• Home
  • Search
  • Artificial Intelligence for CT and MRI Protocoling: A Meta-Analysis of Traditional Machine Learning, BERT, and Large Language Models.
  • Cite Icon4
  • https://doi.org/10.2214/ajr.25.33759Copy DOI Icon

Artificial Intelligence for CT and MRI Protocoling: A Meta-Analysis of Traditional Machine Learning, BERT, and Large Language Models.

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

BACKGROUND. Examination protocoling is a resource-intensive task. Various artificial intelligence (AI) approaches have been investigated to automate this process. OBJECTIVE. The purpose of this study was to evaluate performance of traditional machine learning (ML) models, bidirectional encoder representations from transformers (BERT) models, and large language models (LLMs) for automated CT and MRI protocoling. EVIDENCE ACQUISITION. MEDLINE, Embase, Scopus, Web of Science, IEEE Xplore, and Google Scholar databases were searched through July 2025 for studies reporting the performance of an AI-based technique in assigning protocols for CT or MRI requisitions. Accuracy results were separately extracted for all models tested in each study and pooled using a random-effects meta-analysis. AI approaches were compared using Welch t tests. Common sources of error were qualitatively summarized. EVIDENCE SYNTHESIS. The final analysis included 23 studies, comprising 1,196,259 imaging requisitions. Requisition subspecialties included body imaging (n = 4), musculoskeletal imaging (n = 3), neuroradiology (n = 6), thoracic imaging (n = 1), and multiple subspecialties (n = 9). Sixteen studies evaluated traditional ML models, eight evaluated BERT models, and five evaluated LLMs. Task-specific model fine-tuning was performed in three studies for traditional ML models, all studies for BERT models, and one study for LLMs. The overall pooled protocoling accuracy was 85% (95% CI, 83-87%). The pooled accuracy was 83% (95% CI, 80-85%) for traditional ML models, 87% (95% CI, 85-89%) for BERT models, and 86% (95% CI, 83-89%) for LLMs; these pooled accuracies were not significantly different between any pairwise combination of the three AI approaches (all p > .05). Among 30 distinct models (14 traditional ML models, nine BERT models, seven LLMs), the top-10 performing models comprised two traditional ML models, six BERT models (including the top performing model [BioBERT, a biomedical-domain BERT; accuracy, 93%]), and two LLMs. Common sources of error included ambiguous requisition text, data imbalance yielding incorrect protocol assignments for low-volume protocols, the presence of multiple clinically reasonable protocols for given requisitions, and difficulty handling requisitions containing terms strongly associated with disparate protocols. CONCLUSION. The top-performing AI models for automated CT and MRI protocoling included predominantly fine-tuned BERT models. CLINICAL IMPACT. AI tools show strong potential to help streamline radiologist workflows, possibly through hybrid AI-radiologist approaches. Fine-tuned LLMs warrant further exploration. TRIAL REGISTRATION. PROSPERO identifier CRD420251088671.

Similar Papers
  • PDF
  • Research Article
  • Citations49

Traditional Machine Learning Models and Bidirectional Encoder Representations From Transformer (BERT)–Based Automatic Classification of Tweets About Eating Disorders: Algorithm Development and Validation Study

  • Feb 24, 2022
  • JMIR Medical Informatics
  • José Alberto Benítez-Andrades +4
  • Research Article

AI-powered surveillance of bronchiolitis in the Nirsevimab era: comparative performance of machine learning, deep learning, and large language models on free-text ED records.

  • Mar 10, 2026
  • BMC emergency medicine
  • Marianna Costa +5
  • Research Article

Empirical Analysis of the State-of-the-Art Models for Handling Polarity Shifts Due to Implicit Negation in Mobile Phone Reviews

  • Jan 01, 2025
  • International Journal of Computing Sciences Research
  • Millicent Murithi +2
  • PDF
  • Research Article
  • Citations37

Classifying the lifestyle status for Alzheimer’s disease from clinical notes using deep learning with weak supervision

  • Jul 01, 2022
  • BMC medical informatics and decision making
  • Zitao Shen +6
  • Conference Article
  • Citations3

Engineering Document Summarization Using Sentence Representations Generated by Bidirectional Language Model

  • Aug 17, 2021
  • Yunjian Qiu +1
  • Research Article
  • Citations82

Evaluating the accuracy of a state-of-the-art large language model for prediction of admissions from the emergency room.

  • May 21, 2024
  • Journal of the American Medical Informatics Association : JAMIA
  • Benjamin S Glicksberg +11
  • Research Article

Novel Convolution Neural Network for Enhanced Network Attack Pattern Recognition

  • Sep 19, 2025
  • University of Bisha Journal for Basic and Applied Sciences
  • Salahaldeen Duraibi
  • PDF
  • Research Article
  • Citations14

Comparing human text classification performance and explainability with large language and machine learning models using eye-tracking

  • Jun 21, 2024
  • Scientific Reports
  • Jeevithashree Divya Venkatesh +2
  • Research Article

Spam Detection Based on BERT Model: A Descriptive Study

  • Jan 01, 2023
  • Theory and practice of science and technology
  • Hailin Xu
  • Research Article
  • Citations1

Enhancing Stock Price Prediction through Sentiment Analysis: A Comparative Study of Machine Learning and Deep Learning Models Using Financial News Data

  • Feb 03, 2025
  • Frontline Marketing, Management and Economics Journal
  • Paresh Chandra Nath +6
  • Research Article

GenAI Based Identification and Analysis of Security Risks and Vulnerabilities in 5G Network Traffic

  • Oct 31, 2025
  • World Journal of Advanced Engineering Technology and Sciences
  • Mukesh Kumar Bansal +2
  • Research Article
  • Citations3

(Retracted) News image text classification algorithm with bidirectional encoder representations from transformers model

  • Sep 13, 2022
  • Journal of Electronic Imaging
  • Zhan Shi +3
  • Research Article
  • Citations3

Evaluating the Effectiveness of Large Language Models (LLMs) Versus Machine Learning (ML) in Identifying and Detecting Phishing Email Attempts

  • Sep 25, 2025
  • Algorithms
  • Saed Tarapiah +5
  • Research Article
  • Citations2

CACER: Clinical concept Annotations for Cancer Events and Relations.

  • Sep 03, 2024
  • Journal of the American Medical Informatics Association : JAMIA
  • Yujuan Velvin Fu +7
  • Conference Article
  • Citations1

Comparative Analysis of Transformer Models in Disaster Tweet Classification for Public Safety

  • Sep 23, 2025
  • Sharif Noor Zisad +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.