S7. MORPHOLOGICAL PROFILING OF SCHIZOPHRENIA: CLUSTER ANALYSIS OF MRI-BASED CORTICAL THICKNESS DATA
Bibliographic record
Abstract
The clinical diagnosis of schizophrenia is suspected to include several distinct subgroups of patients, but reliable neurobiological boundaries to differentiate the subgroups remain elusive. These unknown subgroups increase the variance of biological measures within the clinically identified patient group, deflating the group-level estimates of causal factors and treatment effects. Prior studies seeking homogeneous subgroups of schizophrenia based on brain-based measures have not found consistent solutions. A major limitation in prior studies is the assumption that healthy controls form a relatively homogeneous group, that deviates biologically from the patient subgroups. As a result, cluster solutions have been generally sought only within patient samples, without pooling the patient and control data. In the current study, we assessed whether the regional values of cortical thickness estimated from structural MRI are sufficiently sensitive to identify subgroups of patients and healthy controls. We used high resolution (3 Tesla) imaging in 179 patients with schizophrenia and 77 healthy controls, to investigate possible subtypes of schizophrenia. K-means algorithm was applied to perform clustering analysis on cortical thickness data from 68 regions in Desikan-Kiliany Atlas using Freesurfer software, and gap statistics was used to find best cluster solution. General linear models were used to compare cortical thickness, cognitive performance and symptom severity among the identified clusters. A 3-cluster solution provided the most optimal clustering, with the first cluster (C1) comprised almost entirely of patients, while the other 2 clusters (C2 and C3) including a substantial number of patients as well as control subjects. There was no significant difference in cognitive performance among the 3 schizophrenia subtypes. C1 was the most morphologically impoverished group with significantly thin cortex in multiple brain regions but had no more symptom/cognitive burden than the other subtypes. C2 was an intermediate group with significant thinning in selected brain regions, with higher burden of negative symptoms. C3 was the morphologically most intact subgroup with a cortical thickness profile like healthy controls, despite having more severe delusions. In addition, C1 also had higher duration of exposure to medication among the 3 patient groups, with longer illness duration. We report 3 major findings: 1) 3 distinct morphological profiles are observed in schizophrenia 2) A large number of patients with schizophrenia have the cortical morphological profiles of apparently normal healthy controls 3) cortical thickness profiles do not map well to cognitive and symptomatic profiles in schizophrenia. Interestingly, we observed a pattern of morphological preservation among patients with higher levels of delusion. We provide evidence for the presence of morphological subgroups of schizophrenia, the delineation of which may help stratifying patients for future prognostic studies.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.004 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.074 | 0.011 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".