O-052 USING COMPUTER ASSISTED DETECTION TO PROMOTE HEALTH EQUITY THROUGH HIGH-VOLUME IDENTIFICATION OF SILICOSIS AND TUBERCULOSIS IN EX-MINERS: A BIO-ETHICS APPROACH
Bibliographic record
Abstract
Abstract Introduction In Southern Africa, a powerful legacy of social injustice has been the prevalence of occupational lung disease, particularly silicosis and tuberculosis (TB) in those who worked in South Africa’s gold mines. Although Artificial Intelligence (AI) is being increasingly applied for healthcare purposes, distrust about introducing “disruptive” technologies persists. Intrinsic and contextual factors also influence where and how such innovations are initiated. These require careful scrutiny to ensure that health equity is promoted. Methods We describe and appraise an AI application currently being developed, specifically the use of computer assisted detection (CAD) for TB and/or silicosis on chest x-rays, to support more efficient and equitable adjudication of compensation claims from former miners in southern Africa. Using a bio-ethical lens that considers the principles of beneficence, non-maleficence, autonomy and justice and adds explicability as a core principle, this study focuses on the apprehensions of users and stakeholders. Results Issues of concern include funding a sustainable health service delivery model in which CAD can be incorporated, CAD accuracy, possible biases in training of CAD systems, data privacy, impact on human skill development, transparency and accountability in CAD use, as well as intellectual property ownership. Discussion This paper discusses ways in which each of these potential obstacles to successful use of CAD could be mitigated. Conclusion From the outset, efforts to overcome technical implementation challenges must be considered to ensure ethical use. It is timely to take stock of barriers that might undermine the advancement of AI innovation on behalf of those who have been socially marginalized.
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 | 0.030 | 0.030 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.005 | 0.011 |
| Scholarly communication | 0.010 | 0.003 |
| Open science | 0.001 | 0.006 |
| Research integrity | 0.005 | 0.003 |
| Insufficient payload (model declined to judge) | 0.006 | 0.001 |
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".