OCT-FedSIR: Toward Trustworthy Federated Ophthalmic Learning under Annotation Noise
Authors: Sina Gholami, Abdulmoneam Ali, Tania Haghighi, Rashadul H. Badhon, Behafarin Emam, Sally S. Y. Ong, Atalie C. Thompson, Theodore Leng, +3 more
Organizations: Department of Electrical and Computer Engineering, University of North Carolina at Charlotte, Charlotte, NC, USA · Department of Ophthalmology, Wake Forest School of Medicine, Winston-Salem, NC, USA · Byers Eye Institute at Stanford, Stanford University School of Medicine, Stanford, CA, USA · Department of Ophthalmology and Visual Sciences, University of Illinois Chicago, Chicago, IL, USA
Federated learning enables collaborative model development without centralizing patient data, but annotation reliability at participating institutions cannot always be assumed. In ophthalmic imaging, differences in disease prevalence and class composition can resemble changes caused by corrupted supervision. We introduce OCT-FedSIR, a reliability-aware spectral framework for federated OCT classification under client-dependent annotation noise and heterogeneous data distributions. OCT-FedSIR combines class-balanced spectral estimation, Stage-I logit adjustment, complementary spectral descriptors, selective spectral relabeling, and noise-aware federated optimization. We evaluated the framework on the Kermany, University of Illinois Chicago, and Wake Forest datasets under symmetric and structured asymmetric noise and three levels of non-IID heterogeneity. Across 117 experimental conditions, OCT-FedSIR achieved a mean accuracy of 86.73%, compared with 79.94% for RoFL and 78.75% for FedCorr. It correctly separated clients with original and corrupted annotations across all evaluated conditions, while the original FedSIR identification procedure was less robust, particularly under asymmetric noise. Spectral relabeling recovered 77.2% of corrupted annotations with 91.3% correction precision and a 3.5% false-correction rate. Retaining corrected clients outperformed spectral pruning by 9.30 percentage points on average. These findings show that annotation noise can often be identified and corrected without discarding informative client data.
Figures & tables
Figure 1: Overview of the OCT-FedSIR framework. a , Global initialization is distributed to participating clients. b , Local training. c–e , Spectral statistics are estimated, communicated to the server, and used to identify clean and noisy clients. f , Clean-client spectral information is used to construct representative and residual references for relabeling. g , Noise-aware federated optimization using LA, KD, and DaAgg.
Dataset
Method
Symmetric F1 (%)
Asymmetric F1 (%)
Kermany
FedSIR
92.4
71.8
OCT-FedSIR
100.0
100.0
UIC
FedSIR
89.7
68.5
OCT-FedSIR
100.0
100.0
WF
FedSIR
91.2
74.6
OCT-FedSIR
100.0
100.0
Table 2: Clean-client identification F1 score (%) for FedSIR and OCT-FedSIR, averaged across the evaluated label-noise levels and Dirichlet heterogeneity settings. Bold indicates the best-performing method.
Figure 2: Spectral characterization and identification of clients with corrupted labels in representative Kermany experiments under severe statistical heterogeneity ( α=0.1 ). a–c , Class-wise spectral similarity matrices for clean, asymmetric-noise, and symmetric-noise clients, respectively. d , Distributions of mean off-diagonal similarity ( μk ), off-diagonal energy ( ek ), and maximum off-diagonal similarity ( skmax ). e , Joint spectral descriptor space. f , Client-identification confusion matrices across the evaluated symmetric and asymmetric noise levels.
Figure 3: Macro-F1 performance under symmetric label noise. Results are shown across α∈{2.0,0.5,0.1} for a , Kermany OCT classification; b , UIC DR classification; and c , WF GA classification. The dashed line denotes the all-clean FedAvg reference.
Figure 4: Macro-F1 performance under asymmetric label noise. Results are shown across α∈{2.0,0.5,0.1} for a , Kermany OCT classification; b , UIC DR classification; and c , WF GA classification. The dashed line denotes the all-clean FedAvg reference.
Configuration
Mean accuracy (%)
Δ
OCT-FedSIR (Full)
86.73
–
w/o DaAgg
86.21
−0.52
w/o LA
85.68
−1.05
w/o KD
84.91
−1.82
w/o relabeling
79.47
−7.26
Table 4: Ablation analysis of OCT-FedSIR. Mean accuracy is averaged across all evaluated experimental conditions. Δ denotes the decrease from the complete framework in percentage points.
