When Collaborative Federated Learning Meets Blockchain to Preserve Privacy in Healthcare
Bibliographic record
Abstract
Data-driven Machine and Deep Learning (ML/DL) is an emerging approach that uses medical data to build robust and accurate ML/DL models that can improve clinical decisions in some critical tasks ($e.g.,$cancer diagnosis). However, ML/DL-based healthcare models still suffer from poor adoption due to the lack of realistic and recent medical data. The privacy nature of these medical datasets makes it difficult for clinicians and healthcare service providers, to share their sensitive data ($i.e.,$Patient Health Records (PHR)). Thus, privacy-aware collaboration among clinicians and healthcare service providers is expected to become essential to build robust healthcare applications supported by next-generation networking (NGN) technologies, including Beyond sixth-generation (B6G) networks. In this paper, we design a new framework, called HealthFed, that leverages Federated Learning (FL) and blockchain technologies to enable privacy-preserving and distributed learning among multiple clinician collaborators. Specifically, HealthFed enables several distributed SDN-based domains, clinician collaborators, to securely collaborate in order to build robust healthcare ML-based models, while ensuring the privacy of each clinician participant. In addition, HealthFed ensures a secure aggregation of local model updates by leveraging a secure multiparty computation scheme ($i.e.,$Secure Multiparty Computation (SMPC)). Furthermore, we design a novel blockchain-based scheme to facilitate/maintain the collaboration among clinician collaborators, in a fully decentralized, trustworthy, and flexible way. We conduct several experiments to evaluate HealthFed; in-depth experiments results using public Breast Cancer dataset show the efficiency of HealthFed, by not only ensuring the privacy of each collaborator's sensitive data, but also providing an accurate learning models, which makes HealthFed a promising framework for healthcare systems.
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 | 0.023 | 0.072 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.003 | 0.004 |
| Scholarly communication | 0.007 | 0.018 |
| Open science | 0.003 | 0.013 |
| Research integrity | 0.005 | 0.005 |
| Insufficient payload (model declined to judge) | 0.014 | 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".