mlsanityData to 14 Oct 2026

Impact of Labeling Inaccuracy and Image Noise on Tooth Segmentation in Panoramic Radiographs using Federated, Centralized and Local Learning

arXiv 2509.06553 ↗First posted 8 Sep 2025eess.IV cs.CV cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

Objectives: Federated learning (FL) may mitigate privacy constraints, heterogeneous data quality, and inconsistent labeling in dental diagnostic AI. We compared FL with centralized (CL) and local learning (LL) for tooth segmentation in panoramic radiographs across multiple data corruption scenarios. Methods: An Attention U-Net was trained on 2066 radiographs from six institutions across four settings: baseline (unaltered data); label manipulation (dilated/missing annotations); image-quality manipulation (additive Gaussian noise); and exclusion of a faulty client with corrupted data. FL was implemented via the Flower AI framework. Per-client training- and validation-loss trajectories were monitored for anomaly detection and a set of metrics (Dice, IoU, HD, HD95 and ASSD) was evaluated on a hold-out test set. From these metrics significance results were reported through Wilcoxon signed-rank test. CL and LL served as comparators. Results: Baseline: FL achieved a median Dice of 0.94889 (ASSD: 1.33229), slightly better than CL at 0.94706 (ASSD: 1.37074) and LL at 0.93557-0.94026 (ASSD: 1.51910-1.69777). Label manipulation: FL maintained the best median Dice score at 0.94884 (ASSD: 1.46487) versus CL's 0.94183 (ASSD: 1.75738) and LL's 0.93003-0.94026 (ASSD: 1.51910-2.11462). Image noise: FL led with Dice at 0.94853 (ASSD: 1.31088); CL scored 0.94787 (ASSD: 1.36131); LL ranged from 0.93179-0.94026 (ASSD: 1.51910-1.77350). Faulty-client exclusion: FL reached Dice at 0.94790 (ASSD: 1.33113) better than CL's 0.94550 (ASSD: 1.39318). Loss-curve monitoring reliably flagged the corrupted site. Conclusions: FL matches or exceeds CL and outperforms LL across corruption scenarios while preserving privacy. Per-client loss trajectories provide an effective anomaly-detection mechanism and support FL as a practical, privacy-preserving approach for scalable clinical AI deployment.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 11 of its 39 references that are in the map. First posted on arXiv 8 Sep 2025.

Builds on

11 of 39 references.

References · 11

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  3. 2019Federated Learning: Challenges, Methods, and Future Directionscited by 1,538
  4. 2020The Future of Digital Health with Federated Learningcited by 633
  5. 2015Federated Optimization:Distributed Optimization Beyond the Datacentercited by 195
  6. 2022FedDM: Iterative Distribution Matching for Communication-Efficient Federated Learningcited by 66
  7. 2022Rethinking Data Heterogeneity in Federated Learning: Introducing a New Notion and Standard Benchmarkscited by 24
  8. 2022FedMix: Mixed Supervised Federated Learning for Medical Image Segmentationcited by 23
  9. 2023FedA3I: Annotation Quality-Aware Aggregation for Federated Medical Image Segmentation against Heterogeneous Annotation Noisecited by 14
  10. 2024FedIA: Federated Medical Image Segmentation with Heterogeneous Annotation Completenesscited by 2
  11. 2024Supercharging Federated Learning with Flower and NVIDIA FLAREcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026DinoDental: Benchmarking DINOv3 as a Unified Vision Encoder for Dental Image Analysiscited by 2
  2. 2026Deep Learning-based Assessment of the Relation Between the Third Molar and Mandibular Canal on Panoramic Radiographs using Local, Centralized, and Federated Learningcited by 0
  3. 2026Topology-Constrained Quantized nnUNet for Efficient and Anatomically Accurate 3D Tooth Segmentationcited by 0