Automated classification of depression from structural brain measures across two independent community‐based cohorts. Issue 14 (19th June 2020)

• Record Type: Journal Article
• Authors: Stolicyn, Aleks; Harris, Mathew A.; Shen, Xueyi; Barbu, Miruna C.; Adams, Mark J.; Hawkins, Emma L.; de Nooij, Laura; Yeung, Hon Wah; Murray, Alison D.; Lawrie, Stephen M.; Steele, J. Douglas; McIntosh, Andrew M.; Whalley, Heather C.
• Is Part Of: Human brain mapping. Volume 41:Issue 14(2020)
• Page Start: 3922
• Page End: 3937
• Publication Date: 2020-06-19
• Subjects: brain structure -- classification -- depression -- diffusion MRI -- machine learning -- major depressive disorder -- structural MRI; Brain mapping -- Periodicals; 611.81
• DOI: 10.1002/hbm.25095
• Languages: English
• ISSNs: 1065-9471

Abstract

Abstract: 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. Abstract : Previous case‐control classification studies of depression have reported promising accuracies up to 80% and above, but investigated relatively small samples with clinically‐ascertained cases. In the current study we explore multiple classification approaches with brain structural and white matter integrity measures, and achieve up to 75% predictive classification accuracy in a small sample with clinical interview‐based diagnoses ( N < 50 cases, STRADL cohort). This accuracy, however, is not replicated in a larger sample of cases with self‐reported depression ( N > 700 cases, UK Biobank cohort), and also in larger samples with remitted MDD (STRADL) and lifetime‐experienced MDD (UK Biobank). These results indicate that high predictive case‐control classification accuracies may not immediately translate to larger samples with broader diagnostic criteria for depression.