Efficient Federated Aggregation Technique for Brain Tumor Classification Using Transfer Learning Approaches
Bibliographic record
Abstract
Federated learning (FL) is a decentralized machine learning technique enabling machine learning models to be trained on local datasets, preserving data privacy.Aggregation techniques play a vital role in FL since updates from multiple clients located at remote places are shared and combined to build a global model without directly accessing raw data.Most widely used averaging Algorithms such as FedAvg and FedAdam, are often difficult to tune and exhibit unfavorable convergence behavior when the clients participating in averaging vary.We propose a novel approach to allow a selected number of clients to participate in aggregation and hence speed up the convergence by increasing the performance of the machine learning model used to classify the brain tumor types, meningioma, glioma, no tumor, and pituitary utilizing a combination of Figshare, SARTAJ and Br35H dataset.We adopted an innovative approach which modifies the Xception model architecture optimized for Brain tumor classification We compared the performance of Xception, VGG19 and DenseNet201 pretrained classification models, where Xception model outperformed with a measured accuracy of 99.4.The proposed algorithm PC_FedAvg (Priority-based Client selection Federated Learning) is compared with existing FedAvg, FedAdam and FedProx, in terms of the number of communication rounds, along with the accuracy of the classification model and results show that PC_FedAvg has demonstrated improved performance than the selected approaches in terms of model accuracy, precision, and recall.PC_FedAvg gives the best results for the Xception model with a Precision of 90.52, Recall of 91.8, and accuracy of 91.6%.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".