Dataset Information

Capturing heterogeneous group differences using mixture-of-experts: Application to a study of aging.

ABSTRACT: In MRI studies, linear multi-variate methods are often employed to identify regions or connections that are affected due to disease or normal aging. Such linear models inherently assume that there is a single, homogeneous abnormality pattern that is present in all affected individuals. While kernel-based methods can implicitly model a non-linear effect, and therefore the heterogeneity in the affected group, extracting and interpreting information about affected regions is difficult. In this paper, we present a method that explicitly models and captures heterogeneous patterns of change in the affected group relative to a reference group of controls. For this purpose, we use the Mixture-of-Experts (MOE) framework, which combines unsupervised modeling of mixtures of distributions with supervised learning of classifiers. MOE approximates the non-linear boundary between the two groups with a piece-wise linear boundary, thus allowing discovery of multiple patterns of group differences. In the case of patient/control comparisons, each such pattern aims to capture a different dimension of a disease, and hence to identify patient subgroups. We validated our model using multiple simulation scenarios and performance measures. We applied this method to resting state functional MRI data from the Baltimore Longitudinal Study of Aging, to investigate heterogeneous effects of aging on brain function in cognitively normal older adults (>85years) relative to a reference group of normal young to middle-aged adults (<60years). We found strong evidence for the presence of two subgroups of older adults, with similar age distributions in each subgroup, but different connectivity patterns associated with aging. While both older subgroups showed reduced functional connectivity in the Default Mode Network (DMN), increases in functional connectivity within the pre-frontal cortex as well as the bilateral insula were observed only for one of the two subgroups. Interestingly, the subgroup showing this increased connectivity (unlike the other subgroup) was, cognitively similar at baseline to the young and middle-aged subjects in two of seven cognitive domains, and had a faster rate of cognitive decline in one of seven domains. These results suggest that older individuals whose baseline cognitive performance is comparable to that of younger individuals recruit their "cognitive reserve" later in life, to compensate for reduced connectivity in other brain regions.

SUBMITTER: Eavani H

PROVIDER: S-EPMC5460911 | biostudies-literature | 2016 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Capturing heterogeneous group differences using mixture-of-experts: Application to a study of aging.

Eavani Harini H Hsieh Meng Kang MK An Yang Y Erus Guray G Beason-Held Lori L Resnick Susan S Davatzikos Christos C

NeuroImage 20151023

In MRI studies, linear multi-variate methods are often employed to identify regions or connections that are affected due to disease or normal aging. Such linear models inherently assume that there is a single, homogeneous abnormality pattern that is present in all affected individuals. While kernel-based methods can implicitly model a non-linear effect, and therefore the heterogeneity in the affected group, extracting and interpreting information about affected regions is difficult. In this pape ...[more]

PMID: 26525656

Similar Datasets

Project description:Introduction: Research incorporating resilience, a concept featuring a positive outcome despite some type of stressor, has the potential to identify possibilities for promotion of the well-being of older people. This study aims to gain insight into the value and potential applications of resilience in both research and care practice from the perspective of researchers and care professionals. Specifically, the value of two scientific approaches, the a priori (i.e., based on a priori definition of a stressor and outcome) and dynamical systems approaches (i.e., based on mathematically modeled patterns in the real-time response to perturbations), was explored. Methods: Focus groups were performed to explore the thoughts of academic researchers from different disciplines in the fields of aging and care and care professionals on the application of the concept of resilience, including the a priori and dynamical systems approaches. Analysis of these focus groups was based on the framework method. Results: Five focus groups were held with a total of nine researchers from different disciplines (e.g., epidemiology, sociology) and 15 older adult care professionals from different professions (e.g., elderly care physician, physiotherapist). The participants described resilience as a concept with value for both aging research and care through its positive connotation and comprehensiveness. Continued research was thought to play an important role in clearing up some of the existing ambiguity surrounding resilience. The importance of resilience in the context of both high- and low-intensity stressors was underscored. The a priori and dynamical systems approaches were considered to have their specific advantages and disadvantages on both conceptual and feasibility levels. Therefore, the use of both approaches, side by side and in combination, was suggested. Conclusion: This qualitative exploration among researchers and care professionals confirms that the concept of resilience, including the a priori and dynamical systems approaches, is valuable. However, more work is necessary before can be delivered on the potential of resilience in aging research and older adult care practice. Greater conceptual and operational clarity can be achieved through more qualitative studies on the concept that take the perspective of older people into account and through empirical studies that work with both approaches simultaneously and/or in combination.

Dataset Information

Capturing heterogeneous group differences using mixture-of-experts: Application to a study of aging.

Publications

Capturing heterogeneous group differences using mixture-of-experts: Application to a study of aging.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets