PT  - JOURNAL ARTICLE
AU  - Junhao Wen
AU  - Erdem Varol
AU  - Aristeidis Sotiras
AU  - Zhijian Yang
AU  - Ganesh B. Chand
AU  - Guray Erus
AU  - Haochang Shou
AU  - Gyujoon Hwang
AU  - for the Alzheimer’s Disease Neuroimaging Initiative
AU  - Christos Davatzikos
TI  - Multi-scale semi-supervised clustering of brain images: deriving disease subtypes
AID  - 10.1101/2021.04.19.440501
DP  - 2021 Jan 01
TA  - bioRxiv
PG  - 2021.04.19.440501
4099  - http://biorxiv.org/content/early/2021/04/20/2021.04.19.440501.short
4100  - http://biorxiv.org/content/early/2021/04/20/2021.04.19.440501.full
AB  - Disease heterogeneity is a significant obstacle to understanding pathological processes and delivering precision diagnostics and treatment. Clustering methods have gained popularity in stratifying patients into subpopulations (i.e., subtypes) of brain diseases using imaging data. However, unsupervised clustering approaches are often confounded by anatomical and functional variations not related to a disease or pathology of interest. Semi-supervised clustering techniques have been proposed to overcome this and, therefore, capture disease-specific patterns more effectively. An additional limitation of both unsupervised and semi-supervised conventional machine learning methods is that they typically model, learn and infer from data at a basis of feature sets pre-defined at a fixed scale or scales (e.g, an atlas-based regions of interest). Herein we propose a novel method, “Multi-scAle heteroGeneity analysIs and Clustering” (MAGIC), to depict the multi-scale presentation of disease heterogeneity, which builds on a previously proposed semi-supervised clustering method, HYDRA. It derives multi-scale and clinically interpretable feature representations and exploits a double-cyclic optimization procedure to drive inter-scale-consistent disease subtypes or neuroanatomical dimensions effectively. More importantly, to fill in the gap of understanding under what conditions the clustering model can estimate true heterogeneity related to diseases, we conducted extensive and systematic semi-simulated experiments to evaluate the proposed method on a sizeable healthy control sample from the UK Biobank (N=4403). We then applied MAGIC to real imaging data of Alzheimer’s disease (ADNI, N=1728) to demonstrate its potential and challenges in dissecting the neuroanatomical heterogeneity of brain diseases. Taken together, we aim to provide guidelines on when such analyses can succeed or should be taken with caution. The code of the proposed method is publicly available at https://github.com/anbai106/MAGIC.HighlightsWe propose a novel multi-scale semi-supervised clustering method, termed MAGIC, aiming at disentangling the heterogeneity of brain diseases.We perform extensive experiments on large control samples (UK Biobank, N=4403) to precisely quantify performance under various conditions, including varying degrees of brain atrophy, different levels of heterogeneity, overlapping disease subtypes, class imbalance, and varying sample sizes.We demonstrate the strengths of MAGIC relative to other standard clustering methods.We apply MAGIC to MCI and Alzheimer’s disease datasets (ADNI, N=1728) to dissect neuroanatomical heterogeneity in AD and its prodromal stages.Competing Interest StatementThe authors have declared no competing interest.ADAlzheimer’s diseaseGMGrey matterMCIMild cognitive impairmentCNHealthy controlMLMachine learningARIAdjusted Rand indexASLAtrophy strength levelPTPatientsSubSubtypeCVCross-validationQCQuality controlARI_CVARIs during CVARI_GTARIs for ground truthT1w MRIT1-weighted MRIMRIMagnetic resonance imagingNMFNon-negative matrix factorizationVBAVoxel-based analysisMVPAMultivariate pattern analysisSVMSupport vector machinekNumber of clusters/subtypesMNumber of componentsNNumber of subjectsDNumber of voxelsXInput matrixCComponent matrixLLoading coefficient matrixyInput labelwSVM weightbSVM biasSSubtype membership matrixS*Final subtype membership matrix after consensus clusteringiIndex of number of subjectsjIndex of number of clusters/subtypesqIndex of number of blocks in cyclic optimizationQTotal number blocks in cyclic optimization