |
Gan Uddin et al. (2023). (30) | Cross-sectional study | Simulated MCI scenarios | ChatGPT vs. Google Bard | Medical students | Bard: 60%, ChatGPT: 26.7%, Students: 64.3% | Bard comparable to students; ChatGPT significantly lower |
Gan, Ogbodo et al. (2023). (31) | Cross-sectional study | Simulated MCI scenarios | ChatGPT vs. Google Bard | Medical students | Bard: 60%, ChatGPT: 26.7%, Students: 64.3% | Bard comparable to students; ChatGPT significantly lower |
Gebrael et al. (2023). (32) | Original research paper | 56 prostate cancer patients in ED | ChatGPT | ER physicians | ChatGPT showed 95.7% sensitivity for admission decisions, 18.2% specificity for discharges, aligned with physician diagnosis in 87.5% of cases, used more accurate medical terminology, and offered more comprehensive recommendations. | ChatGPT showed cautious but accurate diagnostic support; outperformed physicians in terminology and completeness but lacked discharge precision. |
Adebayo et al. (2023). (33) | Systematic review | Systematic review / Triage patients | AI, ML, DL models | Conventional triage tools | AI-based models significantly improved prediction of mortality, hospitalization, and ICU admission, surpassing traditional triage tools. | AI models statistically outperformed conventional tools in predictive accuracy for critical outcomes. |
Jacob J. (2023). (34) | Original research paper | Polytrauma scenarios | ChatGPT | ESI and Australasian Triage Scales | ChatGPT accurately classified polytrauma patients with one initial misclassification corrected upon review; AI demonstrated potential in rapid classification. | ChatGPT showed strong potential for fast and accurate triage support with self-correction capability. |
Table 2. Overview of the research findings on technological advancements in triage | ||||||
Authors, year | Type of study | Population | AI Models Evaluated | Comparators | Key findings | AI vs. Human performance |
Masoumian et al. (2023). (35) | Systematic review | Triage patients | Various AI applications, specific models not mentioned | Not directly compared | AI used for triage, disease prediction, emergency management; ethical concerns highlighted, particularly regarding transparency. | AI showed clinical potential, but lack of transparency challenges trust and adoption. |
Sarbay et al. (2023). (36) | Cross-sectional study | 50 ESI scenarios | ChatGPT | Emergency Medicine (EM) specialists | ChatGPT showed fair agreement (κ=0.341), 22% over-triage, 18% under-triage, strong in high- acuity cases with 76.2% sensitivity and 93.1% specificity. | ChatGPT effective in high-acuity triage; moderate agreement with specialists suggests supportive role. |
Boonstra, Laven. (2022). (37) | Systematic literature review | ED patients | General AI tools, specific models not mentioned | Not directly compared | AI improved decision- making, triage efficiency, reduced overcrowding and clinician burden; designed to support, not replace clinicians. | AI supports human clinicians by optimizing workflow and resource allocation. |
Ilicki J. (2022). (38) | Systematic review | Triage patients | Patient- operated AI triage systems | Not directly compared | Main limitations were epistemological, ontological, and methodological; caution required in interpreting accuracy claims. | AI triage systems need critical appraisal; performance evaluation is complex and context- dependent. |
Mueller et al. (2022). (39) | Review article | Triage patients | ML applications in emergency medicine, specific models not mentioned | Traditional triage | ML enhances triage by analyzing patient data, improving prioritization, reducing delays, and predicting risk. | AI enhances traditional methods, offering faster and more precise risk assessment. |
Cho et al. (2022). (40) | Original research paper | 1063 triage cases from a Level 1 ED (19 ED nurses) | RMIS-AI (Real- time medical record input assistance system with voice AI, utilizing voice recognition & NLP) | Manual EMR input by nurses | RMIS-AI significantly shortened triage time (204s vs 231s). Mixed results for record completeness and accuracy compared to manual input. | RMIS-AI improved efficiency, but accuracy/ completeness varied, suggesting a supportive rather than replacement role for human nurses. |
Such inconsistency in performance introduces risk when AI systems are used without adequate human oversight. This underscores the need for stringent validation of AI tools before widespread deployment.
Artificial intelligence improves triage by reducing human biases that lead to patient misclassification. In one study, ChatGPT identified high-risk patients more accurately than nurses (87.8% vs. 32.7%) and showed consistent accuracy across age groups, mini- mizing age-related bias (10). Additionally, AI models enhance triage consistency, with ChatGPT-4.0 achiev- ing an inter-rater agreement of κ=0.523, though still lower than human professionals (κ=0.646). However, artificial intelligence struggles with complex cases— ChatGPT-3.5 had poor performance in severe emer- gencies (κ=0.067), highlighting the need for further improvements in high-risk triage scenarios (18).
Artificial intelligence has also demonstrated potential in mass casualty incidents, where rapid and accurate triage is essential for optimizing patient outcomes. Studies have shown that AI can improve triage per- formance in these scenarios. After being trained on the START protocol, ChatGPT’s accuracy in MCI triage reached 80%, surpassing medical students (29). How- ever, when compared to Google Bard, ChatGPT under- performed, achieving only 26.67% accuracy versus Bard’s 60% (30). These results highlight the variability across different models, underscoring the need for fur- ther training and validation before deployment in dis- aster response. In real-world mass casualty incidents, AI’s lack of contextual awareness and inability to adapt dynamically to chaotic environments further compli- cates its practical utility. The risk of over-reliance on AI in such high-pressure settings could delay life-saving interventions without immediate human correction.
Artificial intelligence in triage faces challenges such as data bias, transparency issues, and inconsistent reliability across models (20). Ultimately, AI is most effective in a hybrid model, complementing human
expertise rather than replacing it. Continuous as- sessment and refinement will be essential for its safe and effective use in emergency medicine. Future re- search should focus on improving AI’s performance in complex triage cases, enhancing model interpretabil- ity, and ensuring seamless integration with existing healthcare systems. A promising direction involves collaboration among multiple AI models, as demon- strated in studies where LLMs worked together to achieve a diagnostic accuracy of 98% (24).
Equally important are the practical barriers to imple- mentation. These include high development and inte- gration costs, the need for technical training, resist- ance from clinical staff, and unresolved regulatory and legal frameworks. Establishing trust in AI systems will require transparent reporting, independent validation, and ethical oversight. Comparing the performance of AI models and nurses should be a key focus of future research to clarify the capabilities and limitations of AI. Tasks that require complex thinking and emotional understanding should still be handled by humans, be- cause AI does not yet understand context or have mor- al judgment. To deal with current problems, we need better solutions like making AI more understandable, reducing bias when creating datasets and training AI systems to adjust as medical environments change. We also need real-life studies and pilot programs to test how AI performs in practice. These should assess patient outcomes, staff acceptance, and cost-effec- tiveness across different healthcare settings.
This review shows that artificial intelligence has the potential to improve triage accuracy, reduce certain human biases and assist in clinical decision-making across emergency settings. Studies show artificial intelligence driven models improve risk identifica- tion, triage consistency, and emergency care assess- ments, reinforcing their value in healthcare. Howev- er, artificial intelligence models still face important challenges such as model inconsistency, limited transparency and varying performance across clinical contexts. These challenges can affect patient safety and decision making. Current evidence suggests arti- ficial intelligence is best used as a support tool rather than an independent triage system.
Future research should focus on validating AI tools in real-world clinical environments, improving their per- formance in complex and high-risk cases, and ensuring transparency to build clinician trust. Studies should as- sess not only diagnostic accuracy but also patient out- comes, staff acceptance and cost-effectiveness. Ef- forts are also needed to develop guidelines for ethical use, legal accountability, and integration into existing emergency protocols. Ongoing validation and refine- ment will be essential to ensuring safe and effective deployment in emergency medicine.
