Table 1 Diagnosis of major depressive disorder from structural MRI studies
 Study, yearMRIHealthy controlsPatientsDiagnosisSeverity –HAMD: mean (s.d.)MedicationComparisonClassifierFeatureAccuracySensitivitySpecificityP
N (males)Age, years (s.d.)N (males)Age, years (s.d.)
Costafreda10 20091.5 T37 (9)42.8 (6.7)37 (9)43.2 (8.8)MDD20.6 (2.2)Med freeMDD v. HVSVMGM67.664.970.30.027
Gong113 T23 (12)38.223 (10)39.2 (12.9)MDD24.2 (3.8)Med naïveMDD v. HVSVMGM76.169.682.6<0.001
 2011WM84.773.995.7<0.001
GM+WM76.173.978.3<0.001
23 (14)40.4 (12.6)TRD23.5 (5.4)Med naïveTRD v. HVGM67.465.269.60.01
WM58.760.956.50.13
GM+WM65.265.265.20.02
Liu12 20121.5 T17 (10)24.2 (4.4)17 (10)26.7 (7.7)MDD25.6 (6.3)Med naïveMDD v. HVSearchlight-GM82.4
PCA-SVMWM91.2
RFE-SVMGM70.6
WM76.5
LLE- C MeansGM76.5
WM88.2
LLE-SVMGM82.4
WM88.2
18 (11)27.4 (7.7)TRD23.9 (3.7)On medsaTRD v. HVSearchlight-GM85.7
PCA-SVMWM85.7
RFE-SVMGM77.1
WM85.7
LLE-C MeansGM77.1
WM85.7
LLE-SVMGM77.1
WM85.7
+Mwangi13  2012 1.5 T18 (7)40.6 (10.3)15 (6)46.1 (12.5)TUD23.2 (4.3)On medsTUD v. HVVBM-FBM-SVMGM90.393.387.51×107b
14 (7)43.0 (13.2)15 (5)44.7 (10.0)TRD27.9 (5.8)RVMGM87.186.787.51×107b
++Kipli14 SVM-EMGM+WM+CSF85.3
 2013Information gain-Rand Tree85.3
SVM-K Means82.3
Serpa15 20141.5 T38 (8)29.7 (7.9)19 (4)29.1 (8.3)pMDD16.1cOn medsapMDD v. HVSVMGM+WM+ventricles59.631.673.7
Qiu 201416 3 T32 (23)35.0 (11.2)32 (23)34.9 (11.1)MDD24.3 (5.1)Med naïveMDD v. HVSVMCortical thickness6966720.002
Volume6663690.005
Plial area6969690.001
Curvature4847500.63
Area5966530.10
Sulcal depth5856590.12
Jacobian Metric Distortion6763720.002
+++Combination parametres6969690.002
  • Med, Medication; HAMD, 17-item Hamilton Depression Rating Scale; MDD, major depressive disorder; TRD, treatment-resistant depression; pMDD, psychotic MDD; HV, healthy volunteers; GM, grey matter; WM, white matter; CSF, cerebrospinal fluid; SVM, support vector machines; PCA, principle component analysis; RFE, recursive feature elimination; LLE, locally linear embedding; VBM, voxel based morphometry; RVM, relevance vector machine; FBM, feature based morphometry; EM, expectation-maximisation dustering algorithm; KMeans, simple K means classification via clustering; TUD, treatment unresponsive patients.

    Depression status of MDD patients: first-episode – Liu (2012), Qiu (2014), Serpa (2014); first-episode and recurrent – Costafreda (2009); recurrent: Mwangi (2012); not stated: Gong (2011), Kipli (2013).

  • a Some of the patients were medication free.

  • b χ2P.

  • c 31-item HAMD.

  • + Mwangi (2012): data were randomly divided into two sets (training set, testing set) of equal number of patients and controls (n=31). In patients, depression was considered to be treatment unresponsive. Minimum duration of illness was >3 months with antidepressant medication.

  • ++ Kipli (2013), accuracy of 82.3% also obtained with other classifiers: information gain:-J48, information gain-RandomForest, SVM-K Means, SVM-RandomForest, ReliefF-RandomTree, all-naïve bayes.

  • +++ Combined parameters: Qiu (2014) integrated all the morphometric parameters (i.e. cortical thickness, volume, plial area, curvature area, sulcal depth and Jacobian metric distortion) of the left and right hemispheres within a single model to investigate the discriminative power of the resulting combination.