Skip to main content

Advertisement

SpringerLink
Log in

Depression detection from sMRI and rs-fMRI images using machine learning

  • Published:
Journal of Intelligent Information Systems Aims and scope Submit manuscript

Abstract

Major Depression Disorder (MDD) is a common mental disorder that negatively affects many people’s lives worldwide. Developing an automated method to find useful diagnostic biomarkers from brain imaging data would help clinicians to detect MDD in its early stages. Depression is known to be a brain connectivity disorder problem. In this paper, we present a brain connectivity-based machine learning (ML) workflow that utilizes similarity/dissimilarity of spatial cubes in brain MRI images as features for depression detection. The proposed workflow provides a unified framework applicable to both structural MRI images and resting-state functional MRI images. Several cube similarity measures have been explored, including Pearson or Spearman correlations, Minimum Distance Covariance, or inverse of Minimum Distance Covariance. Discriminative features from the cube similarity matrix are chosen with the Wilcoxon rank-sum test. The extracted features are fed into machine learning classifiers to train MDD prediction models. To address the challenge of data imbalance in MDD detection, oversampling is performed to balance the training data. The proposed workflow is evaluated through experiments on three independent public datasets, all imbalanced, of structural MRI and resting-state fMRI images with depression labels. Experimental results show good performance on all three datasets in terms of prediction accuracy, specificity, sensitivity, and area under the Receiver Operating Characteristic (ROC) curve. The use of features from both structured MRI and resting state functional MRI is also investigated.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

Data Availability

The three datasets that we used, are available online. Web addresses of the datasets are provided as follows.The NKI-Enhanced dataset contains raw SMRI and rs-fMRI data available at http://fcon_1000.proj-ects.nitrc.org/indi/enhanced/neurodata.html. The MPI-Leipzig-Mind-Brain-Body dataset is available online at https://www.neuroconnlab.org/data/. The Closed-eyes dataset is available online at https://openneuro.org/datasets/ds002748/versions/1.0.2.

References

  • Bae, J.N., MacFall, J.R., Krishnan, K.R.R., Payne, M.E., Steffens, D.C., & et al (2006). Dorsolateral prefrontal cortex and anterior cingulate cortex white matter alterations in late-life depression. Biological Psychiatry, 60(12), 1356–1363.

    Article  Google Scholar 

  • Beck, A.T., Steer, R.A., & Brown, G. (1996). Beck depression inventory–ii. Psychological Assessment.

  • Bezmaternykh, D., Melnikov, M., Savelov, A., & Petrovski, E. (2020). Resting state with closed eyes for patients with depression and healthy participants. https://openneuro.org/datasets/ds002748/versions/1.0.2.

  • Biswal, B.B., & Ulmer, J.L. (1999). Blind source separation of multiple signal sources of fmri data sets using independent component analysis. Journal of Computer Assisted Tomography, 23(2), 265–271.

    Article  Google Scholar 

  • Brandt, W.A., Loew, T., von Heymann, F., Stadtmüller, G., Tischinger, M., & et al (2015). How does the icd-10 symptom rating (isr) with four items assess depression compared to the bdi-ii? a validation study. Journal of Affective Disorders, 173, 143–145.

    Article  Google Scholar 

  • Casanova, R., Wagner, B., Whitlow, C.T., Williamson, J.D., Shumaker, S.A., & et al (2011). High dimensional classification of structural mri alzheimer’s disease data based on large scale regularization. Frontiers in Neuroinformatics, 5, 22.

    Article  Google Scholar 

  • Chen, T., Kendrick, K.M., Wang, J., Wu, M., Li, K., & et al (2017). Anomalous single-subject based morphological cortical networks in drug-naive, first-episode major depressive disorder. Human Brain Mapping, 38(5), 2482–2494.

    Article  Google Scholar 

  • De Luca, M., Beckmann, C.F., De Stefano, N., Matthews, P.M., & Smith, S.M. (2006). fmri resting state networks define distinct modes of long-distance interactions in the human brain. NeuroImage, 29(4), 1359–1367.

    Article  Google Scholar 

  • Esteban, O., Markiewicz, C.J., Blair, R.W., Moodie, C.A., Isik, A.I., & et al (2019). fmriprep: a robust preprocessing pipeline for functional mri. Nature Methods, 16(1), 111–116.

    Article  Google Scholar 

  • Faber, J., Antoneli, P.C., Araújo, N S, Pinheiro, D.J., & Cavalheiro, E. (2020). Critical elements for connectivity analysis of brain networks. In Functional brain mapping: methods and aims (pp. 67–107). Springer.

