Schizophrenia is a severe neuropsychiatric disease whose diagnosis, unfortunately, lacks an objective diagnostic tool supporting a thorough psychiatric examination of the patient. We took advantage of today's computational abilities, structural magnetic resonance imaging, and modern machine learning methods, such as stacked autoencoders (SAE) and 3D convolutional neural networks (3D CNN), to teach them to classify 52 patients with schizophrenia and 52 healthy controls. The main aim of this study was to explore whether complex feature extraction methods can help improve the accuracy of deep learning-based classifiers compared to minimally preprocessed data. Our experiments employed three commonly used preprocessing steps to extract three different feature types. They included voxel-based morphometry, deformation-based morphometry, and simple spatial normalization of brain tissue. In addition to classifier models, features and their combination, other model parameters such as network depth, number of neurons, number of convolutional filters, and input data size were also investigated. Autoencoders were trained on feature pools of 1000 and 5000 voxels selected by Mann-Whitney tests, and 3D CNNs were trained on whole images. The most successful model architecture (autoencoders) achieved the highest average accuracy of 69.62% (sensitivity 68.85%, specificity 70.38%). The results of all experiments were statistically compared (the Mann-Whitney test). In conclusion, SAE outperformed 3D CNN, while preprocessing using VBM helped SAE improve the results.

Department of Psychiatry University Hospital Brno Masaryk University Jihlavska 20 625 00 Brno Czech Republic

Faculty of Medicine Institute of Biostatistics and Analyses Masaryk University Kamenice 3 625 00 Brno Czech Republic

Zobrazit více v PubMed

GBD 2017 Disease and Injury Incidence and Prevalence Collaborators Global, regional, and national incidence, prevalence, and years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990–2017: A systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2018;392:1789–1858. doi: 10.1016/S0140-6736(18)32279-7. PubMed DOI PMC

Perkins D.O., Gu H., Boteva K., Lieberman J.A. Relationship between duration of untreated psychosis and outcome in first-episode schizophrenia: A critical review and meta-analysis. Am. J. Psychiatry. 2005;162:1785–1804. doi: 10.1176/appi.ajp.162.10.1785. PubMed DOI

Kubicki M., Shenton M.E., Salisbury D.F., Hirayasu Y., Kasai K., Kikinis R., Jolesz F.A., McCarley R.W. Voxel-Based Morphometric Analysis of Gray Matter in First Episode Schizophrenia. Neuroimage. 2002;17:1711–1719. doi: 10.1006/nimg.2002.1296. PubMed DOI PMC

Shenton M.E., Dickey C.C., Frumin M., McCarley R.W. A review of MRI findings in schizophrenia. Schizophr. Res. 2001;49:1–52. doi: 10.1016/S0920-9964(01)00163-3. PubMed DOI PMC

Ho N.F., Chong P.L.H., Lee D.R., Chew Q.H., Chen G., Sim K. The Amygdala in Schizophrenia and Bipolar Disorder: A Synthesis of Structural MRI, Diffusion Tensor Imaging, and Resting-State Functional Connectivity Findings. Harv. Rev. Psychiatry. 2019;27:150–164. doi: 10.1097/HRP.0000000000000207. PubMed DOI

Lee D.-K., Lee H., Park K., Joh E., Kim C.-E., Ryu S. Common gray and white matter abnormalities in schizophrenia and bipolar disorder. PLoS ONE. 2020;15:e0232826. doi: 10.1371/journal.pone.0232826. PubMed DOI PMC

Zhao C., Zhu J., Liu X., Pu C., Lai Y., Chen L., Yu X., Hong N. Structural and functional brain abnormalities in schizophrenia: A cross-sectional study at different stages of the disease. Prog. Neuro-Psychopharmacol. Biol. Psychiatry. 2018;83:27–32. doi: 10.1016/j.pnpbp.2017.12.017. PubMed DOI

