Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
Evaluating the Accuracy of Artificial Intelligence vs. Nephrologist in Addressing Common Questions about Diabetic Nephropathy from Patients (Preprint)
0
Zitationen
4
Autoren
2024
Jahr
Abstract
<sec> <title>UNSTRUCTURED</title> Diabetic nephropathy (DN) is one of the most frequent and severe complications of diabetes mellitus that requires early reduction for the individuals most at risk. Effective intervention and management of DN requires a patient's deep insight and helpful information. Patients require accurate information to obtain prevention advice about their kidney disease to make appropriate decisions; individuals may seek artificial intelligence (AI) as online sources instead of communicating with clinicians. Therefore, to evaluate ChatGPT and Google Bard's capabilities in providing accurate information regarding DN compared to nephrologists, we assessed their performance in answering questions related to DN. Our study revealed AI would not eliminate the need for open and detailed discussion with healthcare professionals regarding medical concerns and treatment plans, as they indicated inconsistencies in responses. Additionally, it is imperative to provide real-time health information updates to meet the needs of individuals with DN. </sec>
Ähnliche Arbeiten
Biostatistical Analysis
1996 · 35.448 Zit.
UCI Machine Learning Repository
2007 · 24.319 Zit.
An introduction to ROC analysis
2005 · 20.780 Zit.
The use of the area under the ROC curve in the evaluation of machine learning algorithms
1997 · 7.150 Zit.
A method of comparing the areas under receiver operating characteristic curves derived from the same cases.
1983 · 7.072 Zit.