  • Foland-Ross, L.C., Sacchet, M.D., Prasad, G., Gilbert, B., Thompson, P.M., & et al (2015). Cortical thickness predicts the first onset of major depression in adolescence. International Journal of Developmental Neuroscience, 46, 125–131.

    Article  Google Scholar 

  • Fritz, C.O., Morris, P.E., & Richler, J.J. (2012). Effect size estimates: current use, calculations, and interpretation. Journal of Experimental Psychology: General, 141(1), 2.

    Article  Google Scholar 

  • Fu, C.H., Williams, S.C., Cleare, A.J., Brammer, M.J., Walsh, N.D., & et al (2004). Attenuation of the neural response to sad faces in major depressionby antidepressant treatment: a prospective, event-related functional magnetic resonance imagingstudy. Archives of general psychiatry, 61(9), 877–889.

    Article  Google Scholar 

  • Gabrieli, J.D., Ghosh, S.S., & Whitfield-Gabrieli, S. (2015). Prediction as a humanitarian and pragmatic contribution from human cognitive neuroscience. Neuron, 85(1), 11–26.

    Article  Google Scholar 

  • Gao, S., Osuch, E.A., Wammes, M., Théberge, J, Jiang, T.Z., & et al (2017). Discriminating bipolar disorder from major depression based on kernel svm using functional independent components. In 2017 IEEE 27th international workshop on machine learning for signal processing (MLSP) (pp. 1–6). IEEE.

  • Gorgolewski, K.J., Auer, T., Calhoun, V.D., Craddock, R.C., Das, S., & et al (2016). The brain imaging data structure, a format for organizing and describing outputs of neuroimaging experiments. Scientific Data, 3(1), 1–9.

    Article  Google Scholar 

  • Gorgolewski, K.J., Esteban, O., Burns, C., Ziegler, E., Pinsard, B., & et al. (2016). Nipype: a flexible, lightweight and extensible neuroimaging data processing framework in Python. 0.12.0-rc1. https://doi.org/10.5281/zenodo.50186.

  • Greening, S.G., Osuch, E.A., Williamson, P.C., & Mitchell, D.G. (2014). The neural correlates of regulating positive and negative emotions in medication-free major depression. Social Cognitive and Affective Neuroscience, 9(5), 628–637.

    Article  Google Scholar 

  • Guo, S., Huang, C.C., Zhao, W., Yang, A.C., Lin, C.P., Nichols, T., & Tsai, S.J. (2018). Combining multi-modality data for searching biomarkers in schizophrenia. https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0191202.

  • Guo, Wb, Liu, F., Xue, Zm, Xu, X., Wu, Rr, & et al (2012). Alterations of the amplitude of low-frequency fluctuations in treatment-resistant and treatment-response depression: a resting-state fmri study. Progress in neuro-psychopharmacology and amp. Biological Psychiatry, 37(1), 153–160. https://doi.org/10.1016/j.pnpbp.2012.01.011.

    Google Scholar 

  • He, H., Bai, Y., Garcia, E.A., & Li, S. (2008). Adasyn: Adaptive synthetic sampling approach for imbalanced learning. In 2008 IEEE international joint conference on neural networks (IEEE world congress on computational intelligence) (pp. 1322–1328). IEEE.

  • He, H., Sui, J., Du, Y., Yu, Q., Lin, D., & et al (2017). Co-altered functional networks and brain structure in unmedicated patients with bipolar and major depressive disorders. Brain Structure and Function, 222(9), 4051–4064.

    Article  Google Scholar 

  • Hojjati, S.H., Ebrahimzadeh, A., Khazaee, A., & Babajani-Feremi, A. (2018). Predicting conversion from mci to ad by integrating rs-fmri and structural mri. Computers in Biology and Medicine, 102, 30–39. https://doi.org/10.1016/j.compbiomed.2018.09.004.

    Article  Google Scholar 

  • Jabason, E., Ahmad, M.O., & Swamy, M. (2019). Hybrid feature fusion using rnn and pre-trained cnn for classification of alzheimer’s disease (poster). In 2019 22th International Conference On Information Fusion (FUSION) (pp. 1–4). IEEE.

  • Jing, B., Long, Z., Liu, H., Yan, H., Dong, J., & et al (2017). Identifying current and remitted major depressive disorder with the hurst exponent: a comparative study on two automated anatomical labeling atlases. Oncotarget, 8(52), 90452.

