Fabrication and errors in the bibliographic citations generated by ChatGPT.
Ontology highlight
ABSTRACT:
SUBMITTER: Walters WH
PROVIDER: S-EPMC10484980 | biostudies-literature | 2023 Sep
REPOSITORIES: biostudies-literature
Publications
Fabrication and errors in the bibliographic citations generated by ChatGPT.
Walters William H WH Wilder Esther Isabelle EI
Scientific reports 20230907 1
Although chatbots such as ChatGPT can facilitate cost-effective text generation and editing, factually incorrect responses (hallucinations) limit their utility. This study evaluates one particular type of hallucination: fabricated bibliographic citations that do not represent actual scholarly works. We used ChatGPT-3.5 and ChatGPT-4 to produce short literature reviews on 42 multidisciplinary topics, compiling data on the 636 bibliographic citations (references) found in the 84 papers. We then se ...[more]
Similar Datasets
Project description:This paper analyzes the effects of the co-authorship and bibliographic coupling networks on the citations received by scientific articles. It expands prior research that limited its focus on the position of co-authors and incorporates the effects of the use of knowledge sources within articles: references. By creating a network on the basis of shared references, we propose a way to understand whether an article bridges among extant strands of literature and infer the size of its research community and its embeddedness. Thus, we map onto the article--our unit of analysis--the metrics of authors' position in the co-authorship network and of the use of knowledge on which the scientific article is grounded. Specifically, we adopt centrality measures--degree, betweenneess, and closeness centrality--in the co-authorship network and degree, betweenness centrality and clustering coefficient in the bibliographic coupling and show their influence on the citations received in first two years after the year of publication. Findings show that authors' degree positively impacts citations. Also closeness centrality has a positive effect manifested only when the giant component is relevant. Author's betweenness centrality has instead a negative effect that persists until the giant component--largest component of the network in which all nodes can be linked by a path--is relevant. Moreover, articles that draw on fragmented strands of literature tend to be cited more, whereas the size of the scientific research community and the embeddedness of the article in a cohesive cluster of literature have no effect.
| S-EPMC4049820 | biostudies-literature
Project description:As an important biomedical database, PubMed provides users with free access to abstracts of its documents. However, citations between these documents need to be collected from external data sources. Although previous studies have investigated the coverage of various data sources, the quality of citations is underexplored. In response, this study compares the coverage and citation quality of five freely available data sources on 30 million PubMed documents, including OpenCitations Index of CrossRef open DOI-to-DOI citations (COCI), Dimensions, Microsoft Academic Graph (MAG), National Institutes of Health's Open Citation Collection (NIH-OCC), and Semantic Scholar Open Research Corpus (S2ORC). Three gold standards and five metrics are introduced to evaluate the correctness and completeness of citations. Our results indicate that Dimensions is the most comprehensive data source that provides references for 62.4% of PubMed documents, outperforming the official NIH-OCC dataset (56.7%). Over 90% of citation links in other data sources can also be found in Dimensions. The coverage of MAG, COCI, and S2ORC is 59.6%, 34.7%, and 23.5%, respectively. Regarding the citation quality, Dimensions and NIH-OCC achieve the best overall results. Almost all data sources have a precision higher than 90%, but their recall is much lower. All databases have better performances on recent publications than earlier ones. Meanwhile, the gaps between different data sources have diminished for the documents published in recent years. This study provides evidence for researchers to choose suitable PubMed citation sources, which is also helpful for evaluating the citation quality of free bibliographic databases.
| S-EPMC8542188 | biostudies-literature
Project description:ChatGPT and similar generative AI models have attracted hundreds of millions of users and have become part of the public discourse. Many believe that such models will disrupt society and lead to significant changes in the education system and information generation. So far, this belief is based on either colloquial evidence or benchmarks from the owners of the models-both lack scientific rigor. We systematically assess the quality of AI-generated content through a large-scale study comparing human-written versus ChatGPT-generated argumentative student essays. We use essays that were rated by a large number of human experts (teachers). We augment the analysis by considering a set of linguistic characteristics of the generated essays. Our results demonstrate that ChatGPT generates essays that are rated higher regarding quality than human-written essays. The writing style of the AI models exhibits linguistic characteristics that are different from those of the human-written essays. Since the technology is readily available, we believe that educators must act immediately. We must re-invent homework and develop teaching concepts that utilize these AI models in the same way as math utilizes the calculator: teach the general concepts first and then use AI tools to free up time for other learning objectives.