Good C.D., Scahill R., Fox N.C., Ashburnera J., Friston K.J., Chanb D., Crum W.R., Rossor M.N., Frackowiak R.S. Automatic differentiation of anatomical patterns in the human brain: Validation with studies of degenerative dementias. Neuroimage. 2002;17:29–46. doi: 10.1006/nimg.2002.1202. PubMed DOI

Soloff P., Nutche J., Goradia D., Diwadkar V. Structural brain abnormalities in borderline personality disorder: A voxel-based morphometry study. Psychiatry Res. 2008;164:223–236. doi: 10.1016/j.pscychresns.2008.02.003. PubMed DOI PMC

Aguilar-Ortiz S., Salgado-Pineda P., Marco-Pallarés J., Pascual J.C., Vega D., Soler J., Brunel C., Martin-Blanco A., Soto A., Ribas J., et al. Abnormalities in gray matter volume in patients with borderline personality disorder and their relation to lifetime depression: A VBM study. PLoS ONE. 2018;13:e0191946. doi: 10.1371/journal.pone.0191946. PubMed DOI PMC

Labudda K., Kreisel S., Beblo T., Mertens M., Kurlandchikov O., Bien C.G., Driessen M., Woermann F.G. Mesiotemporal volume loss associated with disorder severity: A VBM study in borderline personality disorder. PLoS ONE. 2013;8:e83677. doi: 10.1371/journal.pone.0083677. PubMed DOI PMC

Ashburner J., Friston K.J. Voxel-based morphometry--the methods. Pt 1Neuroimage. 2000;11:805–821. doi: 10.1006/nimg.2000.0582. PubMed DOI

Wright I.C., McGuire P., Poline J.-B., Travere J.M., Murray R., Frith C., Frack-Owiak R.S.J., Friston K. A voxel-based method for the statistical analysis of gray and white matter density applied to schizophrenia. Neuroimage. 1995;2:244–252. doi: 10.1006/nimg.1995.1032. PubMed DOI

Janousova E., Schwarz D., Kasparek T. Combining various types of classifiers and features extracted from magnetic resonance imaging data in schizophrenia recognition. Psychiatry Res. Neuroimaging. 2015;232:237–249. doi: 10.1016/j.pscychresns.2015.03.004. PubMed DOI

Dluhoš P., Schwarz D., Kašpárek T. Wavelet Features for Recognition of First Episode of Schizophrenia from MRI Brain Images. Radioengineering. 2014;23:274–281.

Jafri M.J., Calhoun V.D. Functional classification of schizophrenia using feed forward neural networks; Proceedings of the 2006 International Conference of the IEEE Engineering in Medicine and Biology Society; New York, NY, USA. 30 August–3 September 2006; pp. 6631–6634. PubMed DOI

Savio A.C.J. Neural classifiers for schizophrenia diagnostic support on diffusion imaging data. Neural Netw. World. 2010;20:935–949.

Nieuwenhuis M., van Haren N.E.M., Pol H.E.H., Cahn W., Kahn R.S., Schnack H.G. Classification of schizophrenia patients and healthy controls from structural MRI scans in two large independent samples. Neuroimage. 2012;61:606–612. doi: 10.1016/j.neuroimage.2012.03.079. PubMed DOI

Gould I.C., Shepherd A.M., Laurens K.R., Cairns M.J., Carr V.J., Green M.J. Multivariate neuroanatomical classification of cognitive subtypes in schizophrenia: A support vector machine learning approach. Neuroimage Clin. 2014;6:229–236. doi: 10.1016/j.nicl.2014.09.009. PubMed DOI PMC

Vyškovský R., Schwarz D., Janoušová E., Kašpárek T. Random subspace ensemble artificial neural networks for first-episode schizophrenia classification; Proceedings of the 2016 Federated Conference on Computer Science and Information Systems (FedCSIS); Gdansk, Poland. 11–14 September 2016; pp. 317–321.