    Article  Google Scholar 

  • Johnston, B.A., Steele, J.D., Tolomeo, S., Christmas, D., & Matthews, K. (2015). Structural mri-based predictions in patients with treatment-refractory depression (trd). PloS ONE, 10(7), e0132958.

    Article  Google Scholar 

  • Kambeitz, J., Cabral, C., Sacchet, M.D., Gotlib, I.H., Zahn, R., & et al (2017). Detecting neuroimaging biomarkers for depression: a meta-analysis of multivariate pattern recognition studies. Biological Psychiatry, 82(5), 330–338.

    Article  Google Scholar 

  • Kaye, N.S. (2005). Is your depressed patient bipolar? The Journal of the American Board of Family Practice, 18(4), 271–281.

    Article  Google Scholar 

  • Kipli, K., & Kouzani, A.Z. (2015). Degree of contribution (doc) feature selection algorithm for structural brain mri volumetric features in depression detection. International Journal of Computer Assisted Radiology and Surgery, 10(7), 1003–1016.

    Article  Google Scholar 

  • Kelly, C.J., Karthikesalingam, A., Suleyman, M., Corrado, G., & King, D (2019). Dominic Key challenges for delivering clinical impact with artificial intelligence. BMC Medicine, 17(1), 1–9.

    Article  Google Scholar 

  • Lv, H., Wang, Z., Tong, E., Williams, L.M., Zaharchuk, G., & et al (2018). Resting-state functional mri: everything that nonexperts have always wanted to know. American Journal of Neuroradiology, 39(8), 1390–1399.

    Google Scholar 

  • Mendes, N., Oligschläger, S, Lauckner, M.E., Golchert, J., Huntenburg, J.M., & et al (2019). A functional connectome phenotyping dataset including cognitive state and personality measures. Scientific Data, 6(1), 1–19.

    Article  Google Scholar 

  • Mheich, A., Wendling, F., & Hassan, M. (2020). Brain network similarity: methods and applications. Network Neuroscience, 4(3), 507–527.

    Article  Google Scholar 

  • Mohanty, R., Sethares, W.A., Nair, V.A., & Prabhakaran, V. (2020). Rethinking measures of functional connectivity via feature extraction. Scientific Reports, 10(1), 1–17.

    Article  Google Scholar 

  • Mwangi, B., Ebmeier, K.P., Matthews, K., & Douglas Steele, J. (2012). Multi-centre diagnostic classification of individual structural neuroimaging scans from patients with major depressive disorder. Brain: A Journal of Neurology, 135(5), 1508–1521.

    Article  Google Scholar 

  • Mwangi, B., Tian, T.S., & Soares, J.C. (2014). A review of feature reduction techniques in neuroimaging. Neuroinformatics, 12(2), 229–244.

    Article  Google Scholar 

  • O’Brien, L.M., Ziegler, D.A., Deutsch, C.K., Frazier, J.A., Herbert, M.R., & et al (2011). Statistical adjustments for brain size in volumetric neuroimaging studies: some practical implications in methods. Psychiatry Research: Neuroimaging, 193(2), 113–122.

    Article  Google Scholar 

  • Patel, M.J., Khalaf, A., & Aizenstein, H.J. (2016). Studying depression using imaging and machine learning methods. NeuroImage: Clinical, 10, 115–123.

    Article  Google Scholar 

  • Pominova, M., Artemov, A., Sharaev, M., Kondrateva, E., Bernstein, A., & et al (2018). Voxelwise 3d convolutional and recurrent neural networks for epilepsy and depression diagnostics from structural and functional mri data. In 2018 IEEE International Conference On Data Mining Workshops (ICDMW) (pp. 299–307). IEEE.

  • Queirós, GCdP. (2013). Computational methods for fmri image processing and analysis.

  • Qureshi, M.N.I., Oh, J., & Lee, B. (2019). 3d-cnn based discrimination of schizophrenia using resting-state fmri. Artificial Intelligence in Medicine, 98, 10–17.

    Article  Google Scholar 

  • Ray, D., Bezmaternykh, D., Mel’nikov, M., Friston, K.J., & Das, M. (2021). Altered effective connectivity in sensorimotor cortices: a novel signature of severity and clinical course in depression. bioRxiv.

  • Rubin-Falcone, H., Zanderigo, F., Thapa-Chhetry, B., Lan, M., Miller, J.M., & 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.

