All traces

Deep learning models for ultrasound beamforming can improve image resolution and quality over traditional DAS methods and enable real-time processing when paired with FPGA acceleration.

Imaging AI models achieved high pooled diagnostic accuracy for detecting avascular necrosis of the femoral head in research datasets.

Applying network analysis and process mining to 52,412 YouTube comments distinguished scientific from pseudoscientific health discussions, showing mixed-valence evaluation in scientific threads versus affirmation bonding in pseudoscientific threads.

Scoping review of 35 studies found AI and digital tools applied to predictive models, imaging analysis, and continuous monitoring showed potential for early diagnosis and prediction of neonatal respiratory outcomes.

In a controlled test of 50 abstracts with fictional author identities, ChatGPT and Claude showed no consistent gender or geographic bias and achieved high scoring reproducibility.

Renewing an alliance between the church and artists could increase institutional support and public defense for human-created art as AI-generated art expands.

An ensemble machine learning pipeline detected flowers on herbarium specimens with relatively strong accuracy and labeled 11.1 million of 22 million records, expanding taxonomic and temporal coverage when integrated into Phenobase.

AI-integrated intraoral mobile photographs achieved 90% pooled sensitivity and 89% specificity for noninvasive early detection of oral potentially malignant disorders, offering an accessible screening alternative in low-resource settings.

AI-based algorithms improve detection of scaphoid fractures including occult fractures on plain radiographs to support treatment decisions, while navigation and robotics improve screw accuracy and reduce radiation and guidewire attempts.

AI-assisted conavigation can augment digital health navigators with lived experience by handling triage, documentation, and engagement monitoring in mental health services.

Integration of generative AI into library systems is creating new possibilities for improving how information is accessed, managed, and disseminated.

AI-assisted predictive microbiome design helps identify and deploy beneficial microbial inoculants that lower pest and disease burden and enhance productivity in cereal, legume, and horticulture crops.

AI-based tools including chatbots and clinical decision-support systems were associated with improved accessibility of mental health support and more efficient risk assessment in psychiatric mental health nursing contexts.

Deep learning models applied to H&E whole-slide images can reliably identify melanomas at risk of recurrence and progression, improving diagnosis and personalized risk stratification for stage I-III cutaneous melanoma.

AI-enabled tools may support pharmaceutical functions including target identification, trial design, regulatory intelligence, evidence synthesis, and safety signal detection, expanding reach and efficiency.