Vyškovský R., Schwarz D., Kašpárek T. Brain Morphometry Methods for Feature Extraction in Random Subspace Ensemble Neural Network Classification of First-Episode Schizophrenia. Neural Comput. 2019;31:897–918. doi: 10.1162/neco_a_01180. PubMed DOI

Savio A., Garcia-Sebastian M., Hernández C., Graña M., Villanúa J. Classification Results of Artificial Neural Networks for Alzheimer’s Disease Detection. In: Corchado E., Yin H., editors. Intelligent Data Engineering and Automated Learning-IDEAL 2009. Springer; Berlin/Heidelberg, Germany: 2009. [(accessed on 20 March 2015)]. pp. 641–648. Available online: http://link.springer.com/chapter/10.1007/978-3-642-04394-9_78. DOI

Huang C., Yan B., Jiang H., Wang D. Combining Voxel-based Morphometry with Artifical Neural Network Theory in the Application Research of Diagnosing Alzheimer’s Disease; Proceedings of the 2008 International Conference on BioMedical Engineering and Informatics; Sanya, China. 27–30 May 2008; pp. 250–254. DOI

Rosenblatt F. The perceptron: A probabilistic model for information storage and organization in the brain. Psychol. Rev. 1958;65:386–408. doi: 10.1037/h0042519. PubMed DOI

Rumelhart D.E., Hinton G.E., Williams R.J. Learning Internal Representations by Error Propagation. Volume 1. MIT Press; Cambridge, MA, USA: 1986. pp. 318–362. Parallel Distributed Processing: Explorations in the Microstructure of Cognition.

Ballard D.H. Modular learning in neural networks; Proceedings of the Sixth National Conference on Artificial Intelligence-Seattle; Washington, DC, USA. 13–17 July 1987; pp. 279–284.

Tan C.C., Eswaran C. Using Autoencoders for Mammogram Compression. J. Med. Syst. 2011;35:49–58. doi: 10.1007/s10916-009-9340-3. PubMed DOI

Zeng L.-L., Wang H., Hu P., Yang B., Pu W., Shen H., Chen X., Liu Z., Yin H., Tan Q., et al. Multi-Site Diagnostic Classification of Schizophrenia Using Discriminant Deep Learning with Functional Connectivity MRI. EBioMedicine. 2018;30:74–85. doi: 10.1016/j.ebiom.2018.03.017. PubMed DOI PMC

Moussavi-Khalkhali A., Jamshidi M., Wijemanne S. Feature Fusion for Denoising and Sparse Autoencoders: Application to Neuroimaging Data; Proceedings of the 2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA); Anaheim, CA, USA. 18–20 December 2016; pp. 605–610. DOI

Nayak D.R., Dash R., Majhi B., Pachori R.B., Zhang Y. A deep stacked random vector functional link network autoencoder for diagnosis of brain abnormalities and breast cancer. Biomed. Signal Process. Control. 2020;58:101860. doi: 10.1016/j.bspc.2020.101860. DOI

Mendoza-Léon R., Puentes J., Uriza L.F., Hoyos M.H. Single-slice Alzheimer’s disease classification and disease regional analysis with Supervised Switching Autoencoders. Comput. Biol. Med. 2020;116:103527. doi: 10.1016/j.compbiomed.2019.103527. PubMed DOI

Mallick P.K., Ryu S.H., Satapathy S.K., Mishra S., Nguyen G.N., Tiwari P. Brain MRI Image Classification for Cancer Detection Using Deep Wavelet Autoencoder-Based Deep Neural Network. IEEE Access. 2019;7:46278–46287. doi: 10.1109/ACCESS.2019.2902252. DOI

Pinaya W.H.L., Mechelli A., Sato J.R. Using deep autoencoders to identify abnormal brain structural patterns in neuropsychiatric disorders: A large-scale multi-sample study. Hum. Brain Mapp. 2019;40:944–954. doi: 10.1002/hbm.24423. PubMed DOI PMC