Conceptualization and methodology (BI, HM); Data curation and formal analysis (HM); Investigation and project administration (HM); Writing – original draft (BI, HM); and Review & editing (BI, HM). All authors have approved the final manuscript.
The authors declare no conflicts of interest.
Not applicable.
This research did not receive any specific grant from funding agencies in the public, commercial, or not- for-profit sectors.
Tang X, Li X, Ding Y, Song M, Bu Y. The pace of artifi- cial intelligence innovations: speed, talent, and trial- and-error. J Informetr. 2020;14(4):101094. https://doi. org/10.1016/j.joi.2020.101094
Cho B, Geng E, Arvind V, Valliani A, Tang J, Schwartz J et al. Understanding artificial intelligence and predictive analytics. JBJS Rev. 2022;10(3). https://doi.org/10.2106/ JBJS.RVW.21.00142
LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521:436–44. https://doi.org/10.1145/3448250
Jenkins JL, McCarthy ML, Sauer LM, Green GB, Stuart S, Thomas TL et al. Mass-casualty triage: time for an eviden- ce-based approach. Prehosp Disaster Med. 2008;23(1):3–
8. https://doi.org/10.1017/s1049023x00005471
Malik NS, Chernbumroong S, Xu Y, Vassallo J, Lee J, Bowley DM et al. The BCD triage sieve outperforms all existing major incident triage tools: comparative analysis using the UK national trauma registry population. EclinicalMe- dicine. 2021;36:100888. https://doi.org/10.1016/j. eclinm.2021.100888
Thawley A, Aggar C, Williams N. The educational needs of triage nurses. Health Education in Practice: Journal of Research for Professional Learning. 2020;3(1):26-38. https://doi.org/10.33966/hepj.3.1.14121
Frykberg ER, Tepas JJ. Terrorist bombings: lessons learned from belfast to beirut. Ann Surg. 1988;208(5):569–76. https://doi.org/10.1097/00000658-198811000-00005
Ying Y, Huang B, Zhu Y, Jianx X, Dong J, Ding Yet al. Comparison of five triage tools for identifying mortality risk and injury severity of multiple trauma patients admitted to the emer- gency department in the daytime and nighttime: a retros- pective study. Appl Bionics Biomech. 2022;25;9368920. https://doi.org/10.1155/2022/9368920
Rutschmann OT, Kossovsky M, Geissbühler A, Perneger TV, Vermeulen B, Simon J et al. Interactive triage simulator re- vealed important variability in both process and outcome of emergency triage. J Clin Epidemiol. 2006;59(6):615–21. https://doi.org/10.1016/j.jclinepi.2005.11.003
Ho B, Lu M, Wang X, Butler R, Park J, Ren D. Evaluation of Generative Artificial Intelligence Models in Predic- ting Pediatric Emergency Severity Index Levels. Pediatr Emerg Care. 2025;41(4):251-5. https://doi.org/10.1097/ PEC.0000000000003315.
Lindner G, Ravioli S. Performance of the artificial intelli- gence-based Swiss medical assessment system versus Manchester triage system in the emergency depar- tment: A retrospective analysis. The American Journal of Emergency Medicine. 2025;94:46-9. https://doi. org/10.1016/j.ajem.2025.04.023
Porcellato E, Lanera C, Ocagli H, Danielis M. Exploring Appli- cations of Artificial Intelligence in Critical Care Nursing: A Systematic Review. Nursing Reports. 2025;15(2):55. https://doi.org/10.3390/nursrep15020055
Arslan B, Nuhoglu C, Satici MO, Altinbilek E. Evaluating LLM-based generative AI tools in emergency triage: A comparative study of ChatGPT Plus, Copilot Pro, and tria- ge nurses. Am J Emerg Med. 2025;89:174-81. https://doi. org/10.1016/j.ajem.2024.12.024
Colakca C, Ergın M, Ozensoy HS, Sener A, Guru S, Ozhase- nekler A. Emergency department triaging using ChatGPT based on emergency severity index principles: a cross- sectional study. Sci Rep. 2024:27;14(1):22106. https:// doi.org/10.1038/s41598-024-73229-7
Di Sarno L, Caroselli A, Tonin G, Graglia B, Pansini V, Cau- sio FA et al. Artificial Intelligence in Pediatric Emergency Medicine: Applications, Challenges, and Future Perspec- tives. Biomedicines. 2024:30;12(6):1220. https://doi. org/10.3390/biomedicines12061220
Eraybar S, Dal E, Aydin MO, Begenen M. Transforming emergency triage: A preliminary, scenario-based cross-sectional study comparing artificial intelligence models and clinical expertise for enhanced accuracy. Bratisl Lek Listy. 2024;125(11):738-
43. https://doi.org/10.4149/BLL_2024_114
Franc JM, Hertelendy AJ, Cheng L, Hata R, Verde M. Accu- racy of a Commercial Large Language Model (ChatGPT) to Perform Disaster Triage of Simulated Patients Using the Simple Triage and Rapid Treatment (START) Pro- tocol: Gage Repeatability and Reproducibility Stu- dy. J Med Internet Res. 2024;26:e55648. https://doi. org/10.2196/55648
Franc JM, Cheng L, Hart A, Hata R, Hertelendy A. Repe- atability, reproducibility, and diagnostic accuracy of a commercial large language model (ChatGPT) to perform emergency department triage using the Canadian tria- ge and acuity scale. CJEM. 2024;26(1):40-6. https://doi. org/10.1007/s43678-023-00616-w
Kim JH, Kim SK, Choi J, Lee Y. Reliability of ChatGPT for performing triage task in the emergency department using the Korean Triage and Acuity Scale. Digit He- alth. 2024;10:20552076241227132. https://doi. org/10.1177/20552076241227132
Liu X, Lai R, Wu C, Yan C, Gan Z, Yang Y et al. Assessing the utility of artificial intelligence throughout the triage outpatients: a prospective randomized controlled clinical study. Front Public Health. 2024;12:1391906. https:// doi.org/10.3389/fpubh.2024.1391906
Mani Z, Albagawi B. AI frontiers in emergency care: the next evolution of nursing interventions. Front Public Health. 2024;12:1439412. https://doi.org/10.3389/ fpubh.2024.1439412
Ventura CAI, Denton EE, David JA. Artificial Intelligence in Emergency Trauma Care: A Preliminary Scoping Review. Med Devices (Auckl). 2024;17:191-211. https://doi. org/10.2147/MDER.S467146
Masanneck L, Schmidt L, Seifert A, Kölsche T, Huntemann N, Jansen R, et al. Triage Performance Across Large Lan- guage Models, ChatGPT, and Untrained Doctors in Emer- gency Medicine: Comparative Study. J Med Internet Res. 2024;26:e53297. https://doi.org/10.2196/53297
Preiksaitis C, Ashenburg N, Bunney G, Chu A, Kabeer R, Riley F et al. The Role of Large Language Models in Transforming Emergency Medicine: Scoping Re- view. JMIR Med Inform. 2024;12:e53787. https://doi. org/10.2196/53787
Sorich MJ, Mangoni AA, Bacchi S, Menz BD, Hopkins AM. The Triage and Diagnostic Accuracy of Frontier Large Language Models: Updated Comparison to Physician Per- formance. J Med Internet Res. 2024;26:e67409. https:// doi.org/10.2196/67409
Tyler S, Olis M, Aust N, Patel L, Simon L, Triantafyllidis C et al. Use of Artificial Intelligence in Triage in Hospi- tal Emergency Departments: A Scoping Review. Cure- us. 2024;16(5):e59906. https://doi.org/10.7759/cure- us.59906
Yi N, Baik D, Baek G. The effects of applying artificial in- telligence to triage in the emergency department: A systematic review of prospective studies. J Nurs Scholarsh. 2024;57(1):105-118. https://doi.org/10.1111/jnu.13024
Mayerhoffer H. The Future of Triage: The Analysis of Traditional Methods Compared to ChatGPT. Cro- atian Nursing Journal. 2024;8(1):29-36. https://doi. org/10.24141/2/8/1/3
Fraser H, Crossland D, Bacher I, Ranney M, Madsen T, Hilliard R. Comparison of Diagnostic and Triage Accuracy of Ada Health and WebMD Symptom Checkers, ChatGPT, and Physicians for Patients in an Emergency Depar- tment: Clinical Data Analysis Study. JMIR Mhealth Uhe- alth. 2023;11:e49995. https://doi.org/10.2196/49995
Gan RK, Uddin H, Gan AZ, Yew YY, González PA. ChatGPT’s performance before and after teaching in mass casualty incident triage. Sci Rep. 2023;13(1):20350. https://doi. org/10.1038/s41598-023-46986-0
