Search Results

1 - 2 of 2 items

  • Author: Henry Völzke x
Clear All Modify Search
Subpopulation Discovery in Epidemiological Data with Subspace Clustering


A prerequisite of personalized medicine is the identification of groups of people who share specific risk factors towards an outcome. We investigate the potential of subspace clustering for finding such groups in epidemiological data. We propose a workflow that encompasses clusterability assessment before cluster discovery and quality assessment after learning the clusters. Epidemiological usually do not have a ground truth for the verification of clusters found in subspaces. Hence, we introduce quality assessment through juxtaposition of the learned models to “models-of-randomness”, i.e. models that do not reflect a true cluster structure. On the basis of this workflow, we select subspace clustering methods, compare and discuss their performance. We use a dataset with hepatic steatosis as outcome, but our findings apply on arbitrary epidemiological cohort data that have tenths of variables and exhibit class skew.

Open access
The Impact of Childhood Trauma and Depressive Symptoms on Body Mass Index



Obesity is associated with several somatic diseases and increased psychological burden. This study focused on two potential psychological predictors of the body mass index (BMI), childhood trauma and depressive symptoms.


We used three independent populations: two general population samples (Study of Health in Pomerania, SHIP-2, N = 1,657; SHIP-TREND-0, N = 3,278) and one patient sample (GANI_MED, N = 1,742). Childhood trauma was measured with the childhood trauma questionnaire (CTQ) and depression with the Beck Depression Inventory (BDI-II) in SHIP-2 and the Patient Health Questionnaire (PHQ-9) in SHIP-TREND-0 and GANI_MED. We investigated the impact of childhood trauma and depression on BMI. Furthermore, we used mediation analysis to assess whether depression was a significant mediator on the path from childhood trauma to adult BMI in each of the three samples.


In all the three populations, depressive symptoms exhibited a significant association towards higher BMI (p < 0.05). Childhood trauma was positively associated with BMI with significant associations in SHIP-TREND-0 (p < 0.001) and GANI_MED (p = 0.005). The relationship between CTQ and BMI was significantly partially mediated (p < 0.05) by depressive symptoms in SHIP-TREND-0 (38.0%) and GANI_MED (16.4%), in SHIP-2 results pointed in the same direction. All the trauma sub-dimensions, except sexual abuse, exhibited at least one significant association towards increased BMI in one of the samples.


Childhood trauma and depressive symptoms may be considered as causes of obesity. These results suggest that psychological treatments against obesity should address childhood maltreatment as well as depressive symptoms in their diagnostic assessment and could facilitate psychotherapeutic treatment when necessary.

Open access