Introduction: Artificial intelligence (AI), particularly ChatGPT developed by OpenAI, has shown the potential to improve diagnostic accuracy and efficiency in emergency department (ED) triage. This study aims to evaluate the diagnostic performance and safety of ChatGPT in prioritizing patients based on urgency in ED settings.
Methods: A systematic review and meta-analysis were conducted following PRISMA guidelines. Comprehensive literature searches were performed in Scopus, Web of Science, PubMed, and Embase. Studies evaluating ChatGPT's diagnostic performance in ED triage were included. Quality assessment was conducted using the QUADAS-2 tool. Pooled accuracy estimates were calculated using a random-effects model, and heterogeneity was assessed with the I² statistic.
Results: Fourteen studies with a total of 1,412 patients or scenarios were included. ChatGPT 4.0 demonstrated a pooled accuracy of 0.86 (95% CI: 0.64-0.98) with substantial heterogeneity (I² = 93%). ChatGPT 3.5 showed a pooled accuracy of 0.63 (95% CI: 0.43-0.81) with significant heterogeneity (I² = 84%). Funnel plots indicated potential publication bias, particularly for ChatGPT 3.5. Quality assessments revealed varying levels of risk of bias and applicability concerns.
Conclusion: ChatGPT, especially version 4.0, shows promise in improving ED triage accuracy. However, significant variability and potential biases highlight the need for further evaluation and enhancement.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11407534 | PMC |
| http://dx.doi.org/10.22037/aaem.v12i1.2384 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Dementia Care Research and Psychosocial Factors.
Alzheimers Dement
December 2024
National University, Muscat, Muscat, Oman.
Background: This study explores Alzheimer's prediction through brain MRI images, utilizing Convolutional Neural Networks (CNNs) and Lime interpretability. Based on an extensive ADNI MRI dataset, we demonstrate promising results in predicting Alzheimer's disease. Local Interpretable Model Agnostic Explanations (LIME) shed light on decision-making processes, enhancing transparency.
View Article and Find Full Text PDFEffective BCDNet-based breast cancer classification model using hybrid deep learning with VGG16-based optimal feature extraction.
BMC Med Imaging
January 2025
Department of Information Technology, Manipal Institute of Technology Bengaluru, Manipal Academy of Higher Education, Manipal, Karnataka, 576104, India.
Problem: Breast cancer is a leading cause of death among women, and early detection is crucial for improving survival rates. The manual breast cancer diagnosis utilizes more time and is subjective. Also, the previous CAD models mostly depend on manmade visual details that are complex to generalize across ultrasound images utilizing distinct techniques.
View Article and Find Full Text PDFCardiac dose comparison of abdominal and thoracic deep inspiration breath hold in adult female with left-sided breast cancer receiving radiation therapy: A systematic review and meta-analysis.
Radiography (Lond)
January 2025
School of Nursing, The Hong Kong Polytechnic University, Hong Kong. Electronic address:
Introduction: Deep inspiration breath hold technique has shown promise in reducing cardiac toxicity and improving patient outcomes. However, there is a lack of consensus regarding the implementation of abdominal breath hold technique and its impact on cardiac dose. This systematic review and meta-analysis aim to provide insights into the comparative effectiveness of abdominal and thoracic breath hold in mitigating cardiac toxicity during radiation therapy for left-sided breast cancer.
View Article and Find Full Text PDFGFRs in Chinese CKD: A systematic review.
Clin Chim Acta
January 2025
Department of Nuclear Medicine, Hebei Medical University Third Hospital, Shijiazhuang 050051, PR China. Electronic address:
Purpose: This was an evidence-based study to assess which creatinine-based equation was most useful for estimated glomerular filtration rate (eGFR) in Chinese adults with chronic kidney disease (CKD).
Methods: Multiple databases were searched to collect relevant studies on creatinine-based eGFR equations for Chinese adults with CKD in Chinese and English from January 2009 to January 2023, using "glomerular filtration rate", "GFR equations", "Chinese CKD", "chronic kidney disease", "equation development" and "equation validation". The quality of each study was assessed using the diagnostic test accuracy review by RevMan 5.
Performance of Convolutional Neural Network Models in Meningioma Segmentation in Magnetic Resonance Imaging: A Systematic Review and Meta-Analysis.
Neuroinformatics
January 2025
Institute of Biophotonics, National Yang Ming Chiao Tung University, 155, Sec. 2, Li-Nong St. Beitou Dist, Taipei, 112304, Taiwan.
Background: Meningioma, the most common primary brain tumor, presents significant challenges in MRI-based diagnosis and treatment planning due to its diverse manifestations. Convolutional Neural Networks (CNNs) have shown promise in improving the accuracy and efficiency of meningioma segmentation from MRI scans. This systematic review and meta-analysis assess the effectiveness of CNN models in segmenting meningioma using MRI.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!