P2–263: Detection and classification of dementias using generalized complexity estimates
Bibliographic record
Abstract
Although quantitative neuroimaging studies of neurodegenerative diseases have identified several promising markers for detecting and classifying dementias, their usefulness for diagnosis is hampered by large numbers of structural, physiological and neurochemical outcomes that are hard to summarize and interpret. In particular measurements of cortical thinning, brain volume loss and cerebral hypoperfusion and hypometabolism depend critically on the choice of processing parameters, exacerbating the problem. Generalized complexity estimates (GCE) constitute sensitive, interpretable summary information that can be extracted from images and are easy to generate. We hypothesized that GCE of structural magnetic resonance images (MRI) produce robust, sensitive separation between cognitive normal (CN), Alzheimer's disease (AD) and Frontal Temporal Dementia (FTD) subjects. An initial test applied GCE to structural, segmented MRI from 50 ICBM database subjects, 25 with the cortex artificially thinned in the right superior temporal gyrus (RSTG). For tests on experimental data, GCE were also obtained from structural MRI from 21 AD, 20 FTD, and 25 CN subjects. Statistical significance was estimated using MANOVA and linear discriminant analysis, and specificity and sensitivity were estimated using 10 fold, 10 times cross validation. MANOVA analysis of the GCE for cortical thinning showed significant separations between the populations for all regions containing the RSTG (with p = 4.1e–09, sensitivity of .91 and specificity of .85 for the RSTG itself). MANOVA for the experimental data also produced a highly significant (p = 2.4e–08) separation between AD, FTD, and CN subjects, using GCE for the hippocampus, subiculum, and putamen. The LDA for GCE from 13 brain regions achieved a classification accuracy of 0.96. Projection onto the first 2 linear discriminants is shown in the figure. Preliminary studies indicate that easy to apply and interpret GCE summary information provides robust classifications between AD, FTD, and CN subjects, and could be useful in classification and early detection of dementias. GCE should in fact be of general utility for producing sensitive and interpretable summary information from multimodal imaging studies in a variety of contexts.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".