Skip to results
Modify your search
NARROW
1-20 of 3009
Availability: Open Access
Image
The screenshot of a patient message indicating a potential medical emergenc...
Published: 12 April 2025
Figure 1.
The screenshot of a patient message indicating a potential medical emergency. The form includes a warning to call 911 or visit the emergency department for emergencies, followed by text fields for “Enter a subject...” (filled in with “Headache”) and “Enter your message...” (containing a detailed not
Image
Study overview. A flowchart showing a study overview where a knowledge grap...
Published: 12 April 2025
Figure 2.
Study overview. A flowchart showing a study overview where a knowledge graph, developed using a GPT-4–powered nurse triage book, feeds into three RAG methods (naive, local, global) and a prompt‐only approach, with all outputs evaluated via statistical tests on key performance metrics. (RAG: retrieva
Journal Article
Revisions to the Safety Assurance Factors for Electronic Health Record Resilience (SAFER) Guides to update national recommendations for safe use of electronic health records
Journal of the American Medical Informatics Association, Volume 32, Issue 4, April 2025, Pages 755–760, https://doi.org/10.1093/jamia/ocaf018
Published: 12 April 2025
Journal Article
Detecting emergencies in patient portal messages using large language models and knowledge graph-based retrieval-augmented generation
Journal of the American Medical Informatics Association, ocaf059, https://doi.org/10.1093/jamia/ocaf059
Published: 12 April 2025
Image
Illustrative examples of ATLAS output in 2 patients with COPD who reported ...
Published: 27 March 2025
Figure 1.
Illustrative examples of ATLAS output in 2 patients with COPD who reported the same peak leg effort and dyspnea scores (5/10 according to the Borg score). Patient #1 (A) is a 58-year-old man with mild-to-moderate COPD-based on resting functional data showing a peak work rate of 115 W. In this case,
Image
The severity of exertional dyspnea and leg effort as measured by our algori...
Published: 27 March 2025
Figure 2.
The severity of exertional dyspnea and leg effort as measured by our algorithm throughout the exercise (ATLAS) vs that indicated by the isolated symptom assessment at peak exercise in patients with COPD. A shows that a large fraction of patients in whom leg effort was the dominant symptom had severe
Image
Overview of our model architecture based on PL-Marker architecture. 40...
Published: 27 March 2025
Figure 2.
Overview of our model architecture based on PL-Marker architecture. 40 We take a 3-step pipeline approach: (1) entity and trigger recognition (named entity recognition—NER), (2) relation extraction (RE), and (3) factuality detection. Input to the model are individual sentences of the document. Th
Image
Label distribution of entity instances in DiMB-RE. A horizontal bar char...
Published: 27 March 2025
Figure 3.
Label distribution of entity instances in DiMB-RE. A horizontal bar chart showing the distribution of the entity labels in the DiMB-RE corpus. Frequency counts are included.
Journal Article
AI-Techniques Loss-Based Algorithm for Severity Classification (ATLAS): a novel approach for continuous quantification of exertional symptoms during incremental exercise testing
Journal of the American Medical Informatics Association, ocaf051, https://doi.org/10.1093/jamia/ocaf051
Published: 27 March 2025
Image
Categories of leg effort (left) and dyspnea (right) severity based on peak ...
Published: 27 March 2025
Figure 3.
Categories of leg effort (left) and dyspnea (right) severity based on peak scores and ATLAS vs markers of severe resting functional impairment (A), exercise functional impairment (B), and disablement (C) in patients with COPD. Regardless of the metrics, ATLAS categories—but not peak scores categorie
Image
Class-level trigger and relation type distributions in DiMB-RE. A horizo...
Published: 27 March 2025
Figure 4.
Class-level trigger and relation type distributions in DiMB-RE. A horizontal bar chart showing the distribution of class-level trigger and relation type distributions in the DiMB-RE corpus. For each relation type, number of corresponding triggers and relation instances are shown.
Image
Distribution of certainty labels in DiMB-RE. A horizontal bar chart show...
Published: 27 March 2025
Figure 5.
Distribution of certainty labels in DiMB-RE. A horizontal bar chart showing the distribution of factuality types in the DiMB-RE corpus.
Journal Article
DiMB-RE: mining the scientific literature for diet-microbiome associations
Journal of the American Medical Informatics Association, ocaf054, https://doi.org/10.1093/jamia/ocaf054
Published: 27 March 2025
Image
Annotation of a sentence from a PubMed abstract (PMID: 33509667). Entity me...
Published: 27 March 2025
Figure 1.
Annotation of a sentence from a PubMed abstract (PMID: 33509667). Entity mentions are shown in colored boxes and triggers and the corresponding relations are highlighted in brown. Relations are assigned certainty levels. Abbreviations: F, Factual; N, Negated. A diagram showing the annotation of e
Journal Article
High-performance automated abstract screening with large language model ensembles
Journal of the American Medical Informatics Association, ocaf050, https://doi.org/10.1093/jamia/ocaf050
Published: 22 March 2025
Image
Precision (positive predictive value) and recall (sensitivity) of 6 LLMs ta...
Published: 22 March 2025
Figure 1.
Precision (positive predictive value) and recall (sensitivity) of 6 LLMs tasked with automated abstract screening on the development dataset (n = 800), across a range of 6 prompts with varying bias towards inclusion. Sensitivity was deemed 100% for all models working with Sulistyo et al, 36 as th
Image
(A) Schematics describing 6 distinct configurations for incorporation of LL...
Published: 22 March 2025
Figure 3.
(A) Schematics describing 6 distinct configurations for incorporation of LLM and human decisions into a binary ensemble system. (B) Precision (positive predictive value) and recall (sensitivity) of every ensemble permutation combining LLMs, prompts, and human researchers in each configuration; teste
Image
Correlational analysis, undertaken on results obtained from the development...
Published: 22 March 2025
Figure 2.
Correlational analysis, undertaken on results obtained from the development dataset, to investigate whether review-centric factors influenced the abstract screening performance of LLMs and human researchers replicating the work of Cochrane systematic review authors. Between 1.2% and 29.9% of the var
Image
PRISMA flow diagram of retrieved, screened, and included review articles. ...
Published: 20 March 2025
Figure 1.
PRISMA flow diagram of retrieved, screened, and included review articles. Diagram showing the number of records that were identified, screened, included, and synthesized in this review, as well as the records that were removed or excluded from this process.
Journal Article
Equity implications of extended reality technologies for health and procedural anxiety: a systematic review and implementation-focused framework
Journal of the American Medical Informatics Association, ocaf047, https://doi.org/10.1093/jamia/ocaf047
Published: 20 March 2025