Figure 5: Simulated symmetric label-noise transitions. For clients designated as noisy, a proportion η of annotations within each locally represented class was selected uniformly at random and reassigned uniformly to one of the remaining C−1 diagnostic classes. Clean clients retained their original annotations.
Figure 6: Dataset-specific asymmetric label-noise transitions. For clients designated as noisy, a proportion η of annotations within each applicable source class was selected and reassigned according to predefined dataset-specific class transitions. Clean clients retained their original annotations.
Figure 7: Spectral relabeling of predicted noisy clients using representative and residual spectral references constructed from predicted clean clients. A label change is accepted only when the representative- and residual-subspace predictions agree.
Federated learning (FL) enables collaborative model training without sharing raw data; however, the presence of noisy labels across distributed clients can severely degrade the learning performance. In this paper, we propose FedSIR, a multi-stage framework for robust FL under noisy labels. Different from existing approaches that mainly rely on designing noise-tolerant loss functions or exploiting loss dynamics during training, our method leverages the spectral structure of client feature representations to identify and mitigate label noise. Our framework consists of three key components. First, we identify clean and noisy clients by analyzing the spectral consistency of class-wise feature subspaces with minimal communication overhead. Second, clean clients provide spectral references that enable noisy clients to relabel potentially corrupted samples using both dominant class directions and residual subspaces. Third, we employ a noise-aware training strategy that integrates logit-adjusted loss, knowledge distillation, and distance-aware aggregation to further stabilize federated optimization. Extensive experiments on standard FL benchmarks demonstrate that FedSIR consistently outperforms state-of-the-art methods for FL with noisy labels. The code is available at https://github.com/sinagh72/FedSIR.
Sina Gholami, Abdulmoneam Ali, Tania Haghighi +2
University of North Carolina at Charlotte Charlotte, NC, USA
Split Federated Learning (SplitFed) combines federated and split learning to preserve privacy while reducing client-side computation. However, in medical image segmentation, heterogeneous label quality across clients can significantly degrade performance. We propose SplitFed-CL, a co-learning framework where a global teacher guides local students to detect and refine unreliable annotations. Reliable labels supervise training directly, while unreliable labels are corrected via weighted student--teacher refinement. SplitFed-CL further incorporates consistency regularization for robustness to input perturbations and a trainable weighting module to balance loss terms adaptively. We also introduce a novel difficulty guided strategy to simulate human like boundary centric annotation errors, where the degree of perturbation is governed by shape complexity and the associated annotation difficulty. Experiments on two multiclass segmentation datasets with controlled synthetic noise, together with a binary segmentation dataset containing real-world annotation errors, demonstrate that SplitFed-CL consistently outperforms seven state-of-the-art baselines, yielding improved segmentation quality and robustness.
Zahra Hafezi Kafshgari, Hadi Hadizadeh, Parvaneh Saeedi
School of Engineering Science, Simon Fraser University, Burnaby, BC, Canada
Federated learning (FL) enables multiple clinical institutions to collaboratively train a shared disease classifier without centralizing patient data. In practice, however, each institution annotates only the pathologies within its area of expertise, so the federation operates under task heterogeneity: each client holds labels for a strict subset of the target disease categories while the remaining classes are entirely unobserved at that site. Existing gradient-based FL methods fail under this setting because they require hundreds of communication rounds to converge and because missing class labels introduce systematic false-negative bias that the model cannot correct without a principled mechanism. We propose an analytic federated learning framework for multi-label medical image classification under task heterogeneity. The proposed method replaces iterative gradient optimization with three closed-form operations: a balanced label projection that neutralizes class-imbalance bias by normalizing positive and negative contributions to equal total mass; a per-class absolute aggregation law that independently assembles the optimal ridge-regression classifier for each disease category from the sufficient statistics uploaded by its annotating clients; and an optional analytic pseudo-label refinement round that propagates missing-class knowledge from a confidence-filtered teacher classifier to non-annotating clients. The entire procedure requires at most two communication rounds, irrespective of the degree of task heterogeneity or the number of participating clients. Experiments on ChestXray14 under four progressively severe missing-class configurations demonstrate that the proposed method consistently outperforms the state-of-the-art federated multi-label method FedMLP by up to 18.44 BACC points and 13.24 AUC points, while reducing the communication.
Afsaneh Mahanipour, Hana Khamfroush
Department of Computer Science University of Kentucky Lexington, KY, USA