FedMix: Mixed Supervised Federated Learning for Medical Image Segmentation

Wicaksana, Jeffry; Yan, Zengqiang; Zhang, Dong; Huang, Xijie; Wu, Huimin; Yang, Xin; Cheng, Kwang-Ting

doi:10.1109/tmi.2022.3233405

Cited by 24 publications

(18 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, all data were preprocessed identically and maintained as high-quality, pixel-level segmentation data. In practice, such ideal conditions may need to be relaxed (22). Additionally, we had the advantage of using a CL model to evaluate FL performance.…”

Section: Discussion (800)mentioning

confidence: 99%

In-Silo Federated Learning vs. Centralized Learning for Segmenting Acute and Chronic Ischemic Brain Lesions

Kim,

Lee,

Park

et al. 2024

Preprint

View full text Add to dashboard Cite

Purpose: To investigate the efficacy of federated learning (FL) compared to industry-level centralized learning (CL) for segmenting acute infarct and white matter hyperintensity. Materials and Methods: This retrospective study included 13,546 diffusion-weighted images (DWI) from 10 hospitals and 8,421 fluid-attenuated inversion recovery images (FLAIR) from 9 hospitals for acute (Task I) and chronic (Task II) lesion segmentation. The mean ages (SD) for the training datasets were 68.1 (12.8) for Task I and 67.4 (13.0) for Task II. The frequency of male participants was 51.5% and 60.4%, respectively. We trained with datasets from 9 and 3 institutions for Task I and Task II, respectively, and externally tested them in datasets from 1 and 9 institutions each. For FL, the central server aggregated training results every four rounds with FedYogi (Task I) and FedAvg (Task II). A batch clipping strategy was tested for the FL models. Performances were evaluated with the Dice similarity coefficient (DSC). Results: In Task I, the FL model employing batch clipping trained for 360 epochs achieved a DSC of 0.754±0.183, surpassing an equivalent CL model (DSC 0.691 ± 0.229; p<0.001) and comparable to the best-performing CL model at 940 epochs (DSC 0.755 ± 0.207; p=0.701). In Task II, no significant differences were observed amongst FL model with clipping, without clipping, and CL model after 48 epochs (DSCs of 0.761 ± 0.299, 0.751 ± 0.304, 0.744 ± 0.304). Few-shot FL showed significantly lower performance. Task II reduced training times with batch clipping (3.5 to 1.75 hours). Conclusion: Comparisons between CL and FL in identical settings suggest the feasibility of FL for medical image segmentation.

show abstract

Section: Discussion (800)mentioning

confidence: 99%

In-Silo Federated Learning vs. Centralized Learning for Segmenting Acute and Chronic Ischemic Brain Lesions

Kim,

Lee,

Park

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…[23] significantly enhances brain tumor segmentation accuracy by employing a scalable FL framework with U-Net model architecture, ensuring advanced data privacy and security measures. On a related note, the FedMix model tailored its approach to address the varying levels of image supervision across local clients by dynamically adjusting aggregation weights [12]. Furthermore, tackling the specific challenges of non-IID distribution and class-heterogeneity, the FedSeg model emerged as a specialized solution for class-heterogeneous semantic segmentation [24].…”

Section: Related Work a Federated Learning In Semantic Segmentationmentioning

confidence: 99%

“…It facilitates the amalgamation of data across various hospitals, ensuring adherence to stringent privacy norms [11]. While studies like [11] and [12] highlight the advantages of FL through approaches such as the decentralized MQTT framework for brain tumor segmentation and the label-agnostic FedMix method for diverse medical image segmentation, they often overlook the critical need for communication efficiency optimization in models with large parameters. However, the increasing model size and associated rise in communication costs in FL restrict its application, limiting the inclusion of data to only a select number of medical institutions.…”

Section: Introductionmentioning

confidence: 99%

FKD-Med: Privacy-Aware, Communication-Optimized Medical Image Segmentation via Federated Learning and Model Lightweighting Through Knowledge Distillation

Sun,

Shu,

Shao

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Advances in deep learning have revolutionized medical image segmentation, facilitating the precise delineation of complex anatomical structures. The scarcity of annotated training samples remains a significant bottleneck. To tackle the data limitation, federated learning (FL) offers the promise of pooling data from multiple healthcare institutions. However, as models grow larger, the increase in communication costs restricts FL to fewer nodes, which constrains the volume of data. This situation necessitates the simultaneous achievement of model lightweighting. To address this problem, this study proposes FKD-Med, a novel framework that integrates FL for privacy-sensitive data amalgamation across multiple healthcare institutions, and uses knowledge distillation (KD) to enhance communication efficiency. The "Med" in FKD-Med refers to medical application computational problems. Our principal contributions encompass the design of an open-source framework that seamlessly blends FL and KD, rendering it applicable to a broad spectrum of medical informatics tasks. Our approach substantially augments the computational data volume, thereby boosting both communication efficiency and training throughput. Tested on two datasets of medical image segmentation using TransUNet and ResUNet as teacher models, FKD-Med achieves data privacy, lowers communication costs, and increases accuracy. The parameters of the student models were reduced to 1/127 and 1/1027 of those in the teacher models. Additionally, the models subjected to KD exhibited accuracy improvements of 0.25%, 0.43%, 1.35%, and 1.46% respectively, given the same parameter volume. This positions FKD-Med not only as a pivotal tool for multi-institutional medical research but also as a versatile platform adaptable to a wide array of real-world medical engineering applications. The code is publicly available at https://github.com/SUN-1024/FKD-Med.

show abstract

“…FedMix is a label-independent adaptive aggregation algorithm that makes good use of labels from pixels, bounding boxes, and images. 41 FedMix enables discriminative feature representation for all participating centres by employing an adaptive weight assignment technique. When tested on brain tumor and skin lesion segmentation tasks, FedMix greatly beats SOTA approaches.…”

Section: Research On Fl For Medical Imagingmentioning

confidence: 99%

Federated learning for medical imaging radiology

Rehman,

Hugo Lopez Pinaya,

Nachev

et al. 2023

The British Journal of Radiology

View full text Add to dashboard Cite

Federated learning (FL) is gaining wide acceptance across the medical AI domains. FL promises to provide a fairly acceptable clinical-grade accuracy, privacy, and generalisability of machine learning models across multiple institutions. However, the research on FL for medical imaging AI is still in its early stages. This paper presents a review of recent research to outline the difference between state-of-the-art [SOTA] (published literature) and state-of-the-practice [SOTP] (applied research in realistic clinical environments). Furthermore, the review outlines the future research directions considering various factors such as data, learning models, system design, governance, and human-in-loop to translate the SOTA into SOTP and effectively collaborate across multiple institutions.

show abstract

FedMix: Mixed Supervised Federated Learning for Medical Image Segmentation

Cited by 24 publications

References 48 publications

In-Silo Federated Learning vs. Centralized Learning for Segmenting Acute and Chronic Ischemic Brain Lesions

In-Silo Federated Learning vs. Centralized Learning for Segmenting Acute and Chronic Ischemic Brain Lesions

FKD-Med: Privacy-Aware, Communication-Optimized Medical Image Segmentation via Federated Learning and Model Lightweighting Through Knowledge Distillation

Federated learning for medical imaging radiology

Contact Info

Product

Resources

About