Citations

Background: Data accessibility remains a significant barrier in healthcare AI due to privacy constraints and logistical challenges. Synthetic data, which mimics real patient information while remaining both realistic and non-identifiable, offers a promising solution. Large Language Models (LLMs) create new opportunities to generate high-fidelity clinical conversations between patients and physicians. However, the value of this synthetic data depends on careful evaluation of its realism, accuracy, and practical relevance. Objective: To assess the performance of four leading LLMs: ChatGPT 4.5, ChatGPT 4o, Claude 3.7 Sonnet, and Gemini Pro 2.5 in generating synthetic transcripts of patient–physician interactions in plastic surgery scenarios. Methods: Each model generated transcripts for ten plastic surgery scenarios. Transcripts were independently evaluated by three clinically trained raters using a seven-criterion rubric: Medical Accuracy, Realism, Persona Consistency, Fidelity, Empathy, Relevancy, and Usability. Raters were blinded to the model identity to reduce bias. Each was rated on a 5-point Likert scale, yielding 840 total evaluations. Descriptive statistics were computed, and a two-way repeated measures ANOVA was used to test for differences across models and metrics. In addition, transcripts were analyzed using automated linguistic and content-based metrics. Results: All models achieved strong performance, with mean ratings exceeding 4.5 across all criteria. Gemini 2.5 Pro received mean scores (5.00 ± 0.00) in Medical Accuracy, Realism, Persona Consistency, Relevancy, and Usability. Claude 3.7 Sonnet matched the scores in Persona Consistency and Relevancy and led in Empathy (4.96 ± 0.18). ChatGPT 4.5 also achieved perfect scores in Relevancy, with high scores in Empathy (4.93 ± 0.25) and Usability (4.96 ± 0.18). ChatGPT 4o demonstrated consistently strong but slightly lower performance across most dimensions. ANOVA revealed no statistically significant differences across models (F(3, 6) = 0.85, p = 0.52). Automated analysis showed substantial variation in transcript length, style, and content richness: Gemini 2.5 Pro generated the longest and most emotionally expressive dialogues, while ChatGPT 4o produced the shortest and most concise outputs. Conclusions: Leading LLMs can generate medically accurate, emotionally appropriate synthetic dialogues suitable for educational and research use. Despite high performance, demographic homogeneity in generated patients highlights the need for improved diversity and bias mitigation in model outputs. These findings support the cautious, context-aware integration of LLM-generated dialogues into medical training, simulation, and research.

Deep brain stimulation (DBS) is an established therapeutic intervention for a range of neurological disorders, including Parkinson’s disease, essential tremor, and epilepsy. The efficacy of DBS is predicated on the modulation of dysfunctional neural circuits through the application of electrical pulses. Recent advancements in MRI technology have facilitated the visualization of brain nuclei and their associated networks, which is critical for proper patient-specific targeting of electrodes in surgical procedures. This advancement aims to personalize surgical planning and postoperative programming by integrating patient-specific anatomical and connectivity models. The advent of clinical ultrahigh-field MRI, particularly 7T MRI, significantly enhances the targeting of DBS by improving spatial resolution, signal-to-noise ratio, and tissue contrast. Compared with lower field strengths, 7T MRI provides superior visualization of deeply situated brain nuclei and their extensive cortical projections, including the subthalamic nucleus, globus pallidus internus, and thalamus. This review synthesizes a multi-institutional consensus regarding the technical and clinical applications of 7T MRI, drawing on experience from more than 1000 procedures using 7T DBS. It encompasses both imaging acquisition and postprocessing techniques aimed at optimizing image quality. Our extensive clinical experience informs best practices for correcting distortions, mitigating image artifacts, and employing specific imaging sequences that enhance the visualization of common DBS targets. The insights and recommendations presented are intended to promote the safe and effective utilization of 7T MRI in DBS.

