Comparative analysis of diagnostic accuracy in adult congenital heart disease: A study of three physician groups and ChatGPT

Edward Itelman; Einat Birk; Ayelet Machtei; Rafael Hirsch; Nili Schamroth Pravda

doi:10.1016/j.ijcchd.2024.100547

Comparative analysis of diagnostic accuracy in adult congenital heart disease: A study of three physician groups and ChatGPT

Edward Itelman^*
, Einat Birk
, Ayelet Machtei
, Rafael Hirsch
, Nili Schamroth Pravda

^*Corresponding author for this work

Research output: Contribution to journal › Letter › peer-review

2 Scopus citations

Abstract

This research letter evaluates the diagnostic accuracy of three physician groups—pediatric cardiologists, adult congenital heart disease (ACHD) specialists, and general cardiologists—and compares it to the performance of an AI (Artificial Intelligence) LLM model. Based on eight targeted questions in adult congenital heart disease, the analysis highlights significant disparities in correct answer rates, with congenital specialists achieving the highest accuracy and the AI model – the lowest.

Original language	English
Article number	100547
Journal	International Journal of Cardiology Congenital Heart Disease
Volume	18
DOIs	https://doi.org/10.1016/j.ijcchd.2024.100547
State	Published - Dec 2024

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Access to Document

10.1016/j.ijcchd.2024.100547

Cite this

@article{66b498bced0847a4b019347384a7f1c7,

title = "Comparative analysis of diagnostic accuracy in adult congenital heart disease: A study of three physician groups and ChatGPT",

abstract = "This research letter evaluates the diagnostic accuracy of three physician groups—pediatric cardiologists, adult congenital heart disease (ACHD) specialists, and general cardiologists—and compares it to the performance of an AI (Artificial Intelligence) LLM model. Based on eight targeted questions in adult congenital heart disease, the analysis highlights significant disparities in correct answer rates, with congenital specialists achieving the highest accuracy and the AI model – the lowest.",

author = "Edward Itelman and Einat Birk and Ayelet Machtei and Rafael Hirsch and Pravda, \{Nili Schamroth\}",

note = "Publisher Copyright: {\textcopyright} 2024 The Authors",

year = "2024",

month = dec,

doi = "10.1016/j.ijcchd.2024.100547",

language = "אנגלית",

volume = "18",

journal = "International Journal of Cardiology Congenital Heart Disease",

issn = "2666-6685",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Comparative analysis of diagnostic accuracy in adult congenital heart disease

T2 - A study of three physician groups and ChatGPT

AU - Itelman, Edward

AU - Birk, Einat

AU - Machtei, Ayelet

AU - Hirsch, Rafael

AU - Pravda, Nili Schamroth

PY - 2024/12

Y1 - 2024/12

N2 - This research letter evaluates the diagnostic accuracy of three physician groups—pediatric cardiologists, adult congenital heart disease (ACHD) specialists, and general cardiologists—and compares it to the performance of an AI (Artificial Intelligence) LLM model. Based on eight targeted questions in adult congenital heart disease, the analysis highlights significant disparities in correct answer rates, with congenital specialists achieving the highest accuracy and the AI model – the lowest.

AB - This research letter evaluates the diagnostic accuracy of three physician groups—pediatric cardiologists, adult congenital heart disease (ACHD) specialists, and general cardiologists—and compares it to the performance of an AI (Artificial Intelligence) LLM model. Based on eight targeted questions in adult congenital heart disease, the analysis highlights significant disparities in correct answer rates, with congenital specialists achieving the highest accuracy and the AI model – the lowest.

UR - https://www.scopus.com/pages/publications/85208516153

U2 - 10.1016/j.ijcchd.2024.100547

DO - 10.1016/j.ijcchd.2024.100547

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.letter???

C2 - 39713227

AN - SCOPUS:85208516153

SN - 2666-6685

VL - 18

JO - International Journal of Cardiology Congenital Heart Disease

JF - International Journal of Cardiology Congenital Heart Disease

M1 - 100547

ER -

Comparative analysis of diagnostic accuracy in adult congenital heart disease: A study of three physician groups and ChatGPT

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this