Back to feed
arXiv cs.LG
arXiv cs.LG
7/28/2026
QFedPolyp: A Communication- and Inference-Efficient Federated Learning Framework for Polyp Segmentation

QFedPolyp: A Communication- and Inference-Efficient Federated Learning Framework for Polyp Segmentation

Short summary

QFedPolyp is a federated learning framework for collaborative polyp segmentation that combines quantization-aware training with low-precision model communication, enabling hospitals to train locally on private data while transmitting only quantized parameters. On Kvasir-SEG and CVC-ClinicVideoDB, full-precision federated training achieves Dice scores of 0.910 and 0.930 respectively, while 8-bit communication reduces transmission cost ~4x with competitive accuracy. Quantized models also achieve up to 1.5x faster inference, making them suitable for real-time clinical deployment.

  • •Combines quantization-aware training with federated learning for polyp segmentation
  • •8-bit communication cuts transmission cost ~4x while preserving accuracy
  • •Dice scores of 0.910 (Kvasir-SEG) and 0.930 (CVC-ClinicVideoDB) with up to 1.5x faster inference

Generated with AI, which can make mistakes.

#research-breakthrough#industry-adoption
Read full article at arXiv cs.LG

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

FedCC: A Low-Resource Federated Adaptation of Foundation Models for Robust Corpus Callosum localization in Fetal Ultrasound Images
arXiv cs.LG
Federated Learning for Object Detection: Enabling Collaborative Drone Learning Without Centralizing Data
arXiv cs.LG
A Drift Stable Quantum Federated Learning for Intelligent Services
arXiv cs.LG
HantaWatch: Federated Learning for Hantavirus Genomic Surveillance
arXiv cs.LG