Objective: Our project aims to use Artificial Intelligent (AI)-based methods to address overlooked issues in adolescent nutrition/diet and microbiome, and to identify biomarkers and risk factors that affect long-term health outcomes. Hypothesis: Many diseases, including diabetes, dementia, and Alzheimer’s disease, are linked to microbiome dysbiosis and diet-related risk factors, which can develop during adolescence and influence health outcomes throughout adulthood. Methods and Data: We conducted a systematic literature review on observational, clinical, and laboratory data to examine the relationships among diet quality, nutrition component, microbiome profiles, and energy expenditure, and their effects on behavior and health outcomes/diseases across adolescence, adulthood, and elderly populations (65+ yrs). We designed an analysis workflow to systematically extract and categorize data. Using AI-based methods we constructed knowledge models to uncover the relationships among key concepts on nutrition, diet, microbiome, genetic factors, age, sex, socioeconomic factors, and disease outcomes. Using the citations from PubMed literature, the Dietary Guidelines for Americans, and recommendations from authoritative health organizations, we developed a novel ChatGPT-based chatbot platform, Teen Fuel Up, designed to provide teens with real-time assistance to published knowledge about diet and nutrition. Conclusions: AI-based methods are highly effective in knowledge discovery and health promotion, identifying connections between adolescent nutrition, the microbiome, and chronic diseases later in life, as well as their differences using sex as a biological variable to increase scientific understanding of conditions that are specific to women. We will continue developing the ChatGPT-based chatbot to provide real-time cited knowledge that adapts to the evolving dietary needs of individuals. National Institute of Allergy And Infectious Diseases of the National Institutes of Health, grant number R01AI130460 and U24AI171008, and CPRIT award RP220244. This abstract was presented at the American Physiology Summit 2025 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.

The increasing integration of large language models (LLMs) into healthcare presents significant opportunities, but also critical risks related to patient safety, accuracy, and ethical alignment. Despite these concerns, no standardized framework exists for systematically evaluating and stress testing LLM behavior in clinical decision-making. The PIEE cycle—Planning and Preparation, Information Gathering and Prompt Generation, Execution, and Evaluation—is a structured red-teaming framework developed specifically to address artificial intelligence (AI) safety risks in healthcare decision-making. PIEE enables clinicians and informatics teams to simulate adversarial prompts, including jailbreaking, social engineering, and distractor attacks, to stress-test language models in real-world clinical scenarios. Model performance is evaluated using specific metrics such as true positive and false positive rates for detecting harmful content, hallucination rates measured through adapted TruthfulQA scoring, safety and reliability assessments, bias detection via adapted BBQ benchmarks, and ethical evaluation using structured Likert-based scoring rubrics. The framework is illustrated using examples from plastic surgery, but is adaptable across specialties, and is intended for use by all medical providers, regardless of their backgrounds or familiarity with artificial intelligence. While the framework is currently conceptual and validation is ongoing, PIEE provides a practical foundation for assessing the clinical reliability and ethical robustness of LLMs in medicine.

Triple-negative breast cancer (TNBC) represents the most malignant subtype of breast cancer. The clinical application of PARP inhibitors (PARPi) is limited by the low frequency of BRCA1/2 mutations in TNBC. Here, we identified that MTAP deletion sensitized genotoxic agents in our clinical cohort of metastatic TNBC. Further study demonstrated that MTAP deficiency or inhibition rendered TNBC susceptibility to chemotherapeutic agents, particularly PARPi. Mechanistically, targeting MTAP that synergized with PARPi by disrupting the METTL16-MAT2A axis involved in methionine metabolism and depleting in vivo s-adenosylmethionine (SAM) levels. Exhausted SAM in turn impaired PARPi-induced DNA damage repair through attenuation of MRE11 recruitment and end resection by diminishing MRE11 methylation. Notably, brain metastatic TNBC markedly benefited from a lower dose of PARPi and MTAP deficiency/inhibition synergy due to the inherently limited methionine environment in the brain. Collectively, our findings revealed a feed-forward loop between methionine metabolism and DNA repair through SAM, highlighting a therapeutic strategy of PARPi combined with MTAP deficiency/inhibition for TNBC.

