Evaluation of Artificial Intelligence Based Clinical Decision Support in Pregnancy Specific Dermatological Lesions

Aim: Artificial intelligence-based systems are increasingly used for recognizing and assessing skin lesions, yet studies focusing on pregnancy-specific dermatological findings are limited. This study aimed to compare the accuracy of artificial intelligence based large language models in diagnosing pregnancy related dermatological lesions, explaining their etiopathogenesis, and providing clinically appropriate treatment recommendations.Materials and Methods: In this observational study, ten pregnancy-specific dermatological lesions were evaluated using ChatGPT (version 5) and Google Gemini (version 2.5). Each case was presented with an image from Williams Obstetrics, 26th edition, and a standardized patient statement: “I am pregnant. I have a skin lesion like this. What is it, why does it occur, and what should I do for treatment?” Responses were blindly rated by obstetrics and gynecology and dermatology specialists for diagnostic accuracy, etiopathogenesis explanation, and treatment appropriateness, using a five-point Likert scale. Cohen’s kappa was used to assess inter-rater agreement; the Wilcoxon test compared the models; Spearman’s correlation examined score relationships. A p-value of less than 0.05 was considered statistically significant.Results: ChatGPT outperformed Gemini in etiopathogenesis explanation (p = 0.002) and treatment recommendations (p = 0.002), with no significant difference in diagnostic accuracy (p = 0.336). ChatGPT generally provided more detailed, applicable responses, whereas Gemini often gave limited or purely directive information. Correlations between criteria were weak and statistically insignificant for both models.Conclusions: ChatGPT was superior to Gemini in etiopathogenesis and treatment recommendations but comparable in diagnostic accuracy. However, no artificial intelligence system can replace clinical evaluation by a physician.

Authors

Institutions

Publication Details

Journal
Genel Tıp Dergisi
Published
2026-11-18
DOI
https://doi.org/10.54005/geneltip.1779446
Primary Topic
Artificial Intelligence in Healthcare and Education
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Evaluation of Artificial Intelligence Based Clinical Decision Support in Pregnancy Specific Dermatological Lesions

Emre Vuraloğlu, Aybüke Tayarer
3 links
Genel Tıp Dergisi
Artificial Intelligence in Healthcare and Education
article

Evaluation of Artificial Intelligence Based Clinical Decision Support in Pregnancy Specific Dermatological Lesions

Emre Vuraloğlu, Aybüke Tayarer
article en

Abstract

Aim: Artificial intelligence-based systems are increasingly used for recognizing and assessing skin lesions, yet studies focusing on pregnancy-specific dermatological findings are limited. This study aimed to compare the accuracy of artificial intelligence based large language models in diagnosing pregnancy related dermatological lesions, explaining their etiopathogenesis, and providing clinically appropriate treatment recommendations.Materials and Methods: In this observational study, ten pregnancy-specific dermatological lesions were evaluated using ChatGPT (version 5) and Google Gemini (version 2.5). Each case was presented with an image from Williams Obstetrics, 26th edition, and a standardized patient statement: “I am pregnant. I have a skin lesion like this. What is it, why does it occur, and what should I do for treatment?” Responses were blindly rated by obstetrics and gynecology and dermatology specialists for diagnostic accuracy, etiopathogenesis explanation, and treatment appropriateness, using a five-point Likert scale. Cohen’s kappa was used to assess inter-rater agreement; the Wilcoxon test compared the models; Spearman’s correlation examined score relationships. A p-value of less than 0.05 was considered statistically significant.Results: ChatGPT outperformed Gemini in etiopathogenesis explanation (p = 0.002) and treatment recommendations (p = 0.002), with no significant difference in diagnostic accuracy (p = 0.336). ChatGPT generally provided more detailed, applicable responses, whereas Gemini often gave limited or purely directive information. Correlations between criteria were weak and statistically insignificant for both models.Conclusions: ChatGPT was superior to Gemini in etiopathogenesis and treatment recommendations but comparable in diagnostic accuracy. However, no artificial intelligence system can replace clinical evaluation by a physician.

Genel Tıp DergisiVol. 36(2026)
Ahi Evran University (TR), Kirsehir Training and Research Hospital (TR)
Good health and well-being
Artificial Intelligence in Healthcare and Education
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon