Dataset Information

Associated factors of white matter hyperintensity volume: a machine-learning approach.

ABSTRACT: To identify the most important parameters associated with cerebral white matter hyperintensities (WMH), in consideration of potential collinearity, we used a data-driven machine-learning approach. We analysed two independent cohorts (KORA and SHIP). WMH volumes were derived from cMRI-images (FLAIR). 90 (KORA) and 34 (SHIP) potential determinants of WMH including measures of diabetes, blood-pressure, medication-intake, sociodemographics, life-style factors, somatic/depressive-symptoms and sleep were collected. Elastic net regression was used to identify relevant predictor covariates associated with WMH volume. The ten most frequently selected variables in KORA were subsequently examined for robustness in SHIP. The final KORA sample consisted of 370 participants (58% male; age 55.7?±?9.1 years), the SHIP sample comprised 854 participants (38% male; age 53.9?±?9.3 years). The most often selected and highly replicable parameters associated with WMH volume were in descending order age, hypertension, components of the social environment (i.e. widowed, living alone) and prediabetes. A systematic machine-learning based analysis of two independent, population-based cohorts showed, that besides age and hypertension, prediabetes and components of the social environment might play important roles in the development of WMH. Our results enable personal risk assessment for the development of WMH and inform prevention strategies tailored to the individual patient.

SUBMITTER: Grosu S

PROVIDER: S-EPMC7840689 | biostudies-literature | 2021 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Associated factors of white matter hyperintensity volume: a machine-learning approach.

Grosu Sergio S Rospleszcz Susanne S Hartmann Felix F Habes Mohamad M Bamberg Fabian F Schlett Christopher L CL Galie Franziska F Lorbeer Roberto R Auweter Sigrid S Selder Sonja S Buelow Robin R Heier Margit M Rathmann Wolfgang W Mueller-Peltzer Katharina K Ladwig Karl-Heinz KH Grabe Hans J HJ Peters Annette A Ertl-Wagner Birgit B BB Stoecklein Sophia S

Scientific reports 20210127 1

To identify the most important parameters associated with cerebral white matter hyperintensities (WMH), in consideration of potential collinearity, we used a data-driven machine-learning approach. We analysed two independent cohorts (KORA and SHIP). WMH volumes were derived from cMRI-images (FLAIR). 90 (KORA) and 34 (SHIP) potential determinants of WMH including measures of diabetes, blood-pressure, medication-intake, sociodemographics, life-style factors, somatic/depressive-symptoms and sleep w ...[more]

PMID: 33504924

Similar Datasets

Project description:White matter hyperintensities of presumed vascular origin (WMH) are a prevalent form of cerebral small-vessel disease and an important risk factor for post-stroke cognitive dysfunction. Despite this prevalence, it is not well understood how WMH contributes to post-stroke cognitive dysfunction. Preliminary findings suggest that increasing WMH volume is associated with total hippocampal volume in chronic stroke patients. The hippocampus, however, is a complex structure with distinct subfields that have varying roles in the function of the hippocampal circuitry and unique anatomical projections to different brain regions. For these reasons, an investigation into the relationship between WMH and hippocampal subfield volume may further delineate how WMH predispose to post-stroke cognitive dysfunction. In a prospective study of acute ischemic stroke patients with moderate/severe WMH burden, we assessed the relationship between quantitative WMH burden and hippocampal subfield volumes. Patients underwent a 3T MRI brain within 2-5 days of stroke onset. Total WMH volume was calculated in a semi-automated manner. Mean cortical thickness and hippocampal volumes were measured in the contralesional hemisphere. Total and subfield hippocampal volumes were measured using an automated, high-resolution, ex vivo computational atlas. Linear regression analyses were performed for predictors of total and subfield hippocampal volumes. Forty patients with acute ischemic stroke and moderate/severe white matter hyperintensity burden were included in this analysis. Median WMH volume was 9.0 cm3. Adjusting for intracranial volume and stroke laterality, age (β = -3.7, P < 0.001), hypertension (β = -44.7, P = 0.04), WMH volume (β = -0.89, P = 0.049), and mean cortical thickness (β = 286.2, P = 0.006) were associated with total hippocampal volume. In multivariable analysis, age (β = -3.3, P < 0.001) and cortical thickness (β = 205.2, P = 0.028) remained independently associated with total hippocampal volume. In linear regression for predictors of hippocampal subfield volume, increasing WMH volume was associated with decreased hippocampal-amygdala transition area volume (β = -0.04, P = 0.001). These finding suggest that in ischemic stroke patients, increased WMH burden is associated with selective hippocampal subfield degeneration in the hippocampal-amygdala transition area.

Dataset Information

Associated factors of white matter hyperintensity volume: a machine-learning approach.

Publications

Associated factors of white matter hyperintensity volume: a machine-learning approach.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets