Automated and code-free development of a risk calculator using ChatGPT-4 for predicting diabetic retinopathy and macular edema without retinal imaging

International Journal of Retina and Vitreous, Journal Year: 2025, Volume and Issue: 11(1)

Published: Jan. 31, 2025

Abstract Background Diabetic retinopathy (DR) and macular edema (DME) are critical causes of vision loss in patients with diabetes. In many communities, access to ophthalmologists retinal imaging equipment is limited, making screening for diabetic complications difficult primary health care centers. We investigated whether ChatGPT-4, an advanced large-language-model chatbot, can develop risk calculators DR DME using check-up tabular data without the need or coding experience. Methods Data-driven prediction models were developed medical history laboratory blood test from Korea National Health Nutrition Examination Surveys (KNHANES). The dataset was divided into training (KNHANES 2017–2020) validation 2021) datasets. ChatGPT-4 used build formulas a web-based calculator tool. Logistic regression analysis performed by predict DME, followed automatic generation Hypertext Markup Language (HTML) code performance evaluated areas under curves receiver operating characteristic curve (ROC-AUCs). Results successfully operational on web browser any set showed ROC-AUCs 0.786 0.835 predicting respectively. comparable those created various machine-learning tools. Conclusion By utilizing code-free prompts, we overcame technical barriers associated skills developing models, it feasible prediction. Our approach offers easily accessible tool DM during check-ups, imaging. Based this automatically workers will be able effectively screen who require examinations only data. Future research should focus validating diverse populations exploring integration more comprehensive clinical enhance predictive performance. Graphical

Language: Английский

---

Evaluating Large Language Models for Burning Mouth Syndrome Diagnosis

Journal of Pain Research, Journal Year: 2025, Volume and Issue: Volume 18, P. 1387 - 1405

Published: March 1, 2025

Large language models have been proposed as diagnostic aids across various medical fields, including dentistry. Burning mouth syndrome, characterized by burning sensations in the oral cavity without identifiable cause, poses challenges. This study explores accuracy of large identifying hypothesizing potential limitations. Clinical vignettes 100 synthesized syndrome cases were evaluated using three (ChatGPT-4o, Gemini Advanced 1.5 Pro, and Claude 3.5 Sonnet). Each vignette included patient demographics, symptoms, history. prompted to provide a primary diagnosis, differential diagnoses, their reasoning. Accuracy was determined comparing responses with expert evaluations. ChatGPT achieved an rate 99%, while Gemini's 89% (p < 0.001). Misdiagnoses Persistent Idiopathic Facial Pain combined diagnoses inappropriate conditions. Differences also observed reasoning patterns additional data requests models. Despite high overall accuracy, exhibited variations approaches occasional errors, underscoring importance clinician oversight. Limitations include nature vignettes, over-reliance on exclusionary criteria, challenges differentiating overlapping disorders. demonstrate strong supplementary tools for especially settings lacking specialist expertise. However, reliability depends thorough assessment verification. Integrating into routine diagnostics could enhance early detection management, ultimately improving clinical decision-making dentists specialists alike.

Language: Английский

---

Are chatbots a reliable source for patient frequently asked questions on neck masses?

European Archives of Oto-Rhino-Laryngology, Journal Year: 2025, Volume and Issue: unknown

Published: April 30, 2025

Language: Английский

---

Large Language Models: Pioneering New Educational Frontiers in Childhood Myopia

Ophthalmology and Therapy, Journal Year: 2025, Volume and Issue: unknown

Published: April 21, 2025

This study aimed to evaluate the performance of three large language models (LLMs), namely ChatGPT-3.5, ChatGPT-4o (o1 Preview), and Google Gemini, in producing patient education materials (PEMs) improving readability online PEMs on childhood myopia. LLM-generated responses were assessed using prompts. Prompt A requested "Write educational material myopia." B added a modifier specifying "a sixth-grade reading level FKGL (Flesch-Kincaid Grade Level) formula." C rewrite existing FKGL. Reponses for quality (DISCERN tool), (FKGL, SMOG (Simple Measure Gobbledygook)), Patient Education Materials Assessment Tool (PEMAT, understandability/actionability), accuracy. (01) ChatGPT-3.5 generated good-quality 52.8 52.7, respectively); however, declined from prompt (p = 0.001 p 0.013). Gemini produced fair-quality 43) but improved with 0.02). All exceeded 70% PEMAT understandability threshold failed actionability (40%). No misinformation was identified. Readability B; achieved or below (FGKL 6 ± 0.6 6.2 0.3), while did not 7 0.6). outperformed < 0.001) comparable 0.846). across all LLMs, Preview) showing most significant gains (FKGL 5.8 1.5; 0.001). demonstrates potential accurate, good-quality, understandable PEMs,

Language: Английский

---

Transforming Brazilian healthcare with AI: progress and future perspectives

Discover Health Systems, Journal Year: 2025, Volume and Issue: 4(1)

Published: May 8, 2025

Language: Английский

---