Search Articles

View query in Help articles search

Search Results (1 to 10 of 3558 Results)

Download search results: CSV END BibTex RIS


Detecting Conversation Topics in Recruitment Calls of African American Participants to the All of Us Research Program Using Machine Learning: Model Development and Validation Study

Detecting Conversation Topics in Recruitment Calls of African American Participants to the All of Us Research Program Using Machine Learning: Model Development and Validation Study

Finally, using built-in functionality in the STM R package, we avoided subjective decisions on the number of topics by using a data-driven approach [20]. This approach helped eliminate human-based bias in our analysis by identifying words that appear in a document only if the document pertains to a specific topic.

Priscilla Pemu, Michael Prude, Atuarra McCaslin, Elizabeth Ojemakinde, Christopher Awad, Kelechi Igwe, Anny Rodriguez, Jasmine Foriest, Muhammed Idris

JMIR Form Res 2025;9:e65320

Human Guide Training to Improve Hospital Accessibility for Patients Who Are Blind: Needs Assessment and Pilot Process Evaluation

Human Guide Training to Improve Hospital Accessibility for Patients Who Are Blind: Needs Assessment and Pilot Process Evaluation

Data were analyzed in the R statistical environment using JASP (Jeffrey’s Amazing Statistics Program). We summarize the reach and knowledge questions using frequencies and percentages. We calculated means to describe behavioral capability and satisfaction. Pre- and posttest scores for behavioral capability were assessed for change using paired samples t tests with Cohen d effect sizes.

Tyler G James, Sarah Hughes, Christa Moran, Sherry Day, Michael M McKee

JMIR Rehabil Assist Technol 2025;12:e64666

Use of Large Language Models to Classify Epidemiological Characteristics in Synthetic and Real-World Social Media Posts About Conjunctivitis Outbreaks: Infodemiology Study

Use of Large Language Models to Classify Epidemiological Characteristics in Synthetic and Real-World Social Media Posts About Conjunctivitis Outbreaks: Infodemiology Study

Mean probabilistic assessments from GPT-4 in the published study correlated well with human raters (r=0.60, 95% CI 0.47-0.70) [18]. The mean of these elicited percentages from the posts about conjunctivitis correlated with both post volume and the occurrence of some known epidemics. We did not assess the ability of LLMs to assess the probability of epidemics based on content sourced from other forms of social media, such as web-based forums or blogs.

Michael S Deiner, Russell Y Deiner, Cherie Fathy, Natalie A Deiner, Vagelis Hristidis, Stephen D McLeod, Thomas J Bukowski, Thuy Doan, Gerami D Seitzman, Thomas M Lietman, Travis C Porco

J Med Internet Res 2025;27:e65226