Construction of a Social-Media Based Clinical Database—Roadmap, Challenges, and Feasibility for ADHD Recognition
Bibliographic record
Abstract
The shortage of available high-quality clinical databases restricts medical diagnostics downstream. Clinical databases are often limited to controlled non-natural environments, they are restricted due to privacy limitations and require complex scoring procedures that ultimately result in rater bias. Social media includes massive amounts of information on subjects through streams of text, audio, and video data that are accessible and currently underutilized for medical research. In this work we propose a method for utilizing this information by constructing databases for medical condition assessment. To this end we have created SMDC (Social Medical Data Constructor), a utility based on medical expert requirements. Data Features and non-confidential demographic information are extracted online, and labels are derived using data mining techniques. We examine the feasibility of the suggested technology with ADHD recognition from a database extracted from YouTube clips using the self-tagging as ADHD labels. The database maintains privacy and copy write limitations, and no personally identifying information is collected. To validate the database, we show a high correlation of the derived model’s predictions with expert labeling (r =0.68) and compatibility of six known ADHD motor biomarker features of hyperactivity to the ones derived using our database. Furthermore, we extracted from the video clips kinematics features and reached ADHD recognition accuracy of 83%, and 81%, for female sand males respectively. The suggested technology has the potential to assess natural real-life behavioral properties of the medical condition, and may further be of use as a pre-training phase allowing fine tuning on actual clinical data with minimal data requirements.
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 | 0.007 | 0.024 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".