| S-EPMC10616290 | biostudies-literature
Project description:ObjectiveTo determine if ChatGPT can generate useful suggestions for improving clinical decision support (CDS) logic and to assess noninferiority compared to human-generated suggestions.MethodsWe supplied summaries of CDS logic to ChatGPT, an artificial intelligence (AI) tool for question answering that uses a large language model, and asked it to generate suggestions. We asked human clinician reviewers to review the AI-generated suggestions as well as human-generated suggestions for improving the same CDS alerts, and rate the suggestions for their usefulness, acceptance, relevance, understanding, workflow, bias, inversion, and redundancy.ResultsFive clinicians analyzed 36 AI-generated suggestions and 29 human-generated suggestions for 7 alerts. Of the 20 suggestions that scored highest in the survey, 9 were generated by ChatGPT. The suggestions generated by AI were found to offer unique perspectives and were evaluated as highly understandable and relevant, with moderate usefulness, low acceptance, bias, inversion, redundancy.ConclusionAI-generated suggestions could be an important complementary part of optimizing CDS alerts, can identify potential improvements to alert logic and support their implementation, and may even be able to assist experts in formulating their own suggestions for CDS improvement. ChatGPT shows great potential for using large language models and reinforcement learning from human feedback to improve CDS alert logic and potentially other medical areas involving complex, clinical logic, a key step in the development of an advanced learning health system.
| S-EPMC10280357 | biostudies-literature
Project description:Informed consent is a crucial requirement of a patient's surgical care but can be a burdensome task. Artificial intelligence (AI) and machine learning language models may provide an alternative approach to writing detailed, readable consent forms in an efficient manner. No studies have assessed the accuracy and completeness of AI-generated consents for aesthetic plastic surgeries. This study aims to compare the length, reading level, accuracy, and completeness of informed consent forms that are AI chatbot (ChatGPT-4; OpenAI, San Francisco, CA) generated vs plastic surgeon generated for the most commonly performed aesthetic plastic surgeries. This study is a cross-sectional design comparing informed consent forms created by the American Association of Plastic Surgeons (ASPS) with informed consent forms generated by ChatGPT-4 for the 5 most commonly performed plastic surgery procedures: liposuction, breast augmentation, abdominoplasty, breast lift, and blepharoplasty. The average word count of ChatGPT forms was lower than the ASPS-generated forms (1023 vs 2901, P = .01). Average reading level for ChatGPT forms was also lower than the ASPS forms (11.2 vs 12.5, P = .02). There was no difference between accuracy and completeness scores for general descriptions of the surgery, risks, benefits, or alternatives. The mean overall impression score for ChatGPT consents was 2.33, whereas it was 2.23 for ASPS consent forms (P = .18). In this study, the authors demonstrate that informed consent forms generated by ChatGPT were significantly shorter and more readable than ASPS forms with no significant difference in completeness and accuracy.Level of evidence 5 risk
| S-EPMC11561908 | biostudies-literature
Project description:ObjectivePatients now have direct access to their radiology reports, which can include complex terminology and be difficult to understand. We assessed ChatGPT's ability to generate summarized MRI reports for patients with prostate cancer and evaluated physician satisfaction with the artificial intelligence (AI)-summarized report.MethodsWe used ChatGPT to summarize five full MRI reports for patients with prostate cancer performed at a single institution from 2021 to 2022. Three summarized reports were generated for each full MRI report. Full MRI and summarized reports were assessed for readability using Flesch-Kincaid Grade Level (FK) score. Radiation oncologists were asked to evaluate the AI-summarized reports via an anonymous questionnaire. Qualitative responses were given on a 1-5 Likert-type scale. Fifty newly diagnosed prostate cancer patient MRIs performed at a single institution were additionally assessed for physician online portal response rates.ResultsFifteen summarized reports were generated from five full MRI reports using ChatGPT. The median FK score for the full MRI reports and summarized reports was 9.6 vs. 5.0, (p < 0.05), respectively. Twelve radiation oncologists responded to our questionnaire. The mean [SD] ratings for summarized reports were factual correctness (4.0 [0.6], understanding 4.0 [0.7]), completeness (4.1 [0.5]), potential for harm (3.5 [0.9]), overall quality (3.4 [0.9]), and likelihood to send to patient (3.1 [1.1]). Current physician online portal response rates were 14/50 (28%) at our institution.ConclusionsWe demonstrate a novel application of ChatGPT to summarize MRI reports at a reading level appropriate for patients. Physicians were likely to be satisfied with the summarized reports with respect to factual correctness, ease of understanding, and completeness. Physicians were less likely to be satisfied with respect to potential for harm, overall quality, and likelihood to send to patients. Further research is needed to optimize ChatGPT's ability to summarize radiology reports and understand what factors influence physician trust in AI-summarized reports.
