Results 211 to 220 of about 155,699 (303)
This study uses gut microbiome data to predict corticosteroid response in patients with immune thrombocytopenia (ITP). Fecal samples from 212 patients with ITP are sequenced. Six machine‐learning algorithms are used to train predictive models. The support vector machine‐based model integrated clinical data and selected microbial species, diversities ...
Feng‐Qi Liu+38 more
wiley +1 more source
Twin neural network regression
We propose to reformulate a regression problem into predicting differences between target values. This allows for leveraging consistency conditions which can be used as uncertainty estimates and enable the production of an ensemble of predictions while training only a single neural network.
Sebastian Johann Wetzel+3 more
wiley +1 more source
Automated Scoring of Alzheimer's Disease Atrophy Scale with Subtype Classification Using Deep Learning-Based T1-Weighted Magnetic Resonance Image Segmentation. [PDF]
Choe YS+13 more
europepmc +1 more source
Deep learning enables automated scoring of liver fibrosis stages. [PDF]
Yu Y+13 more
europepmc +1 more source
Automating Lead Scoring with Machine Learning: An Experimental Study [PDF]
Robert Nygård, József Mezei
openalex +1 more source
Ca2+/calmodulin‐dependent kinase II (CaMKII) activation is likely driven by oxidative stress, particularly excessive reactive oxygen species (ROS) production. This study identifies P21‐activated kinase 2 (Pak2) as a novel regulator of ROS‐induced CaMKII activation and abnormal Ca2⁺ dynamics in acute adrenergic and chronic pressure‐overload stressed ...
Tao Li+21 more
wiley +1 more source
How can we fully understand brain function and disease without uncovering the causal coordination between molecular and neural activity across space and time? MEA‐seqX tackles this challenge by integrating advanced electrophysiology, spatial transcriptomics, and artificial intelligence predictions, revealing multiscale dynamics.
Brett Addison Emery+9 more
wiley +1 more source
An automated machine learning framework integrating environmental and genomic data enhances genetic analysis and genomic prediction in maize. By leveraging dimension‐reduced environmental parameters, it reveals trait‐environment relationships and identifies genetic markers that govern phenotypic plasticity and genotype‐by‐environment interactions.
Kunhui He+12 more
wiley +1 more source
Validation of automated scoring of science assessments
Ou Lydia Liu+4 more
openalex +2 more sources