SU‐E‐P‐43: A Knowledge Based Approach to Guidelines for Software Safety
Bibliographic record
Abstract
Purpose: In the fall of 2012, a survey was distributed to medical physicists across Canada. The survey asked the respondents to comment on various aspects of software development and use in their clinic. The survey revealed that most centers employ locally produced (in‐house) software of some kind. The respondents also indicated an interest in having software guidelines, but cautioned that the realities of cancer clinics include variations, that preclude a simple solution. Traditional guidelines typically involve periodically repeating a set of prescribed tests with defined tolerance limits. However, applying a similar formula to software is problematic since it assumes that the users have a perfect knowledge of how and when to apply the software and that if the software operates correctly under one set of conditions it will operate correctly under all conditions Methods: In the approach presented here the personnel involved with the software are included as an integral part of the system. Activities performed to improve the safety of the software are done with both software and people in mind. A learning oriented approach is taken, following the premise that the best approach to safety is increasing the understanding of those associated with the use or development of the software. Results: The software guidance document is organized by areas of knowledge related to use and development of software. The categories include: knowledge of the underlying algorithm and its limitations; knowledge of the operation of the software, such as input values, parameters, error messages, and interpretation of output; and knowledge of the environment for the software including both data and users. Conclusion: We propose a new approach to developing guidelines which is based on acquiring knowledge‐rather than performing tests. The ultimate goal is to provide robust software guidelines which will be practical and effective.
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 | 0.025 | 0.046 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.006 | 0.004 |
| Science and technology studies | 0.004 | 0.006 |
| Scholarly communication | 0.007 | 0.010 |
| Open science | 0.004 | 0.009 |
| Research integrity | 0.007 | 0.005 |
| Insufficient payload (model declined to judge) | 0.009 | 0.003 |
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".