Dataset Information

Making use of longitudinal information in pattern recognition.

ABSTRACT: Longitudinal designs are widely used in medical studies as a means of observing within-subject changes over time in groups of subjects, thereby aiming to improve sensitivity for detecting disease effects. Paralleling an increased use of such studies in neuroimaging has been the adoption of pattern recognition algorithms for making individualized predictions of disease. However, at present few pattern recognition methods exist to make full use of neuroimaging data that have been collected longitudinally, with most methods relying instead on cross-sectional style analysis. This article presents a principal component analysis-based feature construction method that uses longitudinal high-dimensional data to improve predictive performance of pattern recognition algorithms. The method can be applied to data from a wide range of longitudinal study designs and permits an arbitrary number of time-points per subject. We apply the method to two longitudinal datasets, one containing subjects with mild cognitive impairment along with healthy controls, the other with early dementia subjects and healthy controls. Across both datasets, we show improvements in predictive accuracy relative to cross-sectional classifiers for discriminating disease subjects from healthy controls on the basis of whole-brain structural magnetic resonance image-based voxels. In addition, we can transfer longitudinal information from one set of subjects to make disease predictions in another set of subjects. The proposed method is simple and, as a feature construction method, flexible with respect to the choice of classifier and image registration algorithm. Hum Brain Mapp 37:4385-4404, 2016. © 2016 Wiley Periodicals, Inc.

SUBMITTER: Aksman LM

PROVIDER: S-EPMC5111621 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundThe purpose of this study was to develop and validate the Policymakers' Information Use Questionnaire (POLIQ) to capture the intention of individuals in decision-making positions, such as health policy-makers, to act on research-based evidence in order to inform theory and the application of behaviour change models to decision-making spheres.MethodsThe development and validation comprised three steps: item generation, qualitative face validation with cognitive debriefing and factorial construct validation. Confirmatory factor analysis was applied to estimate item-domain correlations for five predefined constructs relating to content, beliefs, behaviour, control and intent. Cronbach's alpha coefficient was calculated to assess the overall consistency of questionnaire items with the predefined constructs. Participants in the item generation and face validation were health and policy researchers and two former decision-makers (former assistant deputy ministers) from the Canadian provincial level. Participants in the construct validation were 39 Canadian decision-makers at various positions of municipal, provincial and federal jurisdiction who participated in a series of policy dialogues focused on childhood disability.ResultsCognitive debriefing allowed for small adjustments in language for clarity, including simultaneous validation of the English and French questionnaires. Participants found that the questions were clear and addressed the domains being targeted. Internal consistency of items belonging to the respective questionnaire domains was moderate to high, with estimated Cronbach's alpha values ranging from 0.67 to 0.84. Estimated item-domain correlations indicated moderate to high measurement performance for the domains norm, control and beliefs, whereas weak to moderate correlations resulted for the constructs content and intent. Estimated imprecision of factor loadings (95% confidence interval widths) was considerable for the questionnaire domains content and intent.ConclusionMeasuring decision-makers' behaviour in relation to research evidence use is challenging. We provide initial evidence on face validity and appropriate measurement properties of the POLIQ based on a convenience sample of decision-makers in social and health policy. Larger validation studies and further psychometric property testing will support further utility of the POLIQ.

Dataset Information

Making use of longitudinal information in pattern recognition.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets