Design and development of the intelligent voice recognition‐based cognitive assessment WeChat mini‐program
Bibliographic record
Abstract
Abstract Background Early screening of mild cognitive impairment (MCI) is crucial for the elderly to access formal diagnosis and intervention of Alzheimer’s Disease (AD) and related dementias (ADRD) at early stages. Owing to the rapid development of artificial intelligence (AI) and voice recognition technology, the detection method using speech patterns as digital biomarkers collected by digital consumer devices can accommodate a huge demand for early and large‐scale detection. We here reported the design and development of the Shanghai Cognitive Screening (SCS) mini‐program, the first intelligent voice recognition‐based cognitive assessment WeChat mini‐program in China. Method The SCS mini‐program was developed based on the cognitive paradigm and classic pencil‐and‐paper neuropsychological assessments (e.g. Montreal cognitive assessment basic [MoCA‐B]). Through Automatic Speech Recognition (ASR), users speak to complete tasks including Naming, Immediate Recall, Delayed Recall, Symbol Digit Modalities Test (SDMT), Delayed RecallII, and Recognition (Figure 1). The acoustic, phonetics, and linguistic speech pattern variables would also be extracted and processed as digital biomarkers. The SCS intelligent prediction model enrolls speech pattern digital biomarkers, individual task scores, and time use variables with age, gender, and education level as linking variables, which enables to grade the user’s global cognitive function and individual cognitive domains (Figure 2). Result After completing all tasks in the SCS, users can immediately read the comprehensive results which are divided into 4 categories corresponding to 4 grading intervals: “healthy” (cognitively normal), “low risk” (suspected SCD), “medium‐high risk” (suspected MCI), and “high risk” (suspected AD). We also use a radar chart with scores of cognitive domains (i.e., memory, attention, visual perception, language, and executive functions), a memory trend graph (i.e., immediate and delayed recalls), an attention persistence graph, and related advice in the automatically generated report (Figure 3). Conclusion The SCS mini‐program is more efficient and economical compared to traditional paper‐based cognitive assessments. It is suitable for clinical cognitive assessment, remote self‐rating, and community screening for MCI and dementias among Chinese older adults and other ADRD high‐risk groups. Clinical trials are yet to be conducted to validate the accuracy and diagnostic value of the SCS WeChat mini‐program.
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 | 0.005 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 0.002 |
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".