Object Recognition with Gradient-Based Learning|SpringerLink. [(accessed on 29 October 2021)]. Available online: https://link.springer.com/chapter/10.1007/3-540-46805-6_19. DOI

Fukushima K. Neocognitron: A self organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biol. Cybern. 1980;36:193–202. doi: 10.1007/BF00344251. PubMed DOI

Wang Z., Sun Y., Shen Q., Cao L. Dilated 3D Convolutional Neural Networks for Brain MRI Data Classification. IEEE Access. 2019;7:134388–134398. doi: 10.1109/ACCESS.2019.2941912. DOI

Hu M., Sim K., Zhou J.H., Jiang X., Guan C. Brain MRI-based 3D Convolutional Neural Networks for Classification of Schizophrenia and Controls. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. 2020;2020:1742–1745. doi: 10.1109/EMBC44109.2020.9176610. PubMed DOI

Hu M., Qian X., Liu S., Koh A.J., Sim K., Jiang X., Guan C., Zhou J.H. Structural and diffusion MRI based schizophrenia classification using 2D pretrained and 3D naive Convolutional Neural Networks. Schizophr. Res. 2021 doi: 10.1016/j.schres.2021.06.011. in press. PubMed DOI

Oh J., Oh B.-L., Lee K.-U., Chae J.-H., Yun K. Identifying Schizophrenia Using Structural MRI with a Deep Learning Algorithm. Front. Psychiatry. 2020;11:16. doi: 10.3389/fpsyt.2020.00016. PubMed DOI PMC

Campese S., Lauriola I., Scarpazza C., Sartori G., Aiolli F. Recent Advances in Big Data and Deep Learning. Springer; Cham, Switzerland: 2020. Psychiatric Disorders Classification with 3D Convolutional Neural Networks; pp. 48–57. DOI

Oh K., Kim W., Shen G., Piao Y., Kang N.-I., Oh I.-S., Chung Y.C. Classification of schizophrenia and normal controls using 3D convolutional neural network and outcome visualization. Schizophr. Res. 2019;212:186–195. doi: 10.1016/j.schres.2019.07.034. PubMed DOI

Good C.D., Johnsrude I.S., Ashburner J., Henson R.N., Friston K.J., Frackowiak R.S. A voxel-based morphometric study of ageing in 465 normal adult human brains. Pt 1Neuroimage. 2001;14:21–36. doi: 10.1006/nimg.2001.0786. PubMed DOI

Janousova E., Montana G., Kasparek T., Schwarz D. Supervised, Multivariate, Whole-Brain Reduction Did Not Help to Achieve High Classification Performance in Schizophrenia Research. Front. Neurosci. 2016;10:392. doi: 10.3389/fnins.2016.00392. PubMed DOI PMC

Kasparek T., Thomaz C., Sato J.R., Schwarz D., Janousova E., Marecek R., Prikryl R., Vanicek J., Fujita A., Ceskova E. Maximum-uncertainty linear discrimination analysis of first-episode schizophrenia subjects. Psychiatry Res. Neuroimaging. 2011;191:174–181. doi: 10.1016/j.pscychresns.2010.09.016. PubMed DOI

Ashburner J., Friston K.J. Unified segmentation. NeuroImage. 2005;26:839–851. doi: 10.1016/j.neuroimage.2005.02.018. PubMed DOI

Schwarz D., Kasparek T., Provaznik I., Jarkovsky J. A deformable registration method for automated morphometry of MRI brain images in neuropsychiatric research. IEEE Trans. Med. Imaging. 2007;26:452–461. doi: 10.1109/TMI.2007.892512. PubMed DOI

Ioffe S., Szegedy C. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. [(accessed on 30 October 2021)];arXiv. 2015 Available online: http://arxiv.org/abs/1502.03167.1502.03167