Cited by A Cross-Sectional Study Comparing Patient Information Guides Generated by ChatGPT and Google Gemini for Common Radiological Procedures

Comparative Accuracy of ChatGPT 4.0 and Google Gemini in Answering Pediatric Radiology Text-Based Questions DOI

Mohammed Abdul Sami,

Abdul Samad,

Keyur Parekh

et al.

Cureus, Journal Year: 2024, Volume and Issue: unknown

Published: Oct. 5, 2024

This study evaluates the accuracy of two AI language models, ChatGPT 4.0 and Google Gemini (as August 2024), in answering a set 79 text-based pediatric radiology questions from "Pediatric Imaging: A Core Review." Accurate interpretation text images is critical radiology, making tools valuable medical education.

Language: Английский

Citations

Cultivating diagnostic clarity: The importance of reporting artificial intelligence confidence levels in radiologic diagnoses DOI

Mobina Fathi, Kimia Vakili, Ramtin Hajibeygi

et al.

Clinical Imaging, Journal Year: 2024, Volume and Issue: 117, P. 110356 - 110356

Published: Nov. 13, 2024

Language: Английский

Citations

Evaluation of ChatGPT 4.0 in Thoracic Imaging and Diagnostics DOI

Golnaz Lotfian,

Keyur Parekh,

Mohammed Abdul Sami

et al.

Cureus, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 15, 2024

Recent advancements in natural language processing (NLP) have profoundly transformed the medical industry, enhancing large cohort data analysis, improving diagnostic capabilities, and streamlining clinical workflows. Among leading tools this domain is ChatGPT 4.0 (OpenAI, San Francisco, California, US), a commercial NLP model widely used across various applications. This study evaluates performance of specifically thoracic imaging by assessing its ability to answer questions related field. We utilized respond multiple-choice derived from scenarios, followed rigorous statistical analysis assess accuracy variability different subgroups. Our revealed significant Overall, achieved an impressive 84.9% diagnosing radiology questions. It excelled terminology signs, achieving perfect scores, demonstrated strong intensive care normal anatomy categories, with accuracies 90% 80%, respectively. In pathology subgroups, average 89.1%, particularly excelling infectious pneumonia atelectasis, though it scored lower diffuse alveolar disease (66.7%). For disease-related questions, mean was 79.1%, scores several specific subcategories. However, notably for vascular (50%) lung cancer conclusion, while shows potential conditions, identified underscores necessity ongoing research refinement transformer architecture. will enhance reliability applicability broader patient settings.

Language: Английский

Citations

ChatGPT-4 Turbo and Meta’s LLaMA 3.1: A Relative Analysis of Answering Radiology Text-Based Questions DOI

Mohammed Abdul Sami,

Abdul Samad,

Keyur Parekh

et al.

Cureus, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 24, 2024

This study aimed to compare the accuracy of two AI models - OpenAI's GPT-4 Turbo (San Francisco, CA) and Meta's LLaMA 3.1 (Menlo Park, when answering a standardized set pediatric radiology questions. The primary objective was evaluate overall each model, while secondary assess their performance within subsections.

Language: Английский

Citations

A Cross-Sectional Study Comparing Patient Information Guides Generated by ChatGPT and Google Gemini for Common Radiological Procedures DOI

V.J. Phillips,

Nidhi L Rao,

Yashasvi H Sanghvi

et al.

Cureus, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 30, 2024

Artificial intelligence (AI) plays a significant role in creating brochures on radiological procedures for patient education. Thus, this study aimed to evaluate the responses generated by ChatGPT (San Francisco, CA: OpenAI) and Google Gemini (Mountain View, LLC) abdominal ultrasound, CT scan, MRI.

Language: Английский

Citations