Purpose: The purpose of this study was to systematically review the reported performances of ChatGPT, identify potential limitations, and explore future directions for its integration, optimization, and ethical considerations in radiology applications.
Materials And Methods: After a comprehensive review of PubMed, Web of Science, Embase, and Google Scholar databases, a cohort of published studies was identified up to January 1, 2024, utilizing ChatGPT for clinical radiology applications.
Results: Out of 861 studies derived, 44 studies evaluated the performance of ChatGPT; among these, 37 (37/44; 84.1%) demonstrated high performance, and seven (7/44; 15.9%) indicated it had a lower performance in providing information on diagnosis and clinical decision support (6/44; 13.6%) and patient communication and educational content (1/44; 2.3%). Twenty-four (24/44; 54.5%) studies reported the proportion of ChatGPT's performance. Among these, 19 (19/24; 79.2%) studies recorded a median accuracy of 70.5%, and in five (5/24; 20.8%) studies, there was a median agreement of 83.6% between ChatGPT outcomes and reference standards [radiologists' decision or guidelines], generally confirming ChatGPT's high accuracy in these studies. Eleven studies compared two recent ChatGPT versions, and in ten (10/11; 90.9%), ChatGPTv4 outperformed v3.5, showing notable enhancements in addressing higher-order thinking questions, better comprehension of radiology terms, and improved accuracy in describing images. Risks and concerns about using ChatGPT included biased responses, limited originality, and the potential for inaccurate information leading to misinformation, hallucinations, improper citations and fake references, cybersecurity vulnerabilities, and patient privacy risks.
Conclusion: Although ChatGPT's effectiveness has been shown in 84.1% of radiology studies, there are still multiple pitfalls and limitations to address. It is too soon to confirm its complete proficiency and accuracy, and more extensive multicenter studies utilizing diverse datasets and pre-training techniques are required to verify ChatGPT's role in radiology.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1016/j.diii.2024.04.003 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Assembly of collagen fibers into contiguous dense and loose regions of subcutaneous fascia.
Connect Tissue Res
January 2025
Graduate School of Engineering, Kogakuin University, Hachioji, Tokyo, Japan.
Objective: This study aimed to investigate the collagen fiber structure of the subcutaneous fascia, a connective tissue layer between the skin and epimysium.
Methods: Fascia samples with varying extensibility were examined using biochemical and microscopic methods.
Results: Loose fascia, the more extensible type, displayed sparsely distributed collagen fibers, while dense fascia showed tightly packed collagen fiber bundles.
Dopamine genetic risk scores and psychiatric symptoms: Interacting risk factors for impulse control behaviours in de novo Parkinson's disease.
J Neuropsychol
January 2025
Department of Health, Medical and Neuropsychology, Leiden University, Leiden, The Netherlands.
Up to 45% of patients with Parkinson's disease (PD) experience impulse control disorders (ICDs), characterized by a loss of voluntary control over impulses, drives or temptations. This study aimed to investigate whether previously identified genetic and psychiatric risk factors interact towards the development of ICDs in PD. A total of 278 de novo PD patients (ICD-free at enrollment) were selected from the Parkinson's Progression Markers Initiative database.
View Article and Find Full Text PDFBone Marrow Aspirate Concentrate Combined With an Appropriate Carrier Effectively Promotes Bone-Tendon Interface Healing in a Rabbit Model of Chronic Rotator Cuff Tear.
Am J Sports Med
January 2025
Department of Orthopaedic Surgery, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, Republic of Korea.
Background: The efficacy of bone marrow aspirate concentrate (BMAC) in promoting bone-tendon interface (BTI) healing without any carriers remains a subject of debate.
Purpose: To evaluate BMAC effects with different carriers on tendon regeneration in a rabbit model of chronic rotator cuff tear.
Study Design: Controlled laboratory study.
Cultural differences in diagnosis and treatment perceptions: Turkish collectivistic representations of common mental disorders.
Int J Qual Stud Health Well-being
December 2025
Department of Language and Communication, Centre for Language Studies, Radboud University, Nijmegen, The Netherlands.
Purpose: Attention-deficit/hyperactivity disorder (ADHD) is less diagnosed among Turkish children, and Turkish clients drop out more often from depression treatments than Dutch clients. This article proposes that cultural differences in collectivistic versus individualistic perceptions of getting an ADHD diagnosis and being treated for depression might explain these ethnic disparities, which have been explored in this study.
Methods: Nine focus group discussions with Turkish individuals and 18 interviews with primary mental health practitioners were conducted.
Descriptive Epidemiology of Complete ACL Tears in the Skeletally Immature Population: A Prospective Multicenter PLUTO Study.
Am J Sports Med
January 2025
Department of Orthopaedic Surgery, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts, USA.
Background: Anterior cruciate ligament (ACL) injuries are common in pediatric and adolescent patients. Understanding this population's injury characteristics and treatment strategies is vital for managing this high-risk group.
Purpose: To report the descriptive epidemiology and treatment strategies of a large cohort of skeletally immature patients with complete ACL tears.
Want 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!