OpenAlex · Aktualisierung stündlich · Letzte Aktualisierung: 27.03.2026, 04:16

Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.

PD45-03 ASSESSMENT OF ChatGPT-GENERATED PATIENT-CENTERED PROSTATE BIOPSY REPORTS

2024·0 Zitationen·The Journal of Urology
Volltext beim Verlag öffnen

0

Zitationen

10

Autoren

2024

Jahr

Abstract

You have accessJournal of UrologyProstate Cancer: Staging II (PD45)1 May 2024PD45-03 ASSESSMENT OF ChatGPT-GENERATED PATIENT-CENTERED PROSTATE BIOPSY REPORTS Erin S. Proctor, David J. Nusbaum, Gregory Raster, Alexander P. Glaser, Craig V. Labbate, Andrew Higgins, Brian T. Helfand, Robert C. Benirschke, John V. Groth, and James T. Kearns Erin S. ProctorErin S. Proctor , David J. NusbaumDavid J. Nusbaum , Gregory RasterGregory Raster , Alexander P. GlaserAlexander P. Glaser , Craig V. LabbateCraig V. Labbate , Andrew HigginsAndrew Higgins , Brian T. HelfandBrian T. Helfand , Robert C. BenirschkeRobert C. Benirschke , John V. GrothJohn V. Groth , and James T. KearnsJames T. Kearns View All Author Informationhttps://doi.org/10.1097/01.JU.0001008792.09108.b4.03AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: The highly specialized language used in prostate biopsy pathology reports coupled with low rates of adult basic health literacy leaves most patients unable to comprehend their own medical information. Patients' use of online search engines can lead to misinterpretation of results and emotional distress. New generative artificial intelligence (AI) tools like ChatGPT can simplify and summarize such complex texts. This study evaluates patient-centered prostate biopsy reports generated by ChatGPT. METHODS: Thirty-five (35) artificial prostate biopsy reports were synthesized using permutations of each of the five risk groups as designated by the National Comprehensive Cancer Network (NCCN) Guidelines. Each report was individually entered into ChatGPT 4.0 with the same customized instructions: "We would like to create patient reports for prostate biopsies to explain the implications of the results. They should have a 7th grade reading level. Please address each of the following questions in the report: 1. Why do we do biopsies? 2. What did it show? 3. How much cancer is there? 4. How bad is it? 5. What is the overall risk to my life? 6. Was there anything else?". The ChatGPT-generated explanations were evaluated by four attending urologists. Survey responses were compared across risk groups. RESULTS: Overall, survey respondents rated the AI-generated explanations as mostly accurate and complete. For 66% of reports, all 4 respondents reported they would feel comfortable sharing the report with a patient and/or healthcare provider. 49% of reports were deemed incomplete by at least one respondent; however, 91% of reports were deemed complete by at least 3 out of the 4 respondents. Regarding the time needed to edit the report, 91% of reports were deemed acceptable as-is by at least 1 respondent (100% of low-risk reports and 70% of high-risk reports). Only one (3%) report (very low risk) was deemed acceptable as-is by all 4 respondents. For all reports (100%), at least one respondent felt it would take <10 minutes to edit. Reports with a lower NCCN risk group tended to yield more acceptable results than higher risk groups (Figure 1). CONCLUSIONS: ChatGPT was able to produce patient-centered prostate biopsy reports that would be clinically acceptable for urologists to share with patients after minimal editing. Download PPT Source of Funding: None © 2024 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 211Issue 5SMay 2024Page: e968 Advertisement Copyright & Permissions© 2024 by American Urological Association Education and Research, Inc.Metrics Author Information Erin S. Proctor More articles by this author David J. Nusbaum More articles by this author Gregory Raster More articles by this author Alexander P. Glaser More articles by this author Craig V. Labbate More articles by this author Andrew Higgins More articles by this author Brian T. Helfand More articles by this author Robert C. Benirschke More articles by this author John V. Groth More articles by this author James T. Kearns More articles by this author Expand All Advertisement PDF downloadLoading ...

Ähnliche Arbeiten