98%
921
2 minutes
20
The recent progress in machine learning has shifted the trends in artificial intelligence (AI) toward an overreliance on increasing amounts of data, computing power, and model parameters. These trends have resulted in success, but have also created a monolithic perspective for AI, increased the barriers to entry outside of large tech companies, and raised concerns about computational sustainability. Neurosymbolic AI is a growing area that promotes methodological heterogeneity and aims to push the frontiers of AI through affordable data and computing power.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC12084822 | PMC |
http://dx.doi.org/10.1093/pnasnexus/pgaf117 | DOI Listing |
Hypertension
September 2025
Division of Cardiovascular Medicine, Hospital of the University of Pennsylvania, Philadelphia (C.B., H.T., J.A.C.).
Background: Aortic structural degeneration occurs with aging; however, 3-dimensional geometric remodeling has not been well characterized in large populations.
Methods: We segmented the thoracic aorta from magnetic resonance images of 56 164 UKB (UK Biobank) participants and computed tomography images of 9417 PMBB (Penn Medicine Biobank) participants. We quantified structural measurements of elongation, dilation, tortuosity, and curvature across the thoracic aorta.
Diagn Interv Radiol
September 2025
LMU University Hospital, LMU Munich, Department of Radiology, Munich, Germany.
Purpose: Computed tomography fluoroscopy (CTF)-guided biopsy is an established technique for sampling pulmonary lesions, particularly with the growing prevalence of lung nodule screening programs. This study investigated procedural and lesion-related factors affecting success and complication rates in routine CTF-guided lung core-needle biopsies at a tertiary center.
Methods: Consecutive patients undergoing percutaneous CTF-guided lung biopsies over a 10-year period (2007-2016) were retrospectively analyzed.
F1000Res
September 2025
Cambridge Centre for Proteomics, Department of Biochemistry, University of Cambridge, Cambridge, CB2 1QR, UK.
Background: Subcellular localisation is a determining factor of protein function. Mass spectrometry-based correlation profiling experiments facilitate the classification of protein subcellular localisation on a proteome-wide scale. In turn, static localisations can be compared across conditions to identify differential protein localisation events.
View Article and Find Full Text PDFNIHR Open Res
September 2025
Department of Neurology, North Bristol NHS Trust, Westbury on Trym, England, UK.
Background: This study aimed to explore the barriers and facilitators of implementing rehabilitation interventions for visual field loss due to stroke.
Methods: The study was a qualitative exploration using one-to-one interviews coded using template analysis and the COM-B a-priori framework. Participants were five occupational therapists from hospital (n=4) and community (n=1) National Health Service (NHS) stroke care settings in England.
Periodontol 2000
September 2025
Lineberger Comprehensive Cancer Center, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA.
Oral cancer is a major global health burden, ranking sixth in prevalence, with oral squamous cell carcinoma (OSCC) being the most common type. Importantly, OSCC is often diagnosed at late stages, underscoring the need for innovative methods for early detection. The oral microbiome, an active microbial community within the oral cavity, holds promise as a biomarker for the prediction and progression of cancer.
View Article and Find Full Text PDF