Sensing and Imaging

, 20:21 | Cite as

Discriminative Analysis of Depression Patients Studied with Structural MR Images Using Support Vector Machine and Recursive Feature Elimination

  • Jing Wang
  • Hongjun Peng
  • Yue Zhang
  • Kai WuEmail author
Original Paper
Part of the following topical collections:
  1. Recent Developments in Sensing and Imaging


Currently, the diagnosis of depression is largely based on clinical judgments due to the absence of objective biomarkers. There are increasing evidences that depression (DP) is associated with structural abnormalities. However, the previous analyses have a poor predictive power for individuals. To discriminate DP patients from normal controls (NCs) studied with structural magnetic resonance images using the method of support vector machine (SVM) combined with recursive feature elimination (RFE). In this study, 40 DP patients and 40 age- and sex-matched NCs were recruited from Guangzhou Brain Hospital and the local community, respectively. We calculated gray matter volume (GMV) and white matter volume (WMV) of 210 cortical and 36 subcortical regions, defined by the Human Brainnetome Atlas. The group differences between DP patients and NCs were compared. The method of SVM combined with RFE was applied into the discriminative analysis of DP patients from NCs, in which discriminative features were drawn from GMV and WMV. We found that the DP patients showed significant GMV reductions in eight brain regions and showed significant WMV reductions in ten brain regions. The classifier using GMV as input features achieved the best performance (an accuracy of 86.25%, a sensitivity of 85%, and a specificity of 87.5%) in the discriminative analyses between DP patients and NCs. These findings provided evidences that specific structural brain regions associated with DP patients might qualify as a potential biomarker for disease diagnosis, and the machine-learning method of SVM with RFE may reveal neurobiological mechanisms in distinguishing DP patients from NCs.


Depression Region of interest Support vector machine Recursive feature elimination 



This work was supported by the National Natural Science Foundation of China (NSFC; 31400845, 31771074), the National Key Technology Research and Development Program of the Ministry of Science and Technology of China (2015BAI13B01, 2015BAI13B02), the Guangdong Natural Science Foundation (2015A030313800), the Science and Technology Program of Guangdong (2016B010108003, 2016A020216004), the Science and Technology Program of Guangzhou (201604020170, 2017010160496, 201704020168, 201807010064), the Guangzhou Medical and Health Science and Technology Project (20171A011268, 20171A010283), and the Youth Innovation Talent Project of Guangdong Education Department (2017KQNCX259).


