Adaptive Asynchronous Split Federated Learning for Medical Image Segmentation
Bibliographic record
Abstract
Split federated (SplitFed) learning offers promise for collaborative machine learning across decentralized and resource-constrained clients (edge devices, nodes, or organizations) in various applications, including healthcare. However, real-world challenges arise in heterogeneous environments where clients experience communication-related data losses, differ in computational capabilities, and have varying local dataset sizes. In this paper, we introduce adaptive asynchronous split federated learning (AASFL), an innovative training scheme to address such challenges. AASFL blends asynchronous SplitFed learning with client-level adaptability, acknowledging varying client capabilities. In AASFL, each client independently adapts its learning rate and the number of local training epochs based on its local training speed, dataset size, and packet loss probability. To demonstrate the effectiveness of AASFL, we implement and evaluate it on a SplitFed network architecture for medical image segmentation. Moreover, we demonstrate the validity of AASFL by emphasizing the necessity of training each client in the collaborative network and revealing drawbacks of client selection. The results on two public datasets indicate that it greatly enhances global model performance, leading to more accurate segmentation results. We also statistically show that AASFL outperforms the standard SplitFed. Furthermore, we provide a theoretical convergence analysis of AASFL. To the best of our knowledge, this is the first analysis of SplitFed in an adaptive, asynchronous setting. The proposed AASFL training scheme offers a promising avenue for improving medical image segmentation tasks in practical decentralized learning environments.
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 | 0.005 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".