Linguistic biomarkers of dementia in Italian patients living in Lombardy: insights from NLP analysis
Bibliographic record
Abstract
Nowadays, dementia poses a major challenge for healthcare services, with consequences on both the economic and the organizational front. The number of people affected by this disease is steadily increasing, and existing pharmacological and psycho-social therapies only aim to slow its progression. Therefore, a timely diagnosis is crucial for early intervention. For this reason, researchers from different disciplines are trying to find the “biomarkers” of dementia to obtain a detailed profiling of this disease and its etiology. In particular, great attention has been directed towards language, as it is one of the first cognitive domains affected by the pathology. The new frontier in the analysis of spoken language productions is the employment of Natural Language Processing (NLP) techniques and Artificial Intelligence (AI), as they enable an ecological and non-intrusive detection of dementia. This study aims at analyzing the speech of elderly individuals diagnosed with dementia and living in Lombardy (Italy) exploiting NLP techniques. A cohort of 8 participants was enrolled, consisting of 4 patients affected by dementia (i. e., Alzheimer’s disease or mixed dementia) and 4 healthy controls matched by age, level of education, and sex. Participants’ selection was made on four neuropsychological tests (i. e., MMSE – Mini-Mental State Examination, MoCA – Montreal Cognitive Assessment, phonemic and semantic fluences). The speech samples were collected through three elicitation tasks and subsequently manually transcribed using the ELAN software. A multidimensional parameter analysis was performed on the corpus obtained taking into consideration a set of 151 linguistic features. Finally, a statistical analysis was performed by comparing the pathological group and the control group. Results demonstrate the efficacy of computational linguistic analysis in differentiating one group from another. Moreover, given the peculiar sociolinguistic situation in Italy, the study confirms the importance of investigating differences related to diatopic variation in clinical populations.
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) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 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".