Gan RK, Ogbodo JC, Wee YZ, Gan AZ, González PA. Perfor- mance of Google bard and ChatGPT in mass casualty in- cidents triage. Am J Emerg Med. 2024;75:72-78. https:// doi.org/10.1016/j.ajem.2023.10.034
Gebrael G, Sahu KK, Chigarira B, Tripathi N, Mathew Thomas V, Sayegh N et al. Enhancing triage efficiency and accuracy in emergency rooms for patients with me- tastatic prostate cancer: a retrospective analysis of ar- tificial intelligence-assisted triage using ChatGPT 4.0. Cancers. 2023;15(14):3717. https://doi.org/10.3390/ cancers15143717
Adebayo O, Bhuiyan ZA, Ahmed Z. Exploring the effective- ness of artificial intelligence, machine learning and deep learning in trauma triage: A systematic review and meta- analysis. Digit Health. 2023;9:20552076231205736. https://doi.org/10.1177/20552076231205736
Jacob J. ChatGPT: Friend or Foe?-Utility in Trauma Tria- ge. Indian J Crit Care Med. 2023;27(8):563-6. https://doi. org/10.5005/jp-journals-10071-24498
Masoumian Hosseini M, Masoumian Hosseini ST, Qayumi K, Ahmady S, Koohestani HR. The Aspects of Running Artificial Intelligence in Emergency Care; a Scoping Re- view. Arch Acad Emerg Med. 2023;11(1):e38. https://doi. org/10.22037/aaem.v11i1.1974
Sarbay İ, Berikol GB, Özturan İU. Performance of emer- gency triage prediction of an open access natural lan- guage processing based chatbot application (ChatGPT): A preliminary, scenario-based cross-sectional study. Turk J Emerg Med. 2023;23(3):156-61. https://doi.org/10.4103/ tjem.tjem_79_23
Boonstra A, Laven M. Influence of artificial intelligence on the work design of emergency department clinicians a systematic literature review. BMC Health Serv Res. 2022;22(1):669. https://doi.org/10.1186/s12913-022-
08070-7
Ilicki J. Challenges in evaluating the accuracy of AI-conta- ining digital triage systems: A systematic review. PLoS One. 2022;17(12):e0279636. https://doi.org/10.1371/ journal.pone.0279636
Mueller B, Kinoshita T, Peebles A, Graber MA, Lee S. Artificial intelligence and machine learning in emer- gency medicine: a narrative review. Acute Med Surg. 2022;9(1):e740. https://doi.org/10.1002/ams2.740
Cho A, Min IK, Hong S, Chung HS, Lee HS, Kim JH. Effect of applying a real-time medical record input assistance system with voice artificial intelligence on triage task performance in the emergency department: prospec- tive interventional study. JMIR Medical Informatics. 2022;10(8):e39892. https://doi.org/10.2196/39892
Table 1. Excluded studies | |||
Author | Name | Year | Reason of exclusion |
Abdulgader, S. M. | Diagnosing Tuberculosis: What Do New Technologies Allow Us to (Not) Do? | 2022 | Not related to the topic |
Abi-Rafeh, J. | Complications Following Facelift and Neck Lift: Implementation and Assessment of Large Language Model and Artificial Intelligence (ChatGPT) Performance Across 16 Simulated Patient Presentations | 2023 | Not related to the topic |
Abou Chaar, M. K. | ChatGPT vs Expert-Guided Care Pathways for Postesophagectomy Symptom Management | 2024 | Not related to the topic |
Acharji, S. | Prognostic significance of elevated baseline troponin in patients with acute coronary syndromes and chronic kidney disease treated with different antithrombotic regimens: a substudy from the ACUITY trial | 2012 | Published before 2022. |
Acosta, J. N. | The Need for Medical Artificial Intelligence That Incorporates Prior Images | 2022 | Not related to the topic |
Adams, S. J. | Artificial Intelligence Solutions for Analysis of X-ray Images | 2021 | Not related to the topic |
Agarwal, S. | Systematic Review of Artificial Intelligence for Abnormality Detection in High- volume Neuroimaging and Subgroup Meta-analysis for Intracranial Hemorrhage Detection | 2023 | Not related to the topic |
Aggelidis, X. | Tele-Monitoring Applications in Respiratory Allergy | 2024 | Not related to the topic |
Ahmed, A. | Role of Digital Health During Coronavirus Disease 2019 Pandemic and Future Perspectives | 2022 | Not related to the topic |
Akkerhuis, K. M. | Recurrent ischemia during continuous 12-lead ECG-ischemia monitoring in patients with acute coronary syndromes treated with eptifibatide: relation with death and myocardial infarction. PURSUIT ECG-Ischemia Monitoring Substudy Investigators. Platelet... | 2000 | Published before 2022. |
Alizadehsani, R. | Coronary artery disease detection using artificial intelligence techniques: A survey of trends, geographical differences and diagnostic features 1991-2020 | 2021 | Published before 2022. |
AlNuaimi, D. | The role of artificial intelligence in plain chest radiographs interpretation during the Covid-19 pandemic | 2022 | Not related to the topic |
Altamimi, I. | Snakebite Advice and Counseling from Artificial Intelligence: An Acute Venomous Snakebite Consultation With ChatGPT | 2023 | Not related to the topic |
Amundson, S. A. | Transcriptomics for radiation biodosimetry: progress and challenges | 2023 | Not related to the topic |
Anderson, P. | Stress granules: the Tao of RNA triage | 2008 | Published before 2022. |
Ankolekar, A. | Using artificial intelligence and predictive modelling to enable learning healthcare systems (LHS) for pandemic preparedness | 2024 | Not related to the topic |
CADTH Horizon Scans | 2023 | Not related to the topic | |
CADTH Horizon Scans | 2023 | Duplicate | |
Ayoub, M. | Mind + Machine: ChatGPT as a Basic Clinical Decisions Support Tool | 2023 | Not related to the topic |
Bahl, M. | Updates in Artificial Intelligence for Breast Imaging | 2022 | Not related to the topic |
Bahl, M. | Artificial Intelligence for Breast Ultrasound: AJR Expert Panel Narrative Review | 2024 | Not related to the topic |
Baker, A. | A comparison of artificial intelligence and human doctors for the purpose of triage and diagnosis | 2020 | Published before 2022. |
Barlow, A. | Pulmonary arterial hypertension in the emergency department: A focus on medication management | 2021 | Published before 2022. |
Batra, P. | Artificial Intelligence in Teledentistry | 2022 | Not related to the topic |
Baughan, N. | Past, Present, and Future of Machine Learning and Artificial Intelligence for Breast Cancer Screening | 2022 | Not related to the topic |
Behrens, A. J. | Glycosylation profiling to evaluate glycoprotein immunogens against HIV-1 | 2017 | Published before 2022. |
Ben Alaya, I. | Applications of artificial intelligence for DWI and PWI data processing in acute ischemic stroke: Current practices and future directions | 2022 | Not related to the topic |
Ben Alaya, I. | Automatic triaging of acute ischemic stroke patients for reperfusion therapies using Artificial Intelligence methods and multiple MRI features: A review | 2023 | Not related to the topic |
Bhattaram, S. | ChatGPT: The next-gen tool for triaging? | 2023 | Full text not written |
Table 1. Excluded studies | |||
Author | Name | Year | Reason of exclusion |
Biswas, S. | Utility of artificial intelligence-based large language models in ophthalmic care | 2024 | Not related to the topic |
Biswas, S. | Utility of artificial intelligence-based large language models in ophthalmic care | 2024 | Duplicate |
Boochoon, K. | Deep Learning for the Assessment of Facial Nerve Palsy: Opportunities and Challenges | 2023 | Not related to the topic |
Boyd, C. J. | Artificial Intelligence as a Triage Tool during the Perioperative Period: Pilot Study of Accuracy and Accessibility for Clinical Application | 2024 | Not related to the topic |
Buchlak, Q. D. | Charting the potential of brain computed tomography deep learning systems | 2022 | Not related to the topic |
Bydon, M. | Commentary: A Quantitative Assessment of Chat-GPT as a Neurosurgical Triaging Tool | 2024 | Not related to the topic |
