Background: Patient experience is a critical consideration for any health care institution. Leveraging artificial intelligence (AI) to improve health care delivery has rapidly become an institutional priority across the United States. Ambient AI documentation systems such as Dragon Ambient eXperience (DAX) may influence patient perception of health care provider communication and overall experience.
Objective: The objective of this study was to assess the impact of the implementation of an ambient AI documentation system (DAX) on Press Ganey (PG) patient experience scores.
Methods: A retrospective study was conducted to evaluate the relationship between provider use of DAX (N=49) and PG patient satisfaction scores from January 2023 to December 2024. Three domains were analyzed: (1) overall assessment of the experience, (2) concern the care provider showed for patients' questions or worries, and (3) likelihood of recommending the care provider to others. Mean pretest-posttest score differences and P values were calculated.
Results: A total of 49 health care providers across 9 departments participated in the DAX pilot. Aggregate scores for individual items increased between 0.9 and 1.9 points. Care provider concern for a patient's questions or worries increased the most (1.9 points; P=.01), followed by overall assessment of the experience (1.3 points; P=.09) and likelihood of recommending the provider (0.9 points; P=.33). Subgroup analysis showed a larger increase in concern scores among providers using DAX <50% of the time (3.2-point increase; P=.03).
Conclusions: This pilot study aimed to investigate the relationship between provider use of DAX and PG patient experience scores in the outpatient setting at a large academic medical center. Increases in PG scores after implementing DAX were observed across all PG items assessed. As technology and AI continue to improve and become more widespread, these results are encouraging. Health care providers may consider leveraging AI note-taking software as a way to enhance their communication and interactions with patients.
Unlabelled: Artificial intelligence (AI) is increasingly used to support medical interpreting and public health communication, yet current systems introduce serious risks to accuracy, confidentiality, and equity, particularly for speakers of low-resource languages. Automatic translation models often struggle with regional varieties, figurative language, culturally embedded meanings, and emotionally sensitive conversations about reproductive health or chronic disease, which can lead to clinically significant misunderstandings. These limitations threaten patient safety, informed consent, and trust in health systems when clinicians rely on AI as if it were a professional interpreter. At the same time, the large data sets required to train and maintain these systems create new concerns about surveillance, secondary use of linguistic data, and gaps in existing privacy protections. This viewpoint examines the ethical and structural implications of AI-mediated interpreting in clinical and public health settings, arguing that its routine use as a replacement for qualified interpreters would normalize a lower standard of care for people with Non-English Language Preference and reinforce existing health disparities. Instead, AI tools should be treated as optional, carefully evaluated supplements that operate under the supervision of trained clinicians and professional interpreters, within clear regulatory guardrails for transparency, accountability, and community oversight. The paper concludes that language access must remain grounded in human expertise, language rights, and structural commitments to equity, rather than in cost-saving promises of automated systems.
Background: Early-stage clinical findings often appear only as conference posters circulated on social media. Because posters rarely carry structured metadata, their citations are invisible to bibliometric and alternative metric tools, limiting real-time research discovery.
Objective: This study aimed to determine whether a large language model can accurately extract citation data from clinical conference poster images shared on X (formerly known as Twitter) and link those data to the Dimensions and Altmetric databases.
Methods: Poster images associated with the 2024 American Society of Clinical Oncology conference were searched using the terms "#ASCO24," "#ASCO2024," and the conference name. Images ≥100 kB that contained the word "poster" in the post text were retained. A prompt-engineered Gemini 2.0 Flash model classified images, summarized posters, and extracted structured citation elements (eg, authors, titles, and digital object identifiers [DOIs]) in JSON format. A hierarchical linkage algorithm matched extracted elements against Dimensions records, prioritizing persistent identifiers and then title-journal-author composites. Manual validation was performed on a random 20% sample.
Results: We searched within 115,714 posts and 16,574 images, of which 651 (3.9%) met the inclusion criteria, and we obtained 1117 potential citations. The algorithm linked 63.4% (708/1117) of the citations to 616 unique research outputs (n=580, 94.2% journal articles; n=36, 5.8% clinical trial registrations). Manual review of 135 randomly sampled citations confirmed correct linkage in 124 (91.9%) cases. DOI-based matching was mostly flawless; most errors occurred where only partial bibliographic details were available. The linked dataset enabled rapid profiling of topical foci (eg, lung and breast cancer) and identification of the most frequently referenced institutions and clinical trials in shared posters.
Conclusions: This study presents a novel artificial intelligence-driven methodology for enhancing research discovery and attention analysis from nontraditional clinical scholarly outputs. The American Society of Clinical Oncology was used as an example, but this methodology could be used for any conference and clinical poster.
Background: Artificial intelligence (AI) has, in the recent past, experienced a rebirth with the growth of generative AI systems such as ChatGPT and Bard. These systems are trained with billions of parameters and have enabled widespread accessibility and understanding of AI among different user groups. Widespread adoption of AI has led to the need for understanding how machine learning (ML) models operate to build trust in them. An understanding of how these models generate their results remains a huge challenge that explainable AI seeks to solve. Federated learning (FL) grew out of the need to have privacy-preserving AI by having ML models that are decentralized but still share model parameters with a global model.
