UK Biobank
PulseAugur coverage of UK Biobank — every cluster mentioning UK Biobank across labs, papers, and developer communities, ranked by signal.
7 day(s) with sentiment data
-
New MICViT model enhances multimodal brain MRI analysis
Researchers have developed a new 3D vision transformer model called MICViT designed to improve the integration of multimodal brain MRI data. This model explicitly captures both modality-specific features and cross-modal…
-
New latent ODE model enhances heart failure prediction from cardiac MRI
Researchers have developed a novel latent dynamical model using neural ordinary differential equations (ODEs) to analyze cardiac magnetic resonance imaging (CMR) data. This model encodes bi-ventricular anatomy and full-…
-
AI framework REVEAL++ improves Alzheimer's risk prediction using retinal scans
Researchers have developed REVEAL++, a novel framework for predicting Alzheimer's disease risk using retinal imaging and clinical data. This new approach employs a differentiable phenotypic grouping method, allowing for…
-
Measurement noise limits nonlinear models in biomedical prediction, study finds
A new research paper argues that measurement noise, rather than model limitations, is the primary factor hindering the performance of nonlinear models in biomedical prediction tasks. The study suggests that additive noi…
-
New method detects ML data leakage from predictions alone
Researchers have developed a new method for detecting information leakage in machine learning models without requiring access to training data or code. The technique analyzes only the model's predictions and outcomes to…
-
New AI framework generates cardiac cine images from ECG and MRI
Researchers have developed a new framework called Chain of Flow (COF) that generates 4D cardiac cine images using electrocardiography (ECG) and patient-specific MRI data. This method aims to provide functional cardiac a…
-
CardioMorphNet predicts cardiac motion using shape-guided Bayesian deep learning
Researchers have developed CardioMorphNet, a novel Bayesian recurrent deep learning framework for predicting cardiac motion from short-axis cardiac MRI images. This method utilizes a recurrent variational autoencoder an…
-
Bayesian hypergraph inference models disease risk pathways
Researchers have developed a new Bayesian hypergraph inference framework to model complex relationships between diseases and risk factors using electronic health records. This approach moves beyond treating diseases ind…
-
New causal analysis method improves hip fracture risk prediction
Researchers have developed a causal analysis method to better understand the relationship between skeletal phenotypes derived from DXA scans and the risk of hip fractures. By analyzing data from over 21,000 UK Biobank p…
-
New research suggests model-level ensembling improves ML feature importance
A new research paper proposes a method to improve the accuracy of feature importance estimation in machine learning models, particularly for complex models used in critical applications like biomedical research. The stu…
-
New RoVTL Framework Tackles Missing Tabular Data in Multimodal Learning
Researchers have developed RoVTL, a novel framework designed to handle missing tabular data in multimodal learning, particularly for medical biobanks. The framework employs contrastive pretraining with simulated missing…
-
AI generates counterfactual DXA spine images using causal VAE
Researchers have developed a causal hierarchical variational autoencoder (CHVAE) to generate realistic spine DXA images. This model is trained on UK Biobank data and conditioned on participant attributes and lumbar morp…
-
New method optimizes data acquisition for multimodal AI under budget constraints
Researchers have introduced Cohort-based Active Modality Acquisition (CAMA), a new method for optimizing the acquisition of additional data modalities in multimodal machine learning under budget constraints. CAMA focuse…
-
Pan-FM model tackles missing organ data in medical imaging
Researchers have developed Pan-FM, a foundation model designed for medical imaging that can handle missing data across multiple organs. Unlike previous models trained on single organs, Pan-FM learns from seven different…
-
New method tackles missing covariate data in large-scale population studies
Researchers have developed a new augmented transfer regression learning method to address situations where crucial covariates are entirely missing in a target population, a common issue with large datasets like the UK B…
-
New framework integrates diverse data for genotype-phenotype prediction
Researchers have developed EFGPP, a new framework designed to improve genotype-phenotype prediction by integrating diverse data sources. The system was tested on migraine prediction using data from 733 UK Biobank indivi…
-
Deep learning predicts Alzheimer's risk factors from retinal images
Researchers have developed deep learning models capable of predicting 12 Alzheimer's disease risk factors from retinal images. These models, trained on over 62,000 images from the UK Biobank, analyzed retinal structures…
-
China penalizes AI platforms, MoonPay buys Sodot, and AI aids antibiotic resistance
China has imposed penalties on AI platforms for failing to properly label their content, indicating a regulatory move towards transparency in the AI sector. Separately, the Moto Buds 2 Plus have been released, incorpora…
-
EU AI Act faces potential delay amid privacy and civil society concerns
The Italian Data Protection Authority has fined Poste/Postepay €12.5 million for tracking pixel violations. Additionally, the AI Digital Omnibus suggests a potential postponement of the AI Act. A civil society letter, s…
-
UK Biobank data of 500k volunteers listed for sale on Alibaba
Medical data belonging to approximately 500,000 volunteers from the UK Biobank was found listed for sale on the Chinese e-commerce platform Alibaba. While the data was reportedly anonymized, UK technology minister Ian M…