    Article  Google Scholar 

  • Sacchet, M.D., Prasad, G., Foland-Ross, L.C., Thompson, P.M., & Gotlib, I.H. (2015). Support vector machine classification of major depressive disorder using diffusion-weighted neuroimaging and graph theory. Frontiers in Psychiatry, 6, 21.

    Article  Google Scholar 

  • Sankar, A., Zhang, T., Gaonkar, B., Doshi, J., Erus, G., Costafreda, S.G., Marangell, L., Davatzikos, C., & Fu, C.H. (2016). Diagnostic potential of structural neuroimaging for depression from a multi-ethnic community sample. BJPsych Open, 2(4), 247–254.

    Article  Google Scholar 

  • Sato, J.R., Hoexter, M.Q., Castellanos, X.F., & Rohde, L.A. (2012). Abnormal brain connectivity patterns in adults with adhd: a coherence study. PloS ONE, 7(9), e45671.

    Article  Google Scholar 

  • Sheline, Y.I., Barch, D.M., Price, J.L., Rundle, M.M., Vaishnavi, S.N., & et al (2009). The default mode network and self-referential processes in depression. Proceedings of the National Academy of Sciences, 106(6), 1942–1947.

    Article  Google Scholar 

  • Siegle, G.J., Steinhauer, S.R., Thase, M.E., Stenger, V.A., & Carter, C.S. (2002). Can’t shake that feeling: event-related fmri assessment of sustained amygdala activity in response to emotional information in depressed individuals. Biological Psychiatry, 51(9), 693–707.

    Article  Google Scholar 

  • Smith, S.M., Fox, P.T., Miller, K.L., Glahn, D.C., Fox, P.M., & et al (2009). Correspondence of the brain’s functional architecture during activation and rest. Proceedings of the National Academy of Sciences, 106(31), 13040–13045.

    Article  Google Scholar 

  • Smith, S.E., Jakobsen, I., Grønlund, M, & Smith, F.A. (2011). Roles of arbuscular mycorrhizas in plant phosphorus nutrition: interactions between pathways of phosphorus uptake in arbuscular mycorrhizal roots have important implications for understanding and manipulating plant phosphorus acquisition. Plant Physiology, 156(3), 1050–1057.

    Article  Google Scholar 

  • Smith, S.M., Jenkinson, M., Woolrich, M.W., Beckmann, C.F., Behrens, T.E., & et al (2004). Advances in functional and structural mr image analysis and implementation as fsl. NeuroImage, 23, S208–S219.

    Article  Google Scholar 

  • Tai, L.K., Setyonugroho, W., & Chen, A.L. (2020). Finding discriminatory features from electronic health records for depression prediction. Journal of Intelligent Information Systems, 55(2), 371–396.

    Article  Google Scholar 

  • Van Den Heuvel, M.P., & Pol, H.E.H. (2010). Exploring the brain network: a review on resting-state fmri functional connectivity. European Neuropsychopharmacology, 20(8), 519–534.

    Article  Google Scholar 

  • Varoquaux, G., & Craddock, R.C. (2013). Learning and comparing functional connectomes across subjects. NeuroImage, 80, 405–415.

    Article  Google Scholar 

  • Varoquaux, G., Sadaghiani, S., Pinel, P., Kleinschmidt, A., Poline, J.B., & et al (2010). A group model for stable multi-subject ica on fmri datasets. NeuroImage, 51(1), 288–299.

    Article  Google Scholar 

  • Varshney, A., Prakash, C., Mittal, N., & Singh, P. (2016). A multimodel approach for schizophrenia diagnosis using fmri and smri dataset. In JM Corchado Rodriguez, S Mitra, SM Thampi, & ES El-Alfy (Eds.) Intelligent systems technologies and applications 2016 (pp. 869–877). Cham: Springer International Publishing.

  • Wang, Y.P., & Gorenstein, C. (2013). Assessment of depression in medical patients: a systematic review of the utility of the beck depression inventory-ii. Clinics, 68(9), 1274–1287.

    Article  Google Scholar 

  • Wang, X., Ren, Y., & Zhang, W. (2017). Depression disorder classification of fmri data using sparse low-rank functional brain network and graph-based features. Computational and Mathematical Methods in Medicine, pp 2017.

  • Woolrich, M.W., Jbabdi, S., Patenaude, B., Chappell, M., Makni, S., & et al (2009). Bayesian analysis of neuroimaging data in fsl. NeuroImage, 45(1), S173–S186.

