Federated Learning for Microvasculature Segmentation and Diabetic Retinopathy Classification of OCT Data

Lo, Julian; Yu, Timothy T.; Ma, Da; Zang, Pengxiao; Owen, Julia P.; Zhang, Qinqin; Wang, Ke; Beg, Mirza Faisal; Lee, Aaron; Jia, Yali; Sarunic, Marinko V.

doi:10.1016/j.xops.2021.100069

Cited by 58 publications

(33 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach has been successfully implemented in the context of retinal microvasculature segmentation and referable diabetic retinopathy detection on optical coherence tomography (OCT) and OCT angiography images. The authors demonstrated that a federated learning approach achieved similar results as a traditional centralized learning approach ( 25 ). Similarly, instead of transferring data to train a DL model, the model itself can be “brought” to the data for retraining.…”

Section: Discussionmentioning

confidence: 96%

The Ethical and Societal Considerations for the Rise of Artificial Intelligence and Big Data in Ophthalmology

Liu

2022

Front. Med.

View full text Add to dashboard Cite

Medical specialties with access to a large amount of imaging data, such as ophthalmology, have been at the forefront of the artificial intelligence (AI) revolution in medicine, driven by deep learning (DL) and big data. With the rise of AI and big data, there has also been increasing concern on the issues of bias and privacy, which can be partially addressed by low-shot learning, generative DL, federated learning and a “model-to-data” approach, as demonstrated by various groups of investigators in ophthalmology. However, to adequately tackle the ethical and societal challenges associated with the rise of AI in ophthalmology, a more comprehensive approach is preferable. Specifically, AI should be viewed as sociotechnical, meaning this technology shapes, and is shaped by social phenomena.

show abstract

Section: Discussionmentioning

confidence: 96%

The Ethical and Societal Considerations for the Rise of Artificial Intelligence and Big Data in Ophthalmology

Liu

2022

Front. Med.

View full text Add to dashboard Cite

show abstract

“…In their work, Bercea et al (2021) proposed a framework for federated unsupervised brain anomaly segmentation. In their work, Lo et al (2021) showed that a federated learning model could achieve similar results as models trained on fully centralized data for microvasculature segmentation. While these methods mainly focus on regular 2D segmentation tasks for grid images, this work, in contrast, investigates the more challenging 3D tooth segmentation task over complicated and heterogeneous geometrical medical data.…”

Section: Federated Learning For Medical Image Segmentationmentioning

confidence: 97%

Privacy-Preserved Federated Learning for 3D Tooth Segmentation in Intra-Oral Mesh Scans

et al. 2022

View full text Add to dashboard Cite

Semantic segmentation over three-dimensional (3D) intra-oral mesh scans (IOS) is an essential step in modern digital dentistry. Many existing methods usually rely on a limited number of labeled samples as annotating IOS scans is time consuming, while a large-scale dataset of IOS is not yet publicly available due to privacy and regulatory concerns. Moreover, the local data heterogeneity would cause serious performance degradation if we follow the conventional learning paradigms to train local models in individual institutions. In this study, we propose the FedTSeg framework, a federated 3D tooth segmentation framework with a deep graph convolutional neural network, to resolve the 3D tooth segmentation task while alleviating data privacy issues. Moreover, we adopt a general privacy-preserving mechanism with homomorphic encryption to prevent information leakage during parameter exchange between the central server and local clients. Extensive experiments demonstrate that both the local and global models trained with the FedTSeg framework can significantly outperform models trained with the conventional paradigm in terms of the mean intersection over union, dice coefficient, and accuracy metrics. The FedTSeg framework can achieve better performance under imbalanced data distributions with different numbers of clients, and its overall performance is on par with the central model trained with the full dataset aggregated from all distributed clients. The data privacy during parameter exchange of FedTSeg is further enhanced with a homomorphic encryption process. Our work presents the first attempts of federated learning for 3D tooth segmentation, demonstrating its strong potential in challenging federated 3D medical image analysis in multi-centric settings.

show abstract

“…Additionally, in [ 116 ], the authors evaluated the effectiveness of federated neural network-based retinal microvasculature segmentation and classification of referable diabetic retinopathy (RDR) using optical coherence tomography (OCT) and OCT angiography (OCTA). For this purpose, several datasets were used, including SFU prototype swept-source OCTA, RTVue XR Avanti (OptoVue, Inc.), Angioplex (Carl Zeiss Meditec), and PLEX Elite 9000 (Carl Zeiss Meditec).…”

Section: Federated Learning In Actionmentioning

confidence: 99%

Reviewing Federated Machine Learning and Its Use in Diseases Prediction

Moshawrab

Adda

Bouzouane

et al. 2023

Sensors

View full text Add to dashboard Cite

Machine learning (ML) has succeeded in improving our daily routines by enabling automation and improved decision making in a variety of industries such as healthcare, finance, and transportation, resulting in increased efficiency and production. However, the development and widespread use of this technology has been significantly hampered by concerns about data privacy, confidentiality, and sensitivity, particularly in healthcare and finance. The “data hunger” of ML describes how additional data can increase performance and accuracy, which is why this question arises. Federated learning (FL) has emerged as a technology that helps solve the privacy problem by eliminating the need to send data to a primary server and collect it where it is processed and the model is trained. To maintain privacy and improve model performance, FL shares parameters rather than data during training, in contrast to the typical ML practice of sending user data during model development. Although FL is still in its infancy, there are already applications in various industries such as healthcare, finance, transportation, and others. In addition, 32% of companies have implemented or plan to implement federated learning in the next 12–24 months, according to the latest figures from KPMG, which forecasts an increase in investment in this area from USD 107 million in 2020 to USD 538 million in 2025. In this context, this article reviews federated learning, describes it technically, differentiates it from other technologies, and discusses current FL aggregation algorithms. It also discusses the use of FL in the diagnosis of cardiovascular disease, diabetes, and cancer. Finally, the problems hindering progress in this area and future strategies to overcome these limitations are discussed in detail.

show abstract

Federated Learning for Microvasculature Segmentation and Diabetic Retinopathy Classification of OCT Data

Cited by 58 publications

References 30 publications

The Ethical and Societal Considerations for the Rise of Artificial Intelligence and Big Data in Ophthalmology

The Ethical and Societal Considerations for the Rise of Artificial Intelligence and Big Data in Ophthalmology

Privacy-Preserved Federated Learning for 3D Tooth Segmentation in Intra-Oral Mesh Scans

Reviewing Federated Machine Learning and Its Use in Diseases Prediction

Contact Info

Product

Resources

About