Cao, X. F. | Application of artificial intelligence in digital chest radiography reading for pulmonary tuberculosis screening | 2021 | Published before 2022. |
Cascella, M. | The Breakthrough of Large Language Models Release for Medical Applications: 1-Year Timeline and Perspectives | 2024 | Not related to the topic |
Casterella, P. J. | Review of the 2005 American College of Cardiology, American Heart Association, and Society for Cardiovascular Interventions guidelines for adjunctive pharmacologic therapy during percutaneous coronary interventions: practical implications, new clinical... | 2008 | Published before 2022. |
Chandrabhatla, A. S. | Artificial Intelligence and Machine Learning in the Diagnosis and Management of Stroke: A Narrative Review of United States Food and Drug Administration- Approved Technologies | 2023 | Not related to the topic |
Chennareddy, S. | Portable stroke detection devices: a systematic scoping review of prehospital applications | 2022 | Not related to the topic |
Choe, J. | Artificial Intelligence in Lung Imaging | 2022 | Not related to the topic |
Chu, K. | Evaluating risk stratification scoring systems to predict mortality in patients with COVID-19 | 2021 | Published before 2022. |
Chu, L. C. | Pancreatic ductal adenocarcinoma staging: a narrative review of radiologic techniques and advances | 2024 | Not related to the topic |
Cicero, M. X. | 60 seconds to survival: A pilot study of a disaster triage video game for prehospital providers | 2017 | Published before 2022. |
Ciecierski-Holmes, T. | Artificial intelligence for strengthening healthcare systems in low- and middle- income countries: a systematic scoping review | 2022 | Not related to the topic |
Corbacho Abelaira, M. D. | Use of Conventional Chest Imaging and Artificial Intelligence in COVID-19 Infection. A Review of the Literature | 2021 | Published before 2022. |
Dafni, M. F. | Empowering cancer prevention with AI: unlocking new frontiers in prediction, diagnosis, and intervention | 2024 | Not related to the topic |
Daneshjou, R. | Lack of Transparency and Potential Bias in Artificial Intelligence Data Sets and Algorithms: A Scoping Review | 2021 | Published before 2022. |
Dangi, R. R. | Transforming Healthcare in Low-Resource Settings with Artificial Intelligence: Recent Developments and Outcomes | 2024 | Not related to the topic |
Daripa, B. | Artificial Intelligence-Aided Headache Classification Based on a Set of Questionnaires: A Short Review | 2022 | Not related to the topic |
Dasegowda, G. | Suboptimal Chest Radiography and Artificial Intelligence: The Problem and the Solution | 2023 | Not related to the topic |
David, D. | The use of artificial intelligence based chat bots in ophthalmology triage | 2024 | Not related to the topic |
Davidović, M. | Facility-Based Indicators to Manage and Scale Up Cervical Cancer Prevention and Care Services for Women Living With HIV in Sub-Saharan Africa: a Three- Round Online Delphi Consensus Method | 2024 | Not related to the topic |
Delgado, J. | Bias in algorithms of AI systems developed for COVID-19: A scoping review | 2022 | Not related to the topic |
Delsoz, M. | The Use of ChatGPT to Assist in Diagnosing Glaucoma Based on Clinical Case Reports | 2023 | Not related to the topic |
Denkinger, C. M. | Defining the needs for next generation assays for tuberculosis | 2015 | Published before 2022. |
Desai, S. M. | Direct Transfer to the Neuroangiography Suite for Patients with Stroke | 2023 | Not related to the topic |
Table 1. Excluded studies | |||
Author | Name | Year | Reason of exclusion |
Desmet, C. M. | Factors Affecting the Quality of Tooth Enamel for In Vivo EPR-Based Retrospective Biodosimetry | 2016 | Published before 2022. |
Dheda, K. | A position statement and practical guide to the use of particulate filtering facepiece respirators (N95, FFP2, or equivalent) for South African health workers exposed to respiratory pathogens including Mycobacterium tuberculosis and SARS-CoV-2 | 2021 | Published before 2022. |
Dias Gonçalves Lima, F. | The Accuracy of Anal Swab-Based Tests to Detect High-Grade Anal Intraepithelial Neoplasia in HIV-Infected Patients: A Systematic Review and Meta-analysis | 2019 | Published before 2022. |
DiCarlo, A. L. | Radiation injury after a nuclear detonation: medical consequences and the need for scarce resources allocation | 2011 | Published before 2022. |
Dimitsaki, S. | Benchmarking of Machine Learning classifiers on plasma proteomic for COVID-19 severity prediction through interpretable artificial intelligence | 2023 | Not related to the topic |
Doeleman, T. | Artificial intelligence in digital pathology of cutaneous lymphomas: A review of the current state and future perspectives | 2023 | Not related to the topic |
Dossantos, J. | Eyes on AI: ChatGPT’s Transformative Potential Impact on Ophthalmology | 2023 | Not related to the topic |
Eaby-Sandy, B. | Side effects of targeted therapies: rash | 2014 | Published before 2022. |
Ebrahimian, S. | FDA-regulated AI Algorithms: Trends, Strengths, and Gaps of Validation Studies | 2022 | Not related to the topic |
Ellis, M. J. | Ki67 Proliferation Index as a Tool for Chemotherapy Decisions During and After Neoadjuvant Aromatase Inhibitor Treatment of Breast Cancer: Results from the American College of Surgeons Oncology Group Z1031 Trial (Alliance) | 2017 | Published before 2022. |
Escalé-Besa, A. | The Use of Artificial Intelligence for Skin Disease Diagnosis in Primary Care Settings: A Systematic Review | 2024 | Not related to the topic |
Fanni, S. C. | Artificial Intelligence-Based Software with CE Mark for Chest X-ray Interpretation: Opportunities and Challenges | 2023 | Not related to the topic |
Feit, F. | Safety and efficacy of bivalirudin monotherapy in patients with diabetes mellitus and acute coronary syndromes: a report from the ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy) trial | 2008 | Published before 2022. |
Flood, A. B. | Benefits and challenges of in vivo EPR nail biodosimetry in a second tier of medical triage in response to a large radiation event | 2023 | Not related to the topic |
Fox, K. A. | Management of acute coronary syndromes: an update | 2004 | Published before 2022. |
Freeman, K. | Use of artificial intelligence for image analysis in breast cancer screening programmes: systematic review of test accuracy | 2021 | Published before 2022. |
Frosolini, A. | The Role of Large Language Models (LLMs) in Providing Triage for Maxillofacial Trauma Cases: A Preliminary Study | 2024 | Not related to the topic |
Galatsis, P. | Leucine-rich repeat kinase 2 inhibitors: a patent review (2014-2016) | 2017 | Published before 2022. |
Galecio-Castillo, M. | Direct to angiosuite strategy versus standard workflow triage for endovascular therapy: systematic review and meta-analysis | 2023 | Not related to the topic |
Garrido, C. | Heat shock proteins 27 and 70: anti-apoptotic proteins with tumorigenic properties | 2006 | Published before 2022. |
Gershlick, A. H. | The acute management of myocardial infarction | 2001 | Published before 2022. |
Gibler, W. B. | Continuum of Care for Acute Coronary Syndrome: Optimizing Treatment for ST-Elevation Myocardial Infarction and Non-St-Elevation Acute Coronary Syndrome | 2018 | Published before 2022. |
Gilotra, K. | Role of artificial intelligence and machine learning in the diagnosis of cerebrovascular disease | 2023 | Not related to the topic |
Giordano, P. | ChatGPT e il suo utilizzo nel supporto decisionale clinico: una scoping review | 2024 | Not in the language criteria |
Giuffrè, M. | Systematic review: The use of large language models as medical chatbots in digestive diseases | 2024 | Not related to the topic |
