English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Machine learning (ML) has shown great promise for revolutionizing a number ofareas, including healthcare. However, it is also facing a reproducibilitycrisis, especially in medicine. ML models that are carefully constructed fromand evaluated on a training set might not generalize well on data fromdifferent patient populations or acquisition instrument settings and protocols.We tackle this problem in the context of neuroimaging of Alzheimer's disease(AD), schizophrenia (SZ) and brain aging. We develop a weighted empirical riskminimization approach that optimally combines data from a source group, e.g.,subjects are stratified by attributes such as sex, age group, race and clinicalcohort to make predictions on a target group, e.g., other sex, age group, etc.using a small fraction (10%) of data from the target group. We apply thismethod to multi-source data of 15,363 individuals from 20 neuroimaging studiesto build ML models for diagnosis of AD and SZ, and estimation of brain age. Wefound that this approach achieves substantially better accuracy than existingdomain adaptation techniques: it obtains area under curve greater than 0.95 forAD classification, area under curve greater than 0.7 for SZ classification andmean absolute error less than 5 years for brain age prediction on all targetgroups, achieving robustness to variations of scanners, protocols, anddemographic or clinical characteristics. In some cases, it is even better thantraining on all data from the target group, because it leverages the diversityand size of a larger training set. We also demonstrate the utility of ourmodels for prognostic tasks such as predicting disease progression inindividuals with mild cognitive impairment. Critically, our brain ageprediction models lead to new clinical insights regarding correlations withneurophysiological tests.

Adapting Machine Learning Diagnostic Models to New Populations Using a Small Amount of Data: Results from Clinical Neuroscience