other Emergency Medicine Emergency Medicine Systematic Review <span style="font-size:13pt"><span style="text-justify:kashida"><span style="text-kashida:0%"><span style="text-autospace:none"><span style="font-family:&quot;Times New Roman&quot;,serif"><span style="font-style:italic"><b><span style="font-size:9.0pt"><span style="font-style:normal">Background: </span></span></b><span style="font-size:9.0pt"><span style="font-style:normal">Emergency departments face challenges of overcrowding and human error. This systematic review evaluates the effectiveness of artificial intelligence (AI) in improving triage and diagnostic accuracy in emergency settings.</span></span></span></span></span></span></span></span><br> <span style="font-size:13pt"><span style="text-justify:kashida"><span style="text-kashida:0%"><span style="text-autospace:none"><span style="font-family:&quot;Times New Roman&quot;,serif"><span style="font-style:italic"><span style="font-size:9.0pt"><span style="font-style:normal">&nbsp;&nbsp; <b>Methods:</b> This systematic review was conducted by searching PubMed, Scopus, Web of Science, and Google Scholar until September 2025. Inclusion criteria comprised original studies reporting at least one quantitative performance metric. Study quality was assessed using NOS and QUADAS-2. Data were synthesized narratively.</span></span></span></span></span></span></span></span><br> <span style="font-size:13pt"><span style="text-justify:kashida"><span style="text-kashida:0%"><span style="text-autospace:none"><span style="font-family:&quot;Times New Roman&quot;,serif"><span style="font-style:italic"><span style="font-size:9.0pt"><span style="font-style:normal">&nbsp;&nbsp; <b>Results:</b> AI models outperformed traditional triage systems (AUC often >0.90). In clinical interpretation of AUC, its limitations in imbalanced datasets should be considered, and metrics such as PPV, NPV, and AUPRC should also be reported. For instance, despite an acceptable AUC, ChatGPT 4.0 led to significant over-triage due to low specificity (18.2%). However, risks of algorithmic bias and challenges of integrating AI with real-world clinical workflows remain major implementation barriers. User resistance and the need for continuous ethical oversight were also identified.</span></span></span></span></span></span></span></span><br> <span style="font-size:13pt"><span style="text-justify:kashida"><span style="text-kashida:0%"><span style="text-autospace:none"><span style="font-family:&quot;Times New Roman&quot;,serif"><span style="font-style:italic"><span style="font-size:9.0pt"><span style="font-style:normal">&nbsp;&nbsp; <b>Conclusion:</b> AI holds potential to improve emergency triage, but current evidence is largely based on retrospective single-center studies. Clinical implementation should be gradual, starting in high-risk domains (e.g., sepsis screening), and supported by continuous ethical oversight.</span></span></span></span></span></span></span></span><br> &nbsp; Artificial Intelligence, Triage, Diagnosis, Emergency Department 498 504 http://mjiri.iums.ac.ir/browse.php?a_code=A-10-1-1127&slc_lang=other&sid=1 Mehdi Shakoori Shakoorimah9811@Gmail.com 200319475328460098291 0009-0006-4597-4353 No Trauma and Injury Research Center, Iran University of Medical Sciences, Tehran, Iran Fatemeh Esmaili Esmailifateme72@gmail.com 200319475328460098292 0009-0005-2261-4518 No Radiation Injuries Research Center, Systems Biology and Poisonings Institute, Nursing Care research center, Clinical Sciences institute, Nursing Faculty, Baqiyatallah University of Medical Sciences, Tehran, Iran Ezzat Abbariki Abbarikiezat@Gmail.com 200319475328460098293 0009-0000-4761-1528 No Firoozgar Clinical Research and Development Center, Iran University of Medical Sciences, Tehran, Iran Esmail Heidaranlu Heidaranlu@bmsu.ac.ir 200319475328460098294 200319475328460098294 Yes Nursing Care Research Center, Clinical Sciences Institute, Nursing Faculty, Baqiyatallah University of Medical Sciences, Tehran, Iran