Giustino, G. | Safety and Efficacy of Bivalirudin in Patients with Diabetes Mellitus Undergoing Percutaneous Coronary Intervention: From the REPLACE-2, ACUITY and HORIZONS-AMI Trials | 2016 | Published before 2022. |
Table 1. Excluded studies | |||
Author | Name | Year | Reason of exclusion |
Goh, E. | ChatGPT Influence on Medical Decision-Making, Bias, and Equity: A Randomized Study of Clinicians Evaluating Clinical Vignettes | 2023 | Full text not written |
Goldstein, J. | Determinants for scalable adoption of autonomous AI in the detection of diabetic eye disease in diverse practice types: key best practices learned through collection of real-world data | 2023 | Not related to the topic |
Goto, K. | Predictors of outcomes in medically treated patients with acute coronary syndromes after angiographic triage: an Acute Catheterization And Urgent Intervention Triage Strategy (ACUITY) substudy | 2010 | Published before 2022. |
Goto, K. | Prognostic value of angiographic lesion complexity in patients with acute coronary syndromes undergoing percutaneous coronary intervention (from the acute catheterization and urgent intervention triage strategy trial) | 2014 | Published before 2022. |
Gradíssimo, A. | Molecular tests potentially improving HPV screening and genotyping for cervical cancer prevention | 2017 | Published before 2022. |
Guermazi, A. | How AI May Transform Musculoskeletal Imaging | 2024 | Not related to the topic |
Gulati, S. | Artificial intelligence in luminal endoscopy | 2020 | Published before 2022. |
Gunasekera, K. S. | Development of treatment-decision algorithms for children evaluated for pulmonary tuberculosis: an individual participant data meta-analysis | 2023 | Not related to the topic |
Gunzer, F. | Reproducibility of artificial intelligence models in computed tomography of the head: a quantitative analysis | 2022 | Not related to the topic |
Gurgitano, M. | Interventional Radiology ex-machina: impact of Artificial Intelligence on practice | 2021 | Published before 2022. |
Gutierrez, G. | Examining the role of AI technology in online mental healthcare: opportunities, challenges, and implications, a mixed-methods review | 2024 | Not related to the topic |
Haase, L. | Horse Diagnosis and Triage Accuracy of GPT-4o | 2024 | Not related to the topic |
Haider, S. P. | Admission computed tomography radiomic signatures outperform hematoma volume in predicting baseline clinical severity and functional outcome in the ATACH-2 trial intracerebral hemorrhage population | 2021 | Published before 2022. |
Halaseh, F. F. | ChatGPT’s Role in Improving Education Among Patients Seeking Emergency Medical Treatment | 2024 | Not related to the topic |
Hamilton, A. | Artificial Intelligence and Healthcare Simulation: The Shifting Landscape of Medical Education | 2024 | Not related to the topic |
Hamilton, A. J. | Machine learning and artificial intelligence: applications in healthcare epidemiology | 2021 | Published before 2022. |
Hamza, I. | Artificial Intelligence Echocardiography in Resource-Limited Regions: Applications and Challenges | 2024 | Not related to the topic |
Haq, M. | Revolutionizing Acute Stroke Care: A Review of Food and Drug Administration- Approved Software as Medical Devices for Stroke Triage | 2024 | Not related to the topic |
Hayat, J. | The Utility and Limitations of Artificial Intelligence-Powered Chatbots in Healthcare | 2024 | Not related to the topic |
Hickey, M. D. | Effect of a patient-centered hypertension delivery strategy on all-cause mortality: Secondary analysis of SEARCH, a community-randomized trial in rural Kenya and Uganda | 2021 | Published before 2022. |
Hickman, S. E. | Adoption of artificial intelligence in breast imaging: evaluation, ethical constraints and limitations | 2021 | Published before 2022. |
Hickman, S. E. | Machine Learning for Workflow Applications in Screening Mammography: Systematic Review and Meta-Analysis | 2022 | Published before 2022. |
Hirtsiefer, C. | Capabilities of ChatGPT-3.5 as a Urological Triage System | 2024 | Full text not written |
Hogarty D. T. | Current state and future prospects of artificial intelligence in ophthalmology: a review. | 2019 | Published before 2022. |
Hsieh, C. | Using Machine Learning to Predict Response to Image-guided Therapies for Hepatocellular Carcinoma | 2023 | Not related to the topic |
Hsueh, J. | Applications of Artificial Intelligence in Helicopter Emergency Medical Services: A Scoping Review | 2024 | Not related to the topic |
Huang, A. E. | Artificial Intelligence and Pediatric Otolaryngology | 2024 | Not related to the topic |
Table 1. Excluded studies | |||
Author | Name | Year | Reason of exclusion |
Hunter, O. F. | Science fiction or clinical reality: a review of the applications of artificial intelligence along the continuum of trauma care | 2023 | Not related to the topic |
Ingielewicz, A. | Drinking from the Holy Grail-Does a Perfect Triage System Exist? And Where to Look for It? | 2024 | Not related to the topic |
İsmail Mendi, B. | Artificial Intelligence in the Non-Invasive Detection of Melanoma | 2024 | Not related to the topic |
Ittarat, M. | Personalized Care in Eye Health: Exploring Opportunities, Challenges, and the Road Ahead for Chatbots | 2023 | Not related to the topic |
Ittarat, M. | Personalized Care in Eye Health: Exploring Opportunities, Challenges, and the Road Ahead for Chatbots | 2023 | Duplicate |
Iyer, R. | Detection of Suicide Risk Using Vocal Characteristics: Systematic Review | 2022 | Not related to the topic |
Jennings, L. K. | Antiplatelet and anticoagulant agents: key differences in mechanisms of action, clinical application, and therapeutic benefit in patients with non-ST- segment-elevation acute coronary syndromes | 2008 | Published before 2022. |
Joudar, S. S. | Triage and priority-based healthcare diagnosis using artificial intelligence for autism spectrum disorder and gene contribution: A systematic review | 2022 | Published before 2022. |
Jovin, T. G. | Thrombectomy for anterior circulation stroke beyond 6 h from time last known well (AURORA): a systematic review and individual patient data meta-analysis | 2022 | Published before 2022. |
Kachman, M. M. | How artificial intelligence could transform emergency care | 2024 | Full text not written |
Kalisz, K. R. | Immune Checkpoint Inhibitor Therapy-related Pneumonitis: Patterns and Management | 2019 | Published before 2022. |
Kaluski, E. | Glycoprotein IIb/IIIa inhibitors: questioning indications and treatment algorithms | 2007 | Not related to the topic |
Kang, C. | Artificial intelligence for diagnosing exudative age-related macular degeneration | 2024 | Not related to the topic |
Khalsa, R. K. | Artificial intelligence and cardiac surgery during COVID-19 era | 2021 | Published before 2022. |
Kiburg, K. V. | Telemedicine and delivery of ophthalmic care in rural and remote communities: Drawing from Australian experience | 2022 | Not related to the topic |
Kim, K. H. | [Applications of Artificial Intelligence in Mammography from a Development and Validation Perspective] | 2021 | Published before 2022. |
Kim, Y. | Applications of artificial intelligence in the thorax: a narrative review focusing on thoracic radiology | 2021 | Not related to the topic |
Knebel, D. | Assessment of ChatGPT in the Prehospital Management of Ophthalmological Emergencies - An Analysis of 10 Fictional Case Vignettes | 2024 | Not related to the topic |
Koren, J., 3rd | The Right Tool for the Job: An Overview of Hsp90 Inhibitors | 2020 | Published before 2022. |
Koren, J., 3rd | The Right Tool for the Job: An Overview of Hsp90 Inhibitors | 2020 | Duplicate |
