What Can We Learn from Quality Requirements in ISO/TS 82304-2 for Evaluating Conversational Agents in Healthcare?
Bibliographic record
Abstract
Evaluating conversational agents (CA) that are supposed to be applied in healthcare and ensuring their quality is essential to avoid patient harm. However, most researchers only study usability and use the CA in clinical trials before conducting such careful evaluation. In previous work, consensus on metrics for evaluating healthcare CA have been found. However, the metrics are still too generic to form an evaluation framework. In this work, we try to link the ISO technical specification ISO/TS 82304-2 Quality Requirements for Health and Wellness Apps to the set of metrics to come a step closer towards an evaluation framework. We identify three links between ISO requirements and the set of metrics, namely accessibility, usability, and security. Although the technical specification rather lists aspects to be considered during development instead of concrete metrics for studying the quality, we can link to some aspects that are also of interest for health CA evaluation. For example, measuring the readability for ensuring accessibility or implementing the Web Content Accessibility Guidelines are two aspects of relevance for health CA.
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 | 0.241 | 0.434 |
| Meta-epidemiology (narrow) | 0.004 | 0.002 |
| Meta-epidemiology (broad) | 0.004 | 0.004 |
| Bibliometrics | 0.014 | 0.008 |
| Science and technology studies | 0.003 | 0.010 |
| Scholarly communication | 0.024 | 0.025 |
| Open science | 0.007 | 0.008 |
| Research integrity | 0.009 | 0.011 |
| 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".