Clinical, Brain, and Multilevel Clustering in Early Psychosis and Affective Stages

Abstract

IMPORTANCEApproaches are needed to stratify individuals in early psychosis stages beyond positive symptom severity to investigate specificity related to affective and normative variation and to validate solutions with premorbid, longitudinal, and genetic risk measures. OBJECTIVE To use machine learning techniques to cluster, compare, and combine subgroup solutions using clinical and brain structural imaging data from early psychosis and depression stages. DESIGN, SETTING, AND PARTICIPANTSA multisite, naturalistic, longitudinal cohort study (10 sites in 5 European countries; including major follow-up intervals at 9 and 18 months) with a referred patient sample of those with clinical high risk for psychosis (CHR-P), recent-onset psychosis (ROP), recent-onset depression (ROD), and healthy controls were

Ensemble clustering: A practical tutorial

“…Meaningfulness in clusters can be examined using external variables not included in the clustering process (e.g., sociodemographic factors, and other associated outcomes such as diagnoses and long-term trajectories). To confirm that that the cluster solution did not occur by chance, a permutation-based non-parametric test can be incorporated, see details illustrated by Dwyer et al (2022).…”

“…Since the cophenetic correlation coefficient, CDF and PAC methods all yield a single set of estimations for the optimal 𝑘 without accounting for variabilities in these metrics, incorporating these methods into a nested cross-validation framework can that the decision-making process is not impacted by random fluctuations in the data. This approach is detailed in the examples provided Dwyer et al (2022).…”

Differentiating Biomarker Features and Familial Characteristics of B-SNIP Psychosis Biotypes

“…With our refined biologically based sub-groups, we have attempted to capture homogenous disease groups with common biological dysfunctions. This is an approach we will continue to refine and is being more widely adopted in biological psychiatry research [ 53 – 56 ]. Critically, it deserves thorough and rigorous testing.…”

Multivariate brain-behaviour associations in psychiatric disorders

“…Multi-level clustering [ 99 ] is another promising approach whereby patients are grouped based on multiple domains, e.g ., brain and behaviour, simultaneously. For example, Dwyer et al found a subset of early psychosis patients with low functioning and reduced brain volume [ 22 ]. Future studies could investigate how membership of different brain-behaviour subtypes is associated with prognosis.…”

“…The former has enjoyed considerable interest in the last two decades, with a wealth of studies investigating how neural features can predict a range of univariate psychiatric outcomes such as functioning [ 11 , 12 ], diagnosis [ 13 – 15 ] and response to treatment [ 16 , 17 ] using popular approaches such as support vector machine (SVM) [ 18 – 20 ]. Within the second group, there are many approaches that could be used in principle, such as independent component analysis (ICA) and its variants (e.g., parallel ICA, joint ICA or linked ICA) [ 21 ], multilevel clustering [ 22 ], canonical correlation analysis (CCA) [ 23 ] and partial least squares [ 24 ] (PLS). The latter two emerge as the most established and popular techniques in brain-behaviour studies, as evidenced by several recent studies in the general population [ 25 ] and tutorials tailored to brain-behaviour investigations [ 26 , 27 ].…”