Domain adaptive FL for edge-enabled privacy-preserving MRI analysis
Bibliographic record
Abstract
Data heterogeneity, privacy leakage challenges, the ineffectiveness of conventional collaborative learning techniques, and unresolved managing non-IID data distributions are some of the major obstacles to implementing artificial intelligence (AI) in healthcare. Federated learning (FL) frameworks frequently have trouble distinguishing between privacy protection and model accuracy, especially when used for delicate medical imaging applications. This study presents a novel framework that synergizes federated learning (FL) with edge computing to address these issues while safeguarding patient privacy. Our proposed Domain Adaptive Federated (DAD) learning approach effectively mitigates both inter-client and intra-client data heterogeneity, enabling collaborative model training across diverse medical imaging modalities (MRI, CT, PET) through cross-domain adaptation. Experimental evaluations on MRI brain segmentation datasets demonstrate the superior performance of DAD compared to traditional FL methods, as evidenced by significant improvements in F1-score (96.3), sensitivity (96.0), specificity (97.1), and AUC (96.7). This enhanced accuracy and robustness in handling heterogeneous and privacy-sensitive data render DAD an ideal candidate for privacy-preserving AI in consumer healthcare. By pioneering innovative strategies for collaborative model training and data privacy, this research contributes to the emerging field of edge intelligence, paving the way for improved patient outcomes while adhering to stringent confidentiality and ethical mandates.
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 | 0.003 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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".