Korobelnik, J. F. | Guidance for anti-VEGF intravitreal injections during the COVID-19 pandemic | 2020 | Published before 2022. |
Krause, A. J. | An update on current treatment strategies for laryngopharyngeal reflux symptoms | 2022 | Not related to the topic |
Krothapalli, N. | Mobile stroke units: Beyond thrombolysis | 2024 | Not related to the topic |
Krusche, M. | Diagnostic accuracy of a large language model in rheumatology: comparison of physician and ChatGPT-4 | 2024 | Not related to the topic |
Kumar, D. | Comparison of Bivalirudin versus Bivalirudin plus glycoprotein IIb/IIIa inhibitor versus heparin plus glycoprotein IIb/IIIa inhibitor in patients with acute coronary syndromes having percutaneous intervention for narrowed saphenous vein aorto-coronary... | 2010 | Published before 2022. |
Kumar, H. | A clinical perspective on the expanding role of artificial intelligence in age- related macular degeneration | 2022 | Not related to the topic |
Kunze, K. N. | Editorial Commentary: The Scope of Medical Research Concerning ChatGPT Remains Limited by Lack of Originality | 2024 | Not related to the topic |
Kunze, K. N. | The Large Language Model ChatGPT-4 Exhibits Excellent Triage Capabilities and Diagnostic Performance for Patients Presenting With Various Causes of Knee Pain | 2024 | Full text not written |
Table 1. Excluded studies | |||
Author | Name | Year | Reason of exclusion |
Kusunose, K. | Radiomics in Echocardiography: Deep Learning and Echocardiographic Analysis | 2020 | Published before 2022. |
Laino, M. E. | Prognostic findings for ICU admission in patients with COVID-19 pneumonia: baseline and follow-up chest CT and the added value of artificial intelligence | 2022 | Not related to the topic |
Lalla, R. | Assessing the validity of the Triage Risk Screening Tool in a third world setting. | 2018 | Published before 2022. |
Lamb, L. R. | Artificial Intelligence (AI) for Screening Mammography, From the AJR Special Series on AI Applications | 2022 | Not related to the topic |
Lång, K. | Artificial intelligence-supported screen reading versus standard double reading in the Mammography Screening with Artificial Intelligence trial (MASAI): a clinical safety analysis of a randomised, controlled, non-inferiority, single- blinded, screening... | 2023 | Not related to the topic |
Lang, M. | Artificial Intelligence in Cardiovascular Imaging: “Unexplainable” Legal and Ethical Challenges? | 2022 | Not related to the topic |
Le, K. D. R. | Applications of natural language processing tools in the surgical journey | 2024 | Duplicate |
Le, K. D. R. | Applications of natural language processing tools in the surgical journey | 2024 | Not related to the topic |
Leung, E. H. | Ocular and Systemic Complications of COVID-19: Impact on Patients and Healthcare | 2022 | Not related to the topic |
Li, Y. | Emergency trauma care during the outbreak of corona virus disease 2019 (COVID-19) in China | 2020 | Published before 2022. |
Lincoff, A. M. | Influence of timing of clopidogrel treatment on the efficacy and safety of bivalirudin in patients with non-ST-segment elevation acute coronary syndromes undergoing percutaneous coronary intervention: an analysis of the ACUITY (Acute Catheterization and... | 2008 | Published before 2022. |
Liu, Z. | Toward Clinical Implementation of Next-Generation Sequencing-Based Genetic Testing in Rare Diseases: Where Are We? | 2019 | Published before 2022. |
Lo Gullo, R. | AI Applications to Breast MRI: Today and Tomorrow | 2024 | Not related to the topic |
Lodise, N. M. | Hypoactive sexual desire disorder in women: treatment options beyond testosterone and approaches to communicating with patients on sexual health | 2013 | Published before 2022. |
Loggers, S. A. I. | Definition of hemodynamic stability in blunt trauma patients: a systematic review and assessment amongst Dutch trauma team members | 2017 | Published before 2022. |
Lopes, R. D. | Advanced age, antithrombotic strategy, and bleeding in non-ST-segment elevation acute coronary syndromes: results from the ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy) trial | 2009 | Published before 2022. |
Luo, W. | The Influence of the Novel Computer-Aided Triage System Based on Artificial Intelligence on Endovascular Therapy in Patients with Large Vascular Occlusions: A Meta-Analysis | 2024 | Not related to the topic |
Lyons, R. J. | Artificial intelligence chatbot performance in triage of ophthalmic conditions | 2024 | Not related to the topic |
Malycha, J. | Artificial intelligence and clinical deterioration | 2022 | Not related to the topic |
Marko, M. | Management and outcome of patients with acute ischemic stroke and tandem carotid occlusion in the ESCAPE-NA1 trial | 2022 | Not related to the topic |
Marques, M. | The Medicine Revolution Through Artificial Intelligence: Ethical Challenges of Machine Learning Algorithms in Decision-Making | 2024 | Not related to the topic |
Mehran, R. | Impact of chronic kidney disease on early (30-day) and late (1-year) outcomes of patients with acute coronary syndromes treated with alternative antithrombotic treatment strategies: an ACUITY (Acute Catheterization and Urgent Intervention Triage... | 2009 | Published before 2022. |
Meral, G. | Comparative analysis of ChatGPT, Gemini and emergency medicine specialist in ESI triage assessment | 2024 | Full text not written |
Meral, G. | Comparative analysis of ChatGPT, Gemini and emergency medicine specialist in ESI triage assessment | 2024 | Duplicate |
Miller, B. S. | Emergency management of adrenal insufficiency in children: advocating for treatment options in outpatient and field settings | 2020 | Not related to the topic |
Milne-Ives, M. | The Effectiveness of Artificial Intelligence Conversational Agents in Health Care: Systematic Review | 2020 | Not related to the topic |
Table 1. Excluded studies | |||
Author | Name | Year | Reason of exclusion |
Miyata, Y. | Molecular chaperones and regulation of tau quality control: strategies for drug discovery in tauopathies | 2011 | Published before 2022. |
Momenaei, B. | ChatGPT enters the room: what it means for patient counselling, physician education, academics, and disease management | 2024 | Not related to the topic |
Momenaei, B. | ChatGPT enters the room: what it means for patient counselling, physician education, academics, and disease management | 2024 | Duplicate |
Monga, M. | Artificial Intelligence in Endourology: Maximizing the Promise Through Consideration of the Principles of Diffusion of Innovation Theory | 2024 | Not related to the topic |
Moparthi, K. P. | Acute Care Surgery: Navigating Recent Developments, Protocols, and Challenges in the Comprehensive Management of Surgical Emergencies | 2024 | Not related to the topic |
Morgan, M. B. | Applications of Artificial Intelligence in Breast Imaging | 2021 | Published before 2022. |
Moscicki, A. B. | Screening for Anal Cancer in Women | 2015 | Published before 2022. |
Mukherjee, D. | Pharmacotherapy of acute coronary syndrome: the ACUITY trial | 2009 | Published before 2022. |
Mungmunpuntipantip, R. | ChatGPT in Trauma Triage | 2024 | Full text not written |
Murray, N. M. | Artificial intelligence to diagnose ischemic stroke and identify large vessel occlusions: a systematic review | 2020 | Published before 2022. |
Nardell, E. | Turning off the spigot: reducing drug-resistant tuberculosis transmission in resource-limited settings | 2010 | Published before 2022. |
Nathavitharana, R. R. | Reimagining the status quo: How close are we to rapid sputum-free tuberculosis diagnostics for all? | 2022 | Not related to the topic |
