Advertisement

Feature Selection and Imbalanced Data Handling for Depression Detection

  • Marzieh Mousavian
  • Jianhua Chen
  • Steven Greening
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11309)

Abstract

Major Depressive Disorder (MDD) is the most common disorder worldwide. Accurate detection of depression is a challenging problem. Machine learning-based automated depression detection provides useful assistance to the clinicians for effective depression diagnosis. One of the most fundamental steps in any automated detection is feature selection and investigation of the most relevant features. Studies show that regional volumes of the brain are affected in response to depression. Regional volumes are considered as features. The gray matter volumes’ correlation with depression and the most effective gray volumes for depression detection is investigated in this study. Various feature selection techniques are studied, along with the investigation on the importance of resampling to handle imbalanced data, which is typically the case for depression detection, as the number of depressed instances is commonly a fraction of the entire data size. Experimental results using Random Forests (RF) and support vector machines (SVM) with a Gaussian kernel (RBF) as classifiers show that feature selection followed by data resampling gives superior performance measured by Area Under the ROC Curve (AUC) as well as prediction accuracy, and RF outperforms SVM for the depression detection task.

Keywords

Depression detection Feature selection Imbalanced data 

References

  1. 1.
    Gong, Q., et al.: Prognostic prediction of therapeutic response in depression using high-field MR imaging. NeuroImage 55, 1497–1503 (2011)CrossRefGoogle Scholar
  2. 2.
    Kipli, K., Kouzani, A.: Degree of contribution (DoC) feature selection algorithm for structural brain MRI volumetric features in depression detection. Int. J. Comput. Assist. Radiol. Surg. 10, 1003–1016 (2014)CrossRefGoogle Scholar
  3. 3.
    Patel, M., Khalaf, A., Aizenstein, H.: Studying depression using imaging and machine learning methods. NeuroImage: Clin. 10, 115–123 (2016)CrossRefGoogle Scholar
  4. 4.
    Costafreda, S., Chu, C., Ashburner, J., Fu, C.: Prognostic and diagnostic potential of the structural neuroanatomy of depression. PLoS ONE 4, e6353 (2009)CrossRefGoogle Scholar
  5. 5.
    Fcon_1000.projects.nitrc.org. Accessed 28 Mar 2018Google Scholar
  6. 6.
    Mwangi, B., Matthews, K., Steele, J.: Prediction of illness severity in patients with major depression using structural MR brain scans. J. Magn. Reson. Imaging 35, 64–71 (2011)CrossRefGoogle Scholar
  7. 7.
    Mwangi, B., Tian, T., Soares, J.: A review of feature reduction techniques in neuroimaging. Neuroinformatics 12, 229–244 (2013)CrossRefGoogle Scholar
  8. 8.
    Hira, Z., Gillies, D.: A review of feature selection and feature extraction methods applied on microarray data. Adv. Bioinf. 2015, 1–13 (2015)CrossRefGoogle Scholar
  9. 9.
    Feature Selection/Extraction Dimensionality Reduction. http://vision.psych.umn.edu/users/schrater/schrater_lab/courses/PattRecog09/Lec17PattRec09.pdf. Accessed 28 Mar 2018
  10. 10.
    Hemphill, E., Lindsay, J., Lee, C., Măndoiu, I., Nelson, C.: Feature selection and classifier performance on diverse bio- logical datasets. BMC Bioinf. 15, S4 (2014)CrossRefGoogle Scholar
  11. 11.
    Geng, X., Xu, J.: Application of autoencoder in depression diagnosis. In: 3rd International Conference on Computer Science and Mechanical Automation, CSMA (2017)Google Scholar
  12. 12.
    Bosch, N., Paquette, L.: Unsupervised Deep Autoencoders for Feature Extraction with Educational Data (2018)Google Scholar
  13. 13.
    Unbalanced data and cross-validation|Kaggle. https://www.kaggle.com/questions-and-answers/27589. Accessed 28 Mar 2018
  14. 14.
    Scikit-learn-contrib/imbalanced-learn. https://github.com/scikit-learn-contrib/imbalanced-learn. Accessed 28 Mar 2018
  15. 15.
    Malone, I., et al.: Accurate automatic estimation of total intracranial volume: a nuisance variable with less nuisance. NeuroImaging 104, 366–372 (2015)CrossRefGoogle Scholar
  16. 16.
    Amami, R., Ayed, D.B., Ellouze, N.: Practical selection of SVM supervised parameters with different feature representations for vowel recognition. arXiv preprint arXiv:1507.06020 (2015)

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  • Marzieh Mousavian
    • 1
  • Jianhua Chen
    • 1
  • Steven Greening
    • 2
  1. 1.Division of Computer Science and Engineering, School of EECSLouisiana State UniversityBaton RougeUSA
  2. 2.Psychology DepartmentLouisiana State UniversityBaton RougeUSA

Personalised recommendations