Introduction: The adoption of advanced reasoning models, such as ChatGPT O1 and DeepSeek-R1, represents a pivotal step forward in clinical decision support, particularly in pediatrics. ChatGPT O1 employs “chain-of-thought reasoning” (CoT) to enhance structured problem-solving, while DeepSeek-R1 introduces self-reflection capabilities through reinforcement learning. This study aimed to evaluate the diagnostic accuracy and clinical utility of these models in pediatric scenarios using the MedQA dataset. Materials and Methods: A total of 500 multiple-choice pediatric questions from the MedQA dataset were presented to ChatGPT O1 and DeepSeek-R1. Each question included four or more options, with one correct answer. The models were evaluated under uniform conditions, with performance metrics including accuracy, Cohen’s Kappa, and chi-square tests applied to assess agreement and statistical significance. Responses were analyzed to determine the models effectiveness in addressing clinical questions. Results: ChatGPT O1 achieved a diagnostic accuracy of 92.8%, significantly outperforming DeepSeek-R1, which scored 87.0% ( p < 0.00001). The CoT reasoning technique used by ChatGPT O1 allowed for more structured and reliable responses, reducing the risk of errors. Conversely, DeepSeek-R1, while slightly less accurate, demonstrated superior accessibility and adaptability due to its open-source nature and emerging self-reflection capabilities. Cohen’s Kappa (K=0.20) indicated low agreement between the models, reflecting their distinct reasoning strategies. Conclusions: This study highlights the strengths of ChatGPT O1 in providing accurate and coherent clinical reasoning, making it highly suitable for critical pediatric scenarios. DeepSeek-R1, with its flexibility and accessibility, remains a valuable tool in resource-limited settings. Combining these models in an ensemble system could leverage their complementary strengths, optimizing decision support in diverse clinical contexts. Further research is warranted to explore their integration into multidisciplinary care teams and their application in real-world clinical settings.

This review examines the increasing use of artificial intelligence (AI) in rheumatology, focusing on its potential impact in key areas. AI, including machine learning (ML) and deep learning (DL), is revolutionizing diagnosis, treatment personalization, and prognosis prediction in rheumatologic diseases. Specifically, AI models based on convolutional neural networks (CNNs) demonstrate significant efficacy in analyzing medical images for disease classification and severity assessment. Predictive AI models also have the ability to forecast disease trajectories and treatment responses, enabling more informed clinical decisions. The role of wearable devices and mobile applications in continuous disease monitoring is discussed, although their effectiveness varies across studies. Despite existing challenges, such as data privacy concerns and issues of model generalizability, the compelling results highlight the transformative potential of AI in rheumatologic disease management. As AI technologies continue to evolve, further research will be essential to address these challenges and fully harness the potential of AI to improve patient outcomes in rheumatology.

This article examines the challenges and security concerns associated with the use of large language models (LLMs) like ChatGPT in the medical field, focusing particularly on the phenomena known as “LLM jailbreaking”. As LLMs increasingly perform complex tasks involving sensitive information, the risk of their misuse becomes significant. Jailbreaking, originally a concept from software systems, refers to bypassing the restrictions set by developers to unlock new functionalities. This practice has spread to LLMs, where users manipulate model inputs to elicit responses otherwise restricted by ethical and safety guidelines. Our research specifically targets the implications of jailbreaking using ChatGPT versions 3.5 and 4 as case studies in two medical scenarios: pneumonia treatment and a recipe for a drink based on drugs. We demonstrate how modified prompts—such as those used in “Role Playing”—can alter the model’s output, potentially leading to the provision of harmful medical advice or the disclosure of sensitive information. Findings indicate that while newer versions of ChatGPT show improved resistance to such manipulations, significant risks remain. The paper discusses the dual necessity of refining these defensive mechanisms and maintaining ethical oversight to prevent misuse. As LLMs permeate more deeply into critical areas like healthcare, the balance between leveraging their capabilities and safeguarding against risks becomes paramount. This analysis underscores the urgent need for ongoing research into more robust security measures and ethical guidelines to ensure the safe use of transformative artificial technology technologies in sensitive fields.

Background: This study investigates the application and efficacy of large language models (LLMs) in pediatric medicine, focusing on their capability to assist in training and decision support for healthcare professionals. Given the unique challenges in pediatric care, such as age-specific conditions and dosing requirements, we aim to evaluate the performance of various LLMs in this specialized field. Methods: We conducted a comparative analysis of several LLMs, including Claude 3-OPUS, ChatGPT 3.5 and 4, Gemini AI, Llama 2 70B, and Mixtral 8x7B. The models were tested on 227 multiple-choice pediatric questions in Italian before and after undergoing specialized training. The training data consisted of pediatric articles from a medical journal, ensuring compliance with HIPAA regulations by using de-identified and anonymous data. Results: The performance of the LLMs varied significantly. ChatGPT 3.5 improved from 65.20% to 83.70% accuracy (P<0.01) after training, while ChatGPT 4 increased from 77.09% to 91.62% (P<0.01). Gemini 1.0 and Mixtral 8x7B recorded accuracies of 70.48% and 71.37% respectively, both showing significant improvements post-training. Llama 2 70B had a lower performance, improving from 47.58% to 52.86%. Claude 3-OPUS demonstrated robust performance with an 82.82% accuracy pre-training, improving to 95.59% post-training. Conclusions: Our analysis confirms the effectiveness of LLMs in pediatric medicine, highlighting their potential in training and decision support. The study emphasizes the need for specific training datasets that reflect the complexities of pediatric conditions to tailor the models accurately. Moreover, there is a significant opportunity to utilize open-source models as a foundation for developing customized systems through training on dedicated datasets. These strategies promise to enhance the accuracy and accessibility of pediatric healthcare, ultimately improving outcomes for young patients.

A bstract: Large Language Models (LLMs) are rapidly emerging as promising tools in the healthcare field, yet their effectiveness in pediatric contexts remains underexplored. This study evaluated the performance of eight contemporary LLMs, released between 2024 and 2025, in answering multiple-choice questions from the MedQA dataset, stratified into two distinct categories: adult medicine (1461 questions) and pediatrics (1653 questions). Models were tested using a standardized prompting methodology with default hyperparameters, simulating real-world use by non-expert clinical users. Accuracy scores for adult and pediatric subsets were statistically compared using the chi-square test with Yates’ correction. Five models (Amazon Nova Pro 1.0, GPT 3.5-turbo-0125, Gemini 2.0 Flash, Grok 2, and Claude 3 Sonnet) demonstrated significantly lower performance on pediatric questions, with accuracy drops of up to more than 10 percentage points. In contrast, ChatGPT-4o, Gemini 1.5 Pro, and Claude 3.5 Sonnet showed comparable performance across both domains, with ChatGPT-4o achieving the most balanced result (accuracy: 83.57% adult, 83.18% pediatric; p = 0.80). These findings suggest that while some models struggle with pediatric-specific content, more recent and advanced LLMs may offer improved generalizability and domain robustness. The observed variability highlights the critical importance of domain-specific validation prior to clinical implementation, particularly in specialized fields such as pediatrics.