Automated classification of depression from structural brain measures across two independent community‐based cohorts

Major depressive disorder (MDD) has been the subject of many neuroimaging case–control classification studies. Although some studies report accuracies ≥80%, most have investigated relatively small samples of clinically‐ascertained, currently symptomatic cases, and did not attempt replication in larger samples. We here first aimed to replicate previously reported classification accuracies in a small, well‐phenotyped community‐based group of current MDD cases with clinical interview‐based diagnoses (from STratifying Resilience and Depression Longitudinally cohort, ‘STRADL’). We performed a set of exploratory predictive classification analyses with measures related to brain morphometry and white matter integrity. We applied three classifier types—SVM, penalised logistic regression or decision tree—either with or without optimisation, and with or without feature selection. We then determined whether similar accuracies could be replicated in a larger independent population‐based sample with self‐reported current depression (UK Biobank cohort). Additional analyses extended to lifetime MDD diagnoses—remitted MDD in STRADL, and lifetime‐experienced MDD in UK Biobank. The highest cross‐validation accuracy (75%) was achieved in the initial current MDD sample with a decision tree classifier and cortical surface area features. The most frequently selected decision tree split variables included surface areas of bilateral caudal anterior cingulate, left lingual gyrus, left superior frontal, right precentral and paracentral regions. High accuracy was not achieved in the larger samples with self‐reported current depression (53.73%), with remitted MDD (57.48%), or with lifetime‐experienced MDD (52.68–60.29%). Our results indicate that high predictive classification accuracies may not immediately translate to larger samples with broader criteria for depression, and may not be robust across different classification approaches.     

Depression in type 1 diabetes and risk of dementia

Objective: Depression afflicts 14% of individuals with type 1 diabetes (T1D). Depression is a robust risk factor for dementia but it is unknown if this holds true for individuals with T1D, who recently started living to an age conferring dementia risk. We examined if depression is a dementia risk factor among elderly individuals with T1D.

Methods: 3,742 individuals with T1D age ≥50 were followed for dementia from 1/1/96-9/30/2015. Depression, dementia, and comorbidities were abstracted from electronic medical records. Cox proportional hazard models estimated the association between depression and dementia adjusting for demographics, glycosylated hemoglobin, severe dysglycemic epidsodes, stroke, heart disease, nephropathy, and end stage renal disease. The cumulative incidence of dementia by depression was estimated conditional on survival dementia-free to age 55.

Results: Five percent (N = 182) were diagnosed with dementia and 20% had baseline depression. Depression was associated with a 72% increase in dementia (fully adjusted HR = 1.72; 95% CI:1.12-2.65). The 25-year cumulative incidence of dementia was more than double for those with versus without depression (27% vs. 12%).

Conclusions: For people with T1D, depression significantly increases dementia risk. Given the pervasiveness of depression in T1D, this has major implications for successful aging in this population recently living to old age.     

Laterality of cross-modal spatial processing

This study examined the laterality of spatial-form perception in normal adults using a cross-modal matching paradigm involving visual and tactile processing. Randomly generated eight-point and 12-point Vanderplas and Garvin (1959) patterns were used as the stimuli. In a visual-to-tactile task, a lateralized visual pattern was followed immediately by a haptic task requiring subjects to discriminate which of two simultaneously felt shapes matched the visual pattern. In a tactile-to-visual task, subjects decided which of two simultaneously felt shapes matched a lateralized visual pattern presented after haptic manipulation. There were no main effects for laterality or for sex differences. Matching accuracy was better in the visual-to-tactile task and for less complex stimuli. A visual field by feeling hand interaction showed best recognition accuracy when visual-feeling hand combinations on the same side of the body were used in the two matching tasks. These data reflect a stimulus-response compatibility explanation of spatial-form perception that is consistent with a behavioral and not a cerebral asymmetry model.     

Youth Tobacco Access and Possession Policy Interventions: Effects on Observed and Perceived Tobacco Use

This study evaluated the effects of tobacco Purchase, Use and Possession (PUP) laws on student perceptions of adolescent tobacco use within towns and schools. Twenty‐four towns were randomly assigned into two conditions, the experimental condition (E PUP) involved efforts to increase both PUP law enforcement and reduce minors' access to commercial sources of tobacco, whereas the control condition (C) focused only on efforts to reduce minors' access to commercial sources of tobacco. A hierarchical linear modeling analytical approach was selected due to the multilevel data and nested design. The present study found that over time, youth in the experimental PUP condition observed less youth tobacco usage at school and in their town, and perceived lower rates of tobacco among their peers at school and among friends than youth in the control condition. The findings suggest that PUP law enforcement might be used to strengthen community norms against youth tobacco use.     

Preconception Care for Improving Perinatal Outcomes: The Time to Act

Maternal and Child Health Journal -