(1) Burn injuries demand multidisciplinary, evidence-based care, yet the extensive literature complicates timely decision making. Retrieval-augmented generation (RAG) synthesizes research while addressing inaccuracies in pretrained models. However, citation bias in sourcing for RAG often prioritizes highly cited studies, overlooking less-cited but valuable research. This study examines RAG’s performance in burn management, comparing citation levels to enhance evidence synthesis, reduce selection bias, and guide decisions. (2) Two burn management datasets were assembled: 30 highly cited (mean: 303) and 30 less-cited (mean: 21). The Gemini-1.0-Pro-002 RAG model addressed 30 questions, ranging from foundational principles to advanced surgical approaches. Responses were evaluated for accuracy (5-point scale), readability (Flesch–Kincaid metrics), and response time with Wilcoxon rank sum tests (p < 0.05). (3) RAG achieved comparable accuracy (4.6 vs. 4.2, p = 0.49), readability (Flesch Reading Ease: 42.8 vs. 46.5, p = 0.26; Grade Level: 9.9 vs. 9.5, p = 0.29), and response time (2.8 vs. 2.5 s, p = 0.39) for the highly and less-cited datasets. (4) Less-cited research performed similarly to highly cited sources. This equivalence broadens clinicians’ access to novel, diverse insights without sacrificing quality. As plastic surgery evolves, RAG’s inclusive approach fosters innovation, improves patient care, and reduces cognitive burden by integrating underutilized studies. Embracing RAG could propel the field toward dynamic, forward-thinking care.

PURPOSE Tumor treating fields (TTFields) use alternating electric fields to disrupt cancer cell proliferation. Feasibility of TTFields therapy with gemcitabine/nab-paclitaxel was previously demonstrated in patients with advanced pancreatic adenocarcinoma. PANOVA-3 was designed to confirm safety and efficacy of TTFields in patients with unresectable locally advanced pancreatic adenocarcinoma (LA-PAC). METHODS In this global phase III trial, 571 patients with newly diagnosed LA-PAC were randomly assigned to receive gemcitabine 1,000 mg/m 2 and nab-paclitaxel 125 mg/m 2 by intravenous infusion once a day on days 1, 8, and 15 of a 28-day cycle with or without TTFields. The primary end point was overall survival (OS). Secondary end points included progression-free survival (PFS), local PFS, pain-free survival, and overall response rate (ORR). Distant PFS was analyzed post hoc. RESULTS OS was significantly prolonged using TTFields with gemcitabine/nab-paclitaxel versus gemcitabine/nab-paclitaxel (median, 16.2 months [95% CI, 15.0 to 18.0] v 14.2 months [95% CI, 12.8 to 15.4]; hazard ratio [HR], 0.82 [95% CI, 0.68 to 0.99]; P = .039). PFS, local PFS, and ORR were not improved. Pain-free survival was significantly prolonged with TTFields with gemcitabine/nab-paclitaxel (median, 15.2 months [95% CI, 10.3 to 22.8] v 9.1 months [95% CI, 7.4 to 12.7]; HR, 0.74 [95% CI, 0.56 to 0.97]; P = .027), as was distant PFS (median, 13.9 months [95% CI, 12.2 to 16.8] v 11.5 months [95% CI, 10.4 to 12.9]; HR, 0.74 [95% CI, 0.57 to 0.96]; P = .022). Device-related skin adverse events (AEs) were experienced by 76.3% of patients. Most device-related skin AEs were mild to moderate, with 7.7% of patients reporting a grade 3 AE. CONCLUSION This study demonstrated significant OS, pain-free survival, and distant PFS benefits for TTFields with gemcitabine/nab-paclitaxel versus gemcitabine/nab-paclitaxel in patients with unresectable LA-PAC, with no additive systemic toxicity.