Systematic Review of Recent ChatGPT updates on Reporting of Radiology Cases with Reference to Magnetic Resonance Imaging of the Lumbar Spine

Abstract

This systematic review aims to evaluate the recent advancements and applications of ChatGPT in reporting radiology cases, focusing on magnetic resonance imaging (MRI) of the lumbar spine. The review adheres to the PRISMA guidelines and explores how artificial intelligence (AI) can improve the accuracy, efficiency, and consistency of radiology reports. A comprehensive literature search was conducted across databases such as PubMed, Scopus, and IEEE Xplore, covering articles published from January 2022 to September 2024. The inclusion criteria were studies that examined the use of ChatGPT or similar AI models in radiology reporting, specifically focusing on MRI of the lumbar spine. Studies not involving ChatGPT or AI, non-radiology-related studies, and articles not available in full text were excluded. Data extraction and synthesis were performed in line with PRISMA guidelines. A total of 18 studies met the inclusion criteria, with 12 focusing on MRI of the lumbar spine. The findings indicate that ChatGPT shows potential in standardizing radiology reports, improving diagnostic accuracy, and reducing the time needed for report generation. However, challenges such as the necessity for clinical context, risks of over-reliance on AI, and issues regarding the interpretability of AI outputs were noted. ChatGPT and similar AI models have considerable potential in radiology reporting, especially for lumbar spine MRI. However, careful implementation, continuous updates, and rigorous validation are crucial to ensure these tools enhance rather than replace the expertise of radiologists.

Ähnliche Arbeiten