| S-EPMC10734360 | biostudies-literature
Project description:IntroductionChatGPT, a sophisticated large language model (LLM), has garnered widespread attention for its ability to mimic human-like communication. As recent studies indicate a potential supportive role of ChatGPT in academic writing, we assessed the LLM's capacity to generate accurate and comprehensive scientific abstracts from published Randomised Controlled Trial (RCT) data, focusing on the adherence to the Consolidated Standards of Reporting Trials for Abstracts (CONSORT-A) statement, in comparison to the original authors' abstracts.MethodologyRCTs, identified in a PubMed/MEDLINE search post-September 2021 across various medical disciplines, were subjected to abstract generation via ChatGPT versions 3.5 and 4, following the guidelines of the respective journals. The overall quality score (OQS) of each abstract was determined by the total number of adequately reported components from the 18-item CONSORT-A checklist. Additional outcome measures included percent adherence to each CONOSORT-A item, readability, hallucination rate, and regression analysis of reporting quality determinants.ResultsOriginal abstracts achieved a mean OQS of 11.89 (95% CI: 11.23-12.54), outperforming GPT 3.5 (7.89; 95% CI: 7.32-8.46) and GPT 4 (5.18; 95% CI: 4.64-5.71). Compared to GPT 3.5 and 4 outputs, original abstracts were more adherent with 10 and 14 CONSORT-A items, respectively. In blind assessments, GPT 3.5-generated abstracts were deemed most readable in 62.22% of cases which was significantly greater than the original (31.11%; P = 0.003) and GPT 4-generated (6.67%; P<0.001) abstracts. Moreover, ChatGPT 3.5 exhibited a hallucination rate of 0.03 items per abstract compared to 1.13 by GPT 4. No determinants for improved reporting quality were identified for GPT-generated abstracts.ConclusionsWhile ChatGPT could generate more readable abstracts, their overall quality was inferior to the original abstracts. Yet, its proficiency to concisely relay key information with minimal error holds promise for medical research and warrants further investigations to fully ascertain the LLM's applicability in this domain.
| S-EPMC10866463 | biostudies-literature
Project description:BackgroundEvaluating the accuracy and educational utility of artificial intelligence-generated medical cases, especially those produced by large language models such as ChatGPT-4 (developed by OpenAI), is crucial yet underexplored.ObjectiveThis study aimed to assess the educational utility of ChatGPT-4-generated clinical vignettes and their applicability in educational settings.MethodsUsing a convergent mixed methods design, a web-based survey was conducted from January 8 to 28, 2024, to evaluate 18 medical cases generated by ChatGPT-4 in Japanese. In the survey, 6 main question items were used to evaluate the quality of the generated clinical vignettes and their educational utility, which are information quality, information accuracy, educational usefulness, clinical match, terminology accuracy (TA), and diagnosis difficulty. Feedback was solicited from physicians specializing in general internal medicine or general medicine and experienced in medical education. Chi-square and Mann-Whitney U tests were performed to identify differences among cases, and linear regression was used to examine trends associated with physicians' experience. Thematic analysis of qualitative feedback was performed to identify areas for improvement and confirm the educational utility of the cases.ResultsOf the 73 invited participants, 71 (97%) responded. The respondents, primarily male (64/71, 90%), spanned a broad range of practice years (from 1976 to 2017) and represented diverse hospital sizes throughout Japan. The majority deemed the information quality (mean 0.77, 95% CI 0.75-0.79) and information accuracy (mean 0.68, 95% CI 0.65-0.71) to be satisfactory, with these responses being based on binary data. The average scores assigned were 3.55 (95% CI 3.49-3.60) for educational usefulness, 3.70 (95% CI 3.65-3.75) for clinical match, 3.49 (95% CI 3.44-3.55) for TA, and 2.34 (95% CI 2.28-2.40) for diagnosis difficulty, based on a 5-point Likert scale. Statistical analysis showed significant variability in content quality and relevance across the cases (P<.001 after Bonferroni correction). Participants suggested improvements in generating physical findings, using natural language, and enhancing medical TA. The thematic analysis highlighted the need for clearer documentation, clinical information consistency, content relevance, and patient-centered case presentations.ConclusionsChatGPT-4-generated medical cases written in Japanese possess considerable potential as resources in medical education, with recognized adequacy in quality and accuracy. Nevertheless, there is a notable need for enhancements in the precision and realism of case details. This study emphasizes ChatGPT-4's value as an adjunctive educational tool in the medical field, requiring expert oversight for optimal application.