  1. 1.
    Gong, Q., & He, Y. (2015). Depression, neuroimaging and connectomics: A selective overview. Biological Psychiatry, 77(3), 223–235.CrossRefGoogle Scholar
  2. 2.
    Pannekoek, J. N., van der Werff, S. J. A., van den Bulk, B. G., et al. (2014). Reduced anterior cingulate gray matter volume in treatment-naive clinically depressed adolescents. NeuroImage: Clinical, 4, 336–342.CrossRefGoogle Scholar
  3. 3.
    Grieve, S. M., Korgaonkar, M. S., Koslow, S. H., et al. (2013). Widespread reductions in gray matter volume in depression. NeuroImage: Clinical, 3, 332–339.CrossRefGoogle Scholar
  4. 4.
    Zhang, H., Li, L., Wu, M., et al. (2016). Brain gray matter alterations in first episodes of depression: A meta-analysis of whole-brain studies. Neuroscience and Biobehavioral Reviews, 60, 43–50.CrossRefGoogle Scholar
  5. 5.
    Shen, Z., Cheng, Y., Yang, S., et al. (2016). Changes of grey matter volume in first-episode drug-naive adult major depressive disorder patients with different age-onset. NeuroImage: Clinical, 12, 492–498.CrossRefGoogle Scholar
  6. 6.
    Zhang, X., Yao, S., Zhu, X., et al. (2012). Gray matter volume abnormalities in individuals with cognitive vulnerability to depression: A voxel-based morphometry study. Journal of Affective Disorders, 136(3), 443–452.CrossRefGoogle Scholar
  7. 7.
    Yang, S., Cheng, Y., Mo, Y., et al. (2017). Childhood maltreatment is associated with gray matter volume abnormalities in patients with first-episode depression. Psychiatry Research: Neuroimaging, 268, 27–34.CrossRefGoogle Scholar
  8. 8.
    Bracht, T., Linden, D., & Keedwell, P. (2015). A review of white matter microstructure alterations of pathways of the reward circuit in depression. Journal of Affective Disorders, 187, 45–53.CrossRefGoogle Scholar
  9. 9.
    Guo, W., Liu, F., Xun, G., et al. (2014). Disrupted white matter integrity in first-episode, drug-naive, late-onset depression. Journal of Affective Disorders, 163, 70–75.CrossRefGoogle Scholar
  10. 10.
    Yang, X., Wang, Y., Wang, D., et al. (2017). White matter microstructural abnormalities and their association with anticipatory anhedonia in depression. Psychiatry Research: Neuroimaging, 264, 29–34.CrossRefGoogle Scholar
  11. 11.
    Jie, N., Zhu, M., Ma, X., et al. (2015). Discriminating bipolar disorder from major depression based on SVM-FoBa: Efficient feature selection with multimodal brain imaging data. IEEE Transactions on Autonomous Mental Development, 7(4), 320–331.CrossRefGoogle Scholar
  12. 12.
    Dai, Z., Yan, C., Wang, Z., et al. (2012). Discriminative analysis of early Alzheimer’s disease using multi-modal imaging and multi-level characterization with multi-classifier (M3). NeuroImage, 59(3), 2187–2195.CrossRefGoogle Scholar
  13. 13.
    Arbabshirani, M. R., Castro, E., & Calhoun, V, D. (2014). Accurate classification of schizophrenia patients based on novel resting-state fMRI features. In IEEE engineering in medicine and biology society conference proceedings (pp. 6691–6694).Google Scholar
  14. 14.
    Wu, F., Zhang, Y., & Yang, Y., et al. (2018). Structural and functional brain abnormalities in drug-naive, first-episode, and chronic patients with schizophrenia: a multimodal MRI study. Neuropsychiatric Disease and Treatment.Google Scholar
  15. 15.
    Cherkassky, V. (1997). The nature of statistical learning theory. IEEE Transactions on Neural Networks, 8(6), 1564.CrossRefGoogle Scholar
  16. 16.
    Rubin-Falcone, H., Zanderigo, F., Thapa-Chhetry, B., et al. (2018). Pattern recognition of magnetic resonance imaging-based gray matter volume measurements classifies bipolar disorder and major depressive disorder. Journal of Affective Disorders, 227, 498–505.CrossRefGoogle Scholar
  17. 17.
    Xiao, Y., Yan, Z., & Zhao, Y, et al. (2017). Support vector machine-based classification of first episode drug-naïve schizophrenia patients and healthy controls using structural MRI. Schizophrenia Research.Google Scholar
  18. 18.
    Chu, C., Hsu, A., Chou, K., et al. (2012). Does feature selection improve classification accuracy? Impact of sample size and feature selection on classification using anatomical magnetic resonance images. NeuroImage, 60(1), 59–70.CrossRefGoogle Scholar
  19. 19.
    Guyon, I., & Weston, J. (2002). Gene selection for cancer classification using support vector machines. Machine Learning, 46, 389–422.CrossRefGoogle Scholar
  20. 20.
    Ding, X., Yang, Y., Stein, E. A., et al. (2015). Multivariate classification of smokers and nonsmokers using SVM-RFE on structural MRI images. Human Brain Mapping, 36(12), 4869–4879.CrossRefGoogle Scholar
  21. 21.
    Bech, P., Allerup, P., Gram, L. F., et al. (1981). The Hamilton depression scale. Evaluation of objectivity using logistic models. Acta Psychiatrica Scandinavica, 63(3), 290–299.CrossRefGoogle Scholar
  22. 22.
    Bobo, W. V., Anglero, G. C., Jenkins, G., et al. (2016). Validation of the 17-item Hamilton depression rating scale definition of response for adults with major depressive disorder using equipercentile linking to clinical global impression scale ratings: Analysis of pharmacogenomic research network antidepressant medication pharmacogenomic study (PGRN-AMPS) data. Human Psychopharmacology, 31(3), 185–192.CrossRefGoogle Scholar
  23. 23.
    Ashburner, J. (2007). A fast diffeomorphic image registration algorithm. NeuroImage, 38(1), 95–113.CrossRefGoogle Scholar
  24. 24.
    Good, C. D., Johnsrude, I., Ashburner, J., et al. (2001). Cerebral asymmetry and the effects of sex and handedness on brain structure: A voxel-based morphometric analysis of 465 normal adult human brains. NeuroImage, 14(3), 685–700.CrossRefGoogle Scholar
  25. 25.
    Fan, L., Li, H., Zhuo, J., et al. (2016). The human brainnetome atlas: A new brain atlas based on connectional architecture. Cerebral Cortex, 26(8), 3508–3526.CrossRefGoogle Scholar
  26. 26.
    Cortes, C. V. V. (1995). Support-vector networks. Machine Learning, 20, 273–297.zbMATHGoogle Scholar
  27. 27.
    Chang, C., & Lin, C. (2011). LIBSVM: A library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2(3), 1–27.CrossRefGoogle Scholar
  28. 28.
    Kearns, M., & Ron, D. (1999). Algorithmic stability and sanity-check bounds for leave-one-out cross-validation. Neural Computation, 11(6), 1427–1453.CrossRefGoogle Scholar
  29. 29.
    Tzourio-Mazoyer, N., Landeau, B., Papathanassiou, D., et al. (2002). Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. NeuroImage, 15(1), 273–289.CrossRefGoogle Scholar
  30. 30.
    Chi, M., Guo, S., Ning, Y., et al. (2014). Using support vector machine to identify imaging biomarkers of major depressive disorder and anxious depression (Vol. 472, pp. 63–67). Wuhan: Springer.Google Scholar
  31. 31.
    Hilbert, K., Lueken, U., Muehlhan, M., et al. (2017). Separating generalized anxiety disorder from major depression using clinical, hormonal, and structural MRI data: A multimodal machine learning study. Brain and Behavior, 7(3), e00633.CrossRefGoogle Scholar
  32. 32.
    Ramasubbu, R., Brown, M. R. G., Cortese, F., et al. (2016). Accuracy of automated classification of major depressive disorder as a function of symptom severity. Neuroimage-Clinical, 12, 320–331.CrossRefGoogle Scholar
  33. 33.
    Lu, X., Yang, Y., Wu, F., et al. (2016). Discriminative analysis of schizophrenia using support vector machine and recursive feature elimination on structural MRI images. Medicine, 95(30), e3973.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.School of Biomedical EngineeringXinhua College of Sun Yat-Sen UniversityGuangzhouChina
  2. 2.Guangzhou Psychiatric HospitalAffiliated Hospital of Guangzhou Medical UniversityGuangzhouChina
  3. 3.Department of Biomedical Engineering, School of Materials Science and EngineeringSouth China University of Technology (SCUT)GuangzhouChina
  4. 4.Guangdong Engineering Technology Research Center for Translational Medicine of Mental DisordersGuangzhouChina
  5. 5.Guangdong Engineering Technology Research Center for Diagnosis and Rehabilitation of DementiaGuangzhouChina

Personalised recommendations