Nazif, T. M. | Comparative effectiveness of upstream glycoprotein IIb/IIIa inhibitors in patients with moderate- and high-risk acute coronary syndromes: an Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) substudy | 2014 | Published before 2022. |
Nazir, T. | Artificial intelligence assisted acute patient journey | 2022 | Not related to the topic |
Ndrepepa, G. | Bivalirudin versus heparin plus a glycoprotein IIb/IIIa inhibitor in patients with non-ST-segment elevation myocardial infarction undergoing percutaneous coronary intervention after clopidogrel pretreatment: pooled analysis from the ACUITY and... | 2012 | Published before 2022. |
Nikolsky, E. | Gastrointestinal bleeding in patients with acute coronary syndromes: incidence, predictors, and clinical implications: analysis from the ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy) trial | 2009 | Published before 2022. |
Nikolsky, E. | Outcomes of patients with prior coronary artery bypass grafting and acute coronary syndromes: analysis from the ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy) trial | 2012 | Published before 2022. |
O Hern, K. | ChatGPT underperforms in triaging appropriate use of Mohs surgery for cutaneous neoplasms | 2023 | Not related to the topic |
Pai, M. | Tuberculosis diagnostics in 2015: landscape, priorities, needs, and prospects | 2015 | Published before 2022. |
Park, J. | Validation of a Natural Language Machine Learning Model for Safety Literature Surveillance | 2024 | Not related to the topic |
Paslı, S. | Assessing the precision of artificial intelligence in ED triage decisions: Insights from a study with ChatGPT | 2024 | Full text not written |
Paslı, S. | Response to: Methodological issues on precision and prediction value of ChatGPT in emergency department triage decisions | 2024 | Full text not written |
Patel, A. V. | Increasing HIV testing engagement through provision of home HIV self- testing kits for patients who decline testing in the emergency department: a pilot randomisation study | 2019 | Published before 2022. |
Peng, H. T. | Artificial intelligence and machine learning for hemorrhagic trauma care | 2023 | Not related to the topic |
Peng, Z. | Development and evaluation of multimodal AI for diagnosis and triage of ophthalmic diseases using ChatGPT and anterior segment images: protocol for a two-stage cross-sectional study | 2023 | Not related to the topic |
Pépin, J. L. | New management pathways for follow-up of CPAP-treated sleep apnoea patients including digital medicine and multimodal telemonitoring | 2024 | Not related to the topic |
Table 1. Excluded studies | |||
Author | Name | Year | Reason of exclusion |
Pham, J. H. | Large language model triaging of simulated nephrology patient inbox messages | 2024 | Not related to the topic |
Pinto, D. S. | Economic evaluation of bivalirudin with or without glycoprotein IIb/IIIa inhibition versus heparin with routine glycoprotein IIb/IIIa inhibition for early invasive management of acute coronary syndromes | 2008 | Published before 2022. |
Posadas, E. M. | Targeting angiogenesis in renal cell carcinoma | 2013 | Published before 2022. |
Potnis, K. C. | Artificial Intelligence in Breast Cancer Screening: Evaluation of FDA Device Regulation and Future Recommendations | 2022 | Not related to the topic |
Preiksaitis, C. | The Role of Large Language Models in Transforming Emergency Medicine: Scoping Review | 2024 | Duplicate, included in the review |
Pressman, S. M. | Clinical and Surgical Applications of Large Language Models: A Systematic Review | 2024 | Not related to the topic |
Pressman, S. M. | Clinical and Surgical Applications of Large Language Models: A Systematic Review | 2024 | Duplicate |
Ramkumar, P. N. | Editorial Commentary: Large Language Models Like ChatGPT Show Promise, but Clinical Use of Artificial Intelligence Requires Physician Partnership | 2024 | Not related to the topic |
Razzaki, S. | A comparative study of artificial intelligence and human doctors for the purpose of triage and diagnosis | 2018 | Published before 2022. |
Rengers, T. A. | Academic Surgery in the Era of Large Language Models: A Review | 2024 | Not related to the topic |
Rengers, T. A. | Academic Surgery in the Era of Large Language Models: A Review | 2024 | Duplicate |
Ricklin, D. | Manipulating the mediator: modulation of the alternative complement pathway C3 convertase in health, disease and therapy | 2012 | Published before 2022. |
Rietjens, S. J. | Pharmacokinetics and pharmacodynamics of 3,4-methylenedioxymethamphetamine (MDMA): interindividual differences due to polymorphisms and drug-drug interactions | 2012 | Published before 2022. |
Sabaner, M. C. | Opportunities and Challenges of Chatbots in Ophthalmology: A Narrative Review | 2024 | Not related to the topic |
Sabaner, M. C. | Opportunities and Challenges of Chatbots in Ophthalmology: A Narrative Review | 2024 | Duplicate |
Sabour, A. | Methodological issues on precision and prediction value of ChatGPT in emergency department triage decisions | 2024 | Full text not written |
Saenger, J. A. | Delayed diagnosis of a transient ischemic attack caused by ChatGPT | 2024 | Not related to the topic |
Salim, M. | AI-based selection of individuals for supplemental MRI in population-based breast cancer screening: the randomized ScreenTrustMRI trial | 2024 | Not related to the topic |
Sammer, M. B. K. | Ensuring Adequate Development and Appropriate Use of Artificial Intelligence in Pediatric Medical Imaging | 2022 | Not related to the topic |
Santoro, E. | [Information technology and digital health to support health in the time of CoViD-19.] | 2020 | Published before 2022. |
Satyamitra, M. | Challenges and Strategies in the Development of Radiation Biodosimetry Tests for Patient Management | 2021 | Published before 2022. |
Shapiro, J. | New Diagnostic Tools for Pulmonary Embolism Detection | 2024 | Not related to the topic |
Shekhar, A. C. | Use of a large language model (LLM) for ambulance dispatch and triage | 2024 | Full text not written |
Shlobin, N. A. | Artificial Intelligence for Large-Vessel Occlusion Stroke: A Systematic Review | 2022 | Not related to the topic |
Singh, N. | Infarcts in a New Territory: Insights From the ESCAPE-NA1 Trial | 2023 | Not related to the topic |
Smith, K. P. | Image analysis and artificial intelligence in infectious disease diagnostics | 2020 | Published before 2022. |
Snow, K. D. | Trends in emergency department visits for bronchiolitis, 1993-2019 | 2024 | Not related to the topic |
Soun, J. E. | Artificial Intelligence and Acute Stroke Imaging | 2021 | Published before 2022. |
Stegeman, I. | Routine laboratory testing to determine if a patient has COVID-19 | 2020 | Published before 2022. |
Stewart, M. | Targeting chaperone modifications: Innovative approaches to cancer treatment | 2024 | Not related to the topic |
Stone, G. W. | Acute Catheterization and Urgent Intervention Triage strategy (ACUITY) trial: study design and rationale | 2004 | Published before 2022. |
Table 1. Excluded studies | |||
Author | Name | Year | Reason of exclusion |
Stone, G. W. | Bivalirudin in patients with acute coronary syndromes undergoing percutaneous coronary intervention: a subgroup analysis from the Acute Catheterization and Urgent Intervention Triage strategy (ACUITY) trial | 2007 | Published before 2022. |
Swartz, H. M. | Overview of the principles and practice of biodosimetry | 2014 | Published before 2022. |
Swartz, H. M. | Scientific and Logistical Considerations When Screening for Radiation Risks by Using Biodosimetry Based on Biological Effects of Radiation Rather than Dose: The Need for Prior Measurements of Homogeneity and Distribution of Dose | 2020 | Published before 2022. |
