The Impact of Adversarial Attacks on Medical Imaging AI Systems
Bibliographic record
Abstract
Bad actors threaten medical imaging AI systems' dependability and safety, affecting patient care and diagnosis accuracy. A novel defense against this expanding threat is adversarial defense via ensemble integration. Explainable Feature-Based Defense (XFBD), Adversarial Training with Transfer Learning (ATTL), and Robust Classifier Augmentation (RCA) are three innovative techniques that have never been combined. RCA enhances training with controlled aggression. This allows the model to distinguish authentic medical imaging data from altered sources. ATTL makes transfer-learning-taught models more resistant to topic-specific hostile approaches. XFBD simplifies the defensive process, helping us comprehend how the model picks and fights different methods. A comparison indicates that ADEI always outperforms tried-and-true approaches. ADEI outperforms typical approaches in accuracy, sensitivity, precision, stability, interpretability, private protection, and computing cost. Strong safeguards are needed as AI-powered healthcare research becomes increasingly popular. As a lighthouse, ADEI protects everything from attack. The math behind each section and how they work together to strengthen the system make it valuable. ADEI advances AI-assisted healthcare's future. Testing instruments must be accurate and dependable in this industry.
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 | 0.006 | 0.027 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 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".