PA-362 Towards explainable AI-based decision support in predicting SARS-CoV-2 breakthrough infections in a SSA context
Bibliographic record
Abstract
Background Vaccinated persons are still prone to SARS-CoV-2 breakthrough infection since vaccines do not offer 100% protection. Thus, quick decision-making on identifying high-risk persons prone to COVID-19 breakthrough infection is essential for effective medical care and cost saving. We explore how one can use Explainable Artificial Intelligence (XAI) to create a decision-making tool that can aid medical professionals in detecting patients prone of SARS-CoV-2 breakthrough in South Africa and beyond. Methods A dataset obtained from an intervention study on volunteers with cardiovascular disease (CVD) risk factors conducted in Cape Town, South Africa, comprising symptoms and feedback from 257 persons — 203 were vaccinated and 54 not — was used for the investigation. Two machine learning algorithms: Deep Multilayer Perceptron (Deep MLP) and the XGBoost classifier were trained on the dataset. The Shapley Additive Explanations (SHAP) was used to investigate the most critical variables influencing breakthrough infection from the ML models’ results. Lastly, a decision-support tool for detecting patients prone to breakthrough infection that leverages the ML model with the best results was created. Results The results show that the XGBoost model performed better (F1= 0.86; AUC = 0.74; G-Mean=0.71; MCC=0.49). Body temperature, total cholesterol, glucose level, blood pressure, waist circumference, body weight, body mass index (BMI), haemoglobin level, and physical activity per week are the most critical variables influencing breakthrough infection. Conclusion We established threshold values for each of them so that we could classify every new value as either high or low, and used these to construct an XAI model that combines machine learning and rule-based reasoning to predict if a patient is prone to breakthrough and provide a rationale/justification for the prediction made. Funding: This research was partially supported by a grant from the South African Medical Research Council (SAMRC).
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 | 0.002 | 0.010 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".