    Article  Google Scholar 

  • Woolrich, M.W., Ripley, B.D., Brady, M., & Smith, S.M. (2001). Temporal autocorrelation in univariate linear modeling of fmri data. NeuroImage, 14(6), 1370–1386.

    Article  Google Scholar 

  • Wu, M.Y., Shen, C.Y., Wang, E.T., & Chen, A.L. (2020). A deep architecture for depression detection using posting, behavior, and living environment data. Journal of Intelligent Information Systems, 54(2), 225–244.

    Article  Google Scholar 

  • Xiao, Y., Fonov, V., Chakravarty, M.M., Beriault, S., Al Subaie, F., & et al (2017). A dataset of multi-contrast population-averaged brain mri atlases of a parkinson’s disease cohort. Data in brief, 12, 370–379.

    Article  Google Scholar 

  • Yan, B., Xu, X., Liu, M., Zheng, K., Liu, J., & et al (2020). Quantitative identification of major depression based on resting-state dynamic functional connectivity: A machine learning approach. Frontiers in Neuroscience, 14, 191. https://doi.org/10.3389/fnins.2020.00191.

    Article  Google Scholar 

  • Yao, Z., Hu, B., Xie, Y., Moore, P., & Zheng, J. (2015). A review of structural and functional brain networks: small world and atlas. Brain Informatics, 2(1), 45–52.

    Article  Google Scholar 

  • Ye, M., Yang, T., Qing, P., Lei, X., Qiu, J., & et al (2015). Changes of functional brain networks in major depressive disorder: a graph theoretical analysis of resting-state fmri. PloS ONE, 10(9), e0133775.

    Article  Google Scholar 

  • Zang, Y.F., Zuo, X.N., Milham, M, & Hallett, M. (2015). Toward a meta-analytic synthesis of the resting-state fmri literature for clinical populations.

  • Zhao, Y., Niu, R., Lei, D., Shah, C., Xiao, Y., & et al. (2020). Aberrant gray matter networks in non-comorbid medication-naive patients with major depressive disorder and those with social anxiety disorder. Frontiers in Human Neuroscience.

  • Zhong, X., Shi, H., Ming, Q., Dong, D., Zhang, X., & et al (2017). Whole-brain resting-state functional connectivity identified major depressive disorder: a multivariate pattern analysis in two independent samples. Journal of Affective Disorders, 218, 346–352.

    Article  Google Scholar 

  • Zhuo, C., Li, G., Lin, X., Jiang, D., Xu, Y., & et al (2019). The rise and fall of MRI studies in major depressive disorder. Translational Psychiatry, 9(1), 1–14.

    Article  Google Scholar 

  • Zisook, S., Lesser, I., Stewart, J.W., Wisniewski, S.R., Balasubramani, G., & et al (2007). Effect of age at onset on the course of major depressive disorder. American Journal of Psychiatry, 164(10), 1539–1546.

    Article  Google Scholar 

  • Zou, L., Zheng, J., Miao, C., Mckeown, M.J., & Wang, Z.J. (2017). 3d cnn based automatic diagnosis of attention deficit hyperactivity disorder using functional and structural mri. IEEE Access, 5, 23626–23636. https://doi.org/10.1109/ACCESS.2017.2762703.

    Article  Google Scholar 

Download references

Acknowledgements

We would like to thank all of those involved in the participation, data collection, and data sharing initiative of the Enhanced Rockland Sample, MPI and closed-eye datasets. Also, we would like to thank the anonymous reviews for their constructive valuable comments.

Author information

Authors and Affiliations

  1. Division of Computer Science and Engineering, School of EECS, Louisiana State University, Baton Rouge, LA, USA

    Marzieh Mousavian & Jianhua Chen

  2. Edward P. Fitts Department of Industrial and Systems Engineering, College of Engineering, North Carolina State University, Raleigh, NC, USA

    Zachary Traylor

  3. Psychology Department, University of Manitoba, Manitoba, Canada

    Steven Greening

Authors
  1. Marzieh Mousavian
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jianhua Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Zachary Traylor
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Steven Greening
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Marzieh Mousavian.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mousavian, M., Chen, J., Traylor, Z. et al. Depression detection from sMRI and rs-fMRI images using machine learning. J Intell Inf Syst 57, 395–418 (2021). https://doi.org/10.1007/s10844-021-00653-w

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10844-021-00653-w

Keywords

Search

Navigation