Subtyping mild cognitive impairment based on imaging and CSF biomarker levels
Bibliographic record
Abstract
Abstract Background Individuals with mild cognitive impairment (MCI) have variable clinical outcomes, with a proportion converting to dementia in a short follow‐up period. Subtyping MCI subjects based on biomarker levels can provide additional insights on why some individuals progress faster, aiding clinical treatment strategy and facilitating cohort enrichment for clinical trials. Method Data included 562 individuals with MCI from the Alzheimer's Disease Neuroimaging Initiative (ADNI) project with baseline MRI, CSF amyloid beta (Aβ), t‐tau, and p‐tau data and a minimum of one‐year follow‐up information on clinical diagnosis. White matter hyperintensities (WMHs) were segmented using a previously validated random forests classifier (Dadar et al. 2017). SNIPE (Scoring by Nonlocal Image Patch Estimator) was used to measure Alzheimer’s‐disease‐like atrophy patterns in the hippocampi (HC) and entorhinal cortex (EC) (Coupé et al. 2019). CSF Aβ, p‐tau and t‐tau levels were obtained from project files. Agglomerative hierarchical clustering as used to cluster the data based on baseline age, education, APOE4 status, CSF Aβ, p‐tau and t‐tau levels, WMH load, and HC and EC grading scores. We then investigated differences in baseline ADAS13 cognitive scores as well as rate of conversion to dementia between different subtypes. Result Figure 1 shows the clustering dendrogram, splitting the data into three subtypes of 152, 200, and 210 participants, respectively. Figure 2 compares biomarker values across subtypes. Baseline ADAS13 scores and conversion rates were significantly different across all subtypes (p<0.0001). Subtype 1 had the oldest subjects with largest WMH loads (p<0.0001), with a conversion rate of 30.2%. Subtype 2 had the highest conversion rate (70.0%), with significantly lower CSF Aβ and EC grading scores, and higher CSF p‐tau and t‐tau values (p<0.0001). Subtype 3 had the lowest conversion rate (15.2%), with significantly higher HC and EC grading scores and lower age and WMH loads than the other two subtypes (p<0.0001). Conclusion Biomarker‐based subtyping of MCI subjects led to three distinct groups with significantly different clinical outcomes: 1) older individuals with cerebrovascular pathology and moderate conversion rates, 2) individuals with abnormal CSF biomarker levels, hippocampal and entorhinal atrophy, and high conversion rates, and 3) younger individuals without abnormal biomarker levels and low conversion rates.
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 | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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".