| S-EPMC11350316 | biostudies-literature
Project description:BackgroundThe diagnostic accuracy of differential diagnoses generated by artificial intelligence chatbots, including ChatGPT models, for complex clinical vignettes derived from general internal medicine (GIM) department case reports is unknown.ObjectiveThis study aims to evaluate the accuracy of the differential diagnosis lists generated by both third-generation ChatGPT (ChatGPT-3.5) and fourth-generation ChatGPT (ChatGPT-4) by using case vignettes from case reports published by the Department of GIM of Dokkyo Medical University Hospital, Japan.MethodsWe searched PubMed for case reports. Upon identification, physicians selected diagnostic cases, determined the final diagnosis, and displayed them into clinical vignettes. Physicians typed the determined text with the clinical vignettes in the ChatGPT-3.5 and ChatGPT-4 prompts to generate the top 10 differential diagnoses. The ChatGPT models were not specially trained or further reinforced for this task. Three GIM physicians from other medical institutions created differential diagnosis lists by reading the same clinical vignettes. We measured the rate of correct diagnosis within the top 10 differential diagnosis lists, top 5 differential diagnosis lists, and the top diagnosis.ResultsIn total, 52 case reports were analyzed. The rates of correct diagnosis by ChatGPT-4 within the top 10 differential diagnosis lists, top 5 differential diagnosis lists, and top diagnosis were 83% (43/52), 81% (42/52), and 60% (31/52), respectively. The rates of correct diagnosis by ChatGPT-3.5 within the top 10 differential diagnosis lists, top 5 differential diagnosis lists, and top diagnosis were 73% (38/52), 65% (34/52), and 42% (22/52), respectively. The rates of correct diagnosis by ChatGPT-4 were comparable to those by physicians within the top 10 (43/52, 83% vs 39/52, 75%, respectively; P=.47) and within the top 5 (42/52, 81% vs 35/52, 67%, respectively; P=.18) differential diagnosis lists and top diagnosis (31/52, 60% vs 26/52, 50%, respectively; P=.43) although the difference was not significant. The ChatGPT models' diagnostic accuracy did not significantly vary based on open access status or the publication date (before 2011 vs 2022).ConclusionsThis study demonstrates the potential diagnostic accuracy of differential diagnosis lists generated using ChatGPT-3.5 and ChatGPT-4 for complex clinical vignettes from case reports published by the GIM department. The rate of correct diagnoses within the top 10 and top 5 differential diagnosis lists generated by ChatGPT-4 exceeds 80%. Although derived from a limited data set of case reports from a single department, our findings highlight the potential utility of ChatGPT-4 as a supplementary tool for physicians, particularly for those affiliated with the GIM department. Further investigations should explore the diagnostic accuracy of ChatGPT by using distinct case materials beyond its training data. Such efforts will provide a comprehensive insight into the role of artificial intelligence in enhancing clinical decision-making.
| S-EPMC10594139 | biostudies-literature
Project description:BackgroundThe persistence of diagnostic errors, despite advances in medical knowledge and diagnostics, highlights the importance of understanding atypical disease presentations and their contribution to mortality and morbidity. Artificial intelligence (AI), particularly generative pre-trained transformers like GPT-4, holds promise for improving diagnostic accuracy, but requires further exploration in handling atypical presentations.ObjectiveThis study aimed to assess the diagnostic accuracy of ChatGPT in generating differential diagnoses for atypical presentations of common diseases, with a focus on the model's reliance on patient history during the diagnostic process.MethodsWe used 25 clinical vignettes from the Journal of Generalist Medicine characterizing atypical manifestations of common diseases. Two general medicine physicians categorized the cases based on atypicality. ChatGPT was then used to generate differential diagnoses based on the clinical information provided. The concordance between AI-generated and final diagnoses was measured, with a focus on the top-ranked disease (top 1) and the top 5 differential diagnoses (top 5).ResultsChatGPT's diagnostic accuracy decreased with an increase in atypical presentation. For category 1 (C1) cases, the concordance rates were 17% (n=1) for the top 1 and 67% (n=4) for the top 5. Categories 3 (C3) and 4 (C4) showed a 0% concordance for top 1 and markedly lower rates for the top 5, indicating difficulties in handling highly atypical cases. The χ2 test revealed no significant difference in the top 1 differential diagnosis accuracy between less atypical (C1+C2) and more atypical (C3+C4) groups (χ²1=2.07; n=25; P=.13). However, a significant difference was found in the top 5 analyses, with less atypical cases showing higher accuracy (χ²1=4.01; n=25; P=.048).ConclusionsChatGPT-4 demonstrates potential as an auxiliary tool for diagnosing typical and mildly atypical presentations of common diseases. However, its performance declines with greater atypicality. The study findings underscore the need for AI systems to encompass a broader range of linguistic capabilities, cultural understanding, and diverse clinical scenarios to improve diagnostic utility in real-world settings.
| S-EPMC11199925 | biostudies-literature