Automated word class and person usage metrics as markers of Alzheimer’s and frontotemporal dementia
Bibliographic record
Abstract
Abstract Background Dementia impacts the way individuals perceive and describe everyday events. Alzheimer's disease (AD) notably affects processing of entities manifested by nouns, while behavioral variant frontotemporal dementia (bvFTD) often presents a detached, third‐person perspective. Yet, the potential of natural language processing tools (NLP) to detect these variations in spontaneous speech remains explored. To tackle this gap, we analyzed both patterns via automated discourse‐level metrics in individuals with AD and bvFTD, contrasting them with healthy controls (HCs). Methods Persons with AD (n = 21), bvFTD (n = 21), as well as HCs (n = 21), narrated a typical day of their lives. We analyzed the frequency of nouns and verbs, along with first‐ or third‐person usage, via part‐of‐speech and morphological tagging, respectively. Inferential statistics and machine learning were used to examine whether these features were useful for discriminating patients from HCs at both the group and the subject level. We further evaluated whether such features correlated with cognitive symptom severity, as captured through the Montreal Cognitive Assessment (MoCA). Results Compared with HCs, AD (but not bvFTD) patients exhibited a lower proportion of nouns, without differences in verb ratio. Conversely, persons with bvFTD (but not those with AD) had a greater proportion of third‐person markers and a reduced proportion of first‐person markers. Machine learning analyses showed that these features robustly identified individuals within each group (AUCs = 0.75). No linguistic feature was significantly correlated with MoCA scores in either patient group. Conclusions Spontaneous daily narratives offer distinct markers for AD and bvFTD, detectable through automated analysis. Focusing on specific linguistic attributes relevant to each type of dementia not only aids in understanding but also enhances diagnosis and tracking of these conditions. Overall, our findings attest to the relevance of NLP tools as a viable, cost‐effective means to identify scalable dementia markers.
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 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.004 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".