Objective: This study sought to examine the extent of development of the explainable AI field within the FL environment in relation to the main contributions made, the types of FL, the sectors it is applied to, the models used, the methods applied by each study, and the databases from which sources are obtained.
Methods: A systematic search in 8 electronic databases, namely, Web of Science Core Collection, Scopus, PubMed, ACM Digital Library, IEEE Xplore, Mendeley, BASE, and Google Scholar, was undertaken.
Results: A review of 26 studies revealed that research on explainable FL is steadily growing despite being concentrated in Europe and Asia. The key determinants of FL use were data privacy and limited training data. Horizontal FL remains the preferred approach for federated ML, whereas post hoc explainability techniques were preferred.
Conclusions: There is potential for development of novel approaches and improvement of existing approaches in the explainable FL field, especially for critical areas.
Trial registration: OSF Registries 10.17605/OSF.IO/Y85WA; https://osf.io/y85wa.
Background: Artificial intelligence (AI) chatbots have become prominent tools in health care to enhance health knowledge and promote healthy behaviors across diverse populations. However, factors influencing the perception of AI chatbots and human-AI interaction are largely unknown.
Objective: This study aimed to identify interaction characteristics associated with the perception of an AI chatbot identity as a human versus an artificial agent, adjusting for sociodemographic status and previous chatbot use in a diverse sample of women.
Methods: This study was a secondary analysis of data from the HeartBot trial in women aged 25 years or older who were recruited through social media from October 2023 to January 2024. The original goal of the HeartBot trial was to evaluate the change in awareness and knowledge of heart attack after interacting with a fully automated AI HeartBot chatbot. All participants interacted with HeartBot once. At the beginning of the conversation, the chatbot introduced itself as HeartBot. However, it did not explicitly indicate that participants would be interacting with an AI system. The perceived chatbot identity (human vs artificial agent), conversation length with HeartBot, message humanness, message effectiveness, and attitude toward AI were measured at the postchatbot survey. Multivariable logistic regression was conducted to explore factors predicting women's perception of a chatbot's identity as a human, adjusting for age, race or ethnicity, education, previous AI chatbot use, message humanness, message effectiveness, and attitude toward AI.
Results: Among 92 women (mean age 45.9, SD 11.9; range 26-70 y), the chatbot identity was correctly identified by two-thirds (n=61, 66%) of the sample, while one-third (n=31, 34%) misidentified the chatbot as a human. Over half (n=53, 58%) had previous AI chatbot experience. On average, participants interacted with the HeartBot for 13.0 (SD 7.8) minutes and entered 82.5 (SD 61.9) words. In multivariable analysis, only message humanness was significantly associated with the perception of chatbot identity as a human compared with an artificial agent (adjusted odds ratio 2.37, 95% CI 1.26-4.48; P=.007).
Conclusions: To the best of our knowledge, this is the first study to explicitly ask participants whether they perceive an interaction as human or from a chatbot (HeartBot) in the health care field. This study's findings (role and importance of message humanness) provide new insights into designing chatbots. However, the current evidence remains preliminary. Future research is warranted to understand the relationship between chatbot identity, message humanness, and health outcomes in a larger-scale study.
Background: Artificial intelligence (AI) models are increasingly being used in medical education. Although models like ChatGPT have previously demonstrated strong performance on USMLE-style questions, newer AI tools with enhanced capabilities are now available, necessitating comparative evaluations of their accuracy and reliability across different medical domains and question formats.
Objective: To evaluate and compare the performance of five publicly available AI models: Grok, ChatGPT-4, Copilot, Gemini, and DeepSeek, on the USMLE Step 1 Free 120-question set, checking their accuracy and consistency across question types and medical subjects.
Methods: This cross-sectional observational study was conducted between February 10 and March 5, 2025. Each of the 119 USMLE-style questions (excluding one audio-based item) was presented to each AI model using a standardized prompt cycle. Models answered each question three times to assess confidence and consistency. Questions were categorized as text-based or image-based, and as case-based or information-based. Statistical analysis was done using Chi-square and Fisher's exact tests, with Bonferroni adjustment for pairwise comparisons.
Results: Grok got the highest score (91.6%), followed by Copilot (84.9%), Gemini (84.0%), ChatGPT-4 (79.8%), and DeepSeek (72.3%). DeepSeek's lower grade was due to an inability to process visual media, resulting in 0% accuracy on image-based items. When limited to text-only questions (n = 96), DeepSeek's accuracy increased to 89.6%, matching Copilot. Grok showed the highest accuracy on image-based (91.3%) and case-based questions (89.7%), with statistically significant differences observed between Grok and DeepSeek on case-based items (p = .011). The models performed best in Biostatistics & Epidemiology (96.7%) and worst in Musculoskeletal, Skin, & Connective Tissue (62.9%). Grok maintained 100% consistency in responses, while Copilot demonstrated the most self-correction (94.1% consistency), improving its accuracy to 89.9% on the third attempt.
Conclusions: AI models showed varying strengths across domains, with Grok demonstrating the highest accuracy and consistency in this dataset, particularly for image-based and reasoning-heavy questions. Although ChatGPT-4 remains widely used, newer models like Grok and Copilot also performed competitively. Continuous evaluation is essential as AI tools rapidly evolve.
Clinicaltrial:
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：