Syrowatka, A. | Leveraging artificial intelligence for pandemic preparedness and response: a scoping review to identify key use cases | 2021 | Published before 2022. |
Tarone, G. | Keep your heart in shape: molecular chaperone networks for treating heart disease | 2014 | Published before 2022. |
Taylor, C. R. | Artificial Intelligence Applications in Breast Imaging: Current Status and Future Directions | 2023 | Not related to the topic |
Thomas, L. B. | Artificial Intelligence: Review of Current and Future Applications in Medicine | 2021 | Published before 2022. |
Thomassin-Naggara, I. | [French breast cancer screening: What’s the place of artificial intelligence?] | 2022 | Not related to the topic |
Ting, D. S. W. | Artificial intelligence and deep learning in ophthalmology | 2019 | Not related to the topic |
Tortum, F. | Exploring the potential of artificial intelligence models for triage in the emergency department | 2024 | Full text not written |
Toy, J. | Use of artificial intelligence to support prehospital traumatic injury care: A scoping review | 2024 | Not related to the topic |
Trzeciak, A. | Biomarkers and Associated Immune Mechanisms for Early Detection and Therapeutic Management of Sepsis | 2020 | Not related to the topic |
Tse, E. | The diagnosis and management of NK/T-cell lymphomas | 2017 | Published before 2022. |
Vaduganathan, M. | Evaluation of Ischemic and Bleeding Risks Associated With 2 Parenteral Antiplatelet Strategies Comparing Cangrelor with Glycoprotein IIb/IIIa Inhibitors: An Exploratory Analysis from the CHAMPION Trials | 2017 | Published before 2022. |
Vandevenne, M. M. | Artificial intelligence for detecting keratoconus | 2023 | Not related to the topic |
Vedantham, S. | Artificial Intelligence in Breast X-Ray Imaging | 2023 | Not related to the topic |
Verheugt, F. W. | Incidence, prognostic impact, and influence of antithrombotic therapy on access and nonaccess site bleeding in percutaneous coronary intervention | 2011 | Published before 2022. |
Vinay, R. | Ethics of ICU triage during COVID-19 | 2021 | Published before 2022. |
Vinny, P. W. | Critical Appraisal of a Machine Learning Paper: A Guide for the Neurologist | 2021 | Published before 2022. |
Walsh, L. | A Systematic Review of Current Teleophthalmology Services in New Zealand Compared to the Four Comparable Countries of the United Kingdom, Australia, United States of America (USA) and Canada | 2021 | Published before 2022. |
Wang, C. | Diagnostic Test Accuracy of Deep Learning Prediction Models on COVID-19 Severity: Systematic Review and Meta-Analysis | 2023 | Not related to the topic |
Wang, X. | ChatGPT: promise and challenges for deployment in low- and middle-income countries | 2023 | Not related to the topic |
Ward, M. | A Quantitative Assessment of ChatGPT as a Neurosurgical Triaging Tool | 2024 | Not related to the topic |
Ward, M. | Analysis of ChatGPT in the Triage of Common Spinal Complaints | 2024 | Not related to the topic |
Warsinske, H. | Host-response-based gene signatures for tuberculosis diagnosis: A systematic comparison of 16 signatures | 2019 | Published before 2022. |
Weisberg, E. M. | The first use of artificial intelligence (AI) in the ER: triage not diagnosis | 2020 | Published before 2022. |
White, H. D. | Safety and efficacy of bivalirudin with and without glycoprotein IIb/IIIa inhibitors in patients with acute coronary syndromes undergoing percutaneous coronary intervention 1-year results from the ACUITY (Acute Catheterization and Urgent Intervention... | 2008 | Published before 2022. |
Wójcik, S. | Beyond ChatGPT: What does GPT-4 add to healthcare? The dawn of a new era | 2023 | Not related to the topic |
Woodfin, M. W. | ChatGPT Effectively Triages Real-World Neoplasms Using Mohs Appropriate Use Criteria | 2024 | Not related to the topic |
Table 1. Excluded studies | |||
Author | Name | Year | Reason of exclusion |
Xavier, D. | Artificial intelligence for triaging of breast cancer screening mammograms and workload reduction: A meta-analysis of a deep learning software | 2024 | Not related to the topic |
Xie, Y. | Reviewing Hit Discovery Literature for Difficult Targets: Glutathione Transferase Omega-1 as an Example | 2018 | Published before 2022. |
Xie, Y. | Reviewing Hit Discovery Literature for Difficult Targets: Glutathione Transferase Omega-1 as an Example | 2018 | Duplicate |
Xu, R. | Generative artificial intelligence in healthcare from the perspective of digital media: Applications, opportunities and challenges | 2024 | Not related to the topic |
Yamasaki, S. | Reprogramming mRNA translation during stress | 2008 | Published before 2022. |
Yang, Z. | Understanding natural language: Potential application of large language models to ophthalmology | 2024 | Not related to the topic |
Yi, X. | Action plan for hit identification (APHID): KAT6A as a case study | 2020 | Published before 2022. |
Yuba, M. | Systematic analysis of the test design and performance of AI/ML-based medical devices approved for triage/detection/diagnosis in the USA and Japan | 2022 | Not related to the topic |
Zaboli, A. | Human intelligence versus Chat-GPT: who performs better in correctly classifying patients in triage? | 2024 | Full text not written |
Zandi, R. | Exploring Diagnostic Precision and Triage Proficiency: A Comparative Study of GPT-4 and Bard in Addressing Common Ophthalmic Complaints | 2024 | Not related to the topic |
Zarella, M. D. | Artificial intelligence and digital pathology: clinical promise and deployment considerations | 2023 | Not related to the topic |
Zhang, Z. | Associations of immunological features with COVID-19 severity: a systematic review and meta-analysis | 2021 | Published before 2022. |
Susanty, S. | Questionnaire-free machine-learning method to predict depressive symptoms among community-dwelling older adults | 2022 | Not related to the topic |
Al-Zaiti, SS. | Machine learning for ECG diagnosis and risk stratification of occlusion myocardial infarction | 2023 | Not related to the topic |
Charan, GS | Impact of Analytics Applying Artificial Intelligence and Machine Learning on Enhancing Intensive Care Unit: A Narrative Review | 2023 | Not related to the topic |
van Maurik, IS | Targeted Development and Validation of Clinical Prediction Models in Secondary Care Settings: Opportunities and Challenges for Electronic Health Record Data | 2024 | Not related to the topic |
Switzer, DF | Ethics Crisis Standards of Care Simulation | 2024 | Not related to the topic |
Zaboli, A. | Human intelligence versus Chat-GPT: who performs better in correctly classifying patients in triage? | 2024 | Full text not written |
Yang, J | Development and evaluation of an artificial intelligence-based workflow for the prioritization of patient portal messages | 2024 | Not related to the topic |
Duncan, SF | Radiograph accelerated detection and identification of cancer in the lung (RADICAL): a mixed methods study to assess the clinical effectiveness and acceptability of Qure.ai artificial intelligence software to prioritise chest X-ray (CXR) interpretation | 2024 | Not related to the topic |
Chenard, SW | ChatGPT provides safe responses to post-operative concerns following total joint arthroplasty | 2024 | Not related to the topic |
Arends, BKO | Barriers, facilitators and strategies for the implementation of artificial intelligence-based electrocardiogram interpretation: A mixed- methods study | 2025 | Not related to the topic |
Zaboli, A | Chat-GPT in triage: Still far from surpassing human expertise - An observational study | 2025 | Full text not written |
Mani, Z | AI frontiers in emergency care: the next evolution of nursing interventions. | 2024 | Duplicate |