Handling Privacy-Sensitive Medical Data With Federated Learning: Challenges and Future Directions

In the paper

“It also improves patient privacy and reduces low latency during the collaboration of multiple hospitals/entities [7].”

What the sources say

• Partly supported

  In “The promise of federated efforts” (page unavailable), the paper explains that collaborating institutions do not directly access one another’s patient data, supporting the privacy claim. “Challenges and considerations” discusses higher-throughput networks and reducing idle time, but does not say FL reduces latency during collaboration; the latency part is unsupported. (LLM, conf=0.88)

  cites [7] sourceExperimental — accuracy not yet measured

In the paper

“For example, [76] presents TrustFed, which uses Industrial IoT devices as federated learning candidates and smart contract technology over the Ethereum blockchain to maintain participants' reputations.”

What the sources say

• Supported

  The introduction describes IIoT devices acting as federated-learning servers or model trainers. The On-Chain FL Components section says a reputation smart contract calculates training devices’ scores, and the Results and Discussion section describes the contracts’ operation on the Ethereum blockchain. Page numbers are unavailable in our copy. (LLM, conf=0.91)

  cites [76] source

In the paper

“Although FL has been well studied in different domains, to the best of our knowledge, no existing work extensively reviews the use of FL in IoMT networks and applications.”

In the paper

“Independently of the FL training strategy, a recent analysis for the medical field has shown that a general FL-trained model can achieve performance levels comparable to the ones trained on centrally hosted datasets and superior to models that only see isolated single-institutional data [27], [28].”

What the sources say

• Not checked yetcites [27] source
• Not checked yetcites [28] source

In the paper

“With FL, a global model is trained collaboratively by each agent of the system (e.g., hospitals, IoMTs, or health care centers) over the decentralized network.This is done via local updates, without exchanging private data.”

What the sources say

• Not checked yetcites [7] source

In the paper

“FL also requires computationally robust devices.”

In the paper

“It also shows that the TFL framework increases the classification time.”

What the sources say

• Not checked yetcites [44] source

In the paper

“The experiment results demonstrate that FedHealth can improve the classification performance compared to the non-federated model and the traditional machine learning model.”

What the sources say

• Not checked yetcites [43] source

In the paper

“Although FL is the enabling technology in this scenario and provides a privacy-preservation capability by allowing the clients to keep the data on local devices, there are still model security and data leakage risks that would compromise the security of the FL system and the data privacy of clients.”

In the paper

“Moreover, recently FL has made a huge contribution to the fight against COVID-19 without sharing patient data.”

In the paper

“Also, the authors showed that when the training data is IID, the difference between FL and non-FL is within 3%.”

What the sources say

• Not checked yetcites [37] source

In the paper

“Their experimental results show that human activity recognition with semi-supervised FL is not affected by the non-IID data and can achieve comparable accuracy to that of the supervised FL.”

What the sources say

• Not checked yetcites [40] Zhao, 2020, Semi-supervised federated learning for activity recognition

In the paper

“Then, they study the trade-off between the communication cost and the model accuracy, which indicates that FL requires less communication and computational resources while using less complex models but at the cost of lower accuracy.”

What the sources say

• Not checked yetcites [37] source

In the paper

“The results demonstrate that the FTL outperforms the non-TFL approach in terms of the Area Under The Curve (AUC) metric and maintains the same accuracy.”

What the sources say

• Not checked yetcites [44] source

In the paper

“From the presented state-of-the-art on FL-IoMT, we can notice that FL plays an important role in facilitating healthcare services.”

In the paper

“For this reason, some IoMT devices offload their model training task to the edge gateways; therefore, helping the FL training process to take advantage of the cloud and edge computing, and using the IoMT devices as simple data collectors.”

In the paper

“It can be learned from [37] that FL requires less communication overhead, and hence it can replace traditional learning approaches in IoMT applications.”

What the sources say

• Not checked yetcites [37] source

In the paper

“Although the efficiency of FL application for healthcare systems, the shared global model trained on the FL server fails in personalization due to the different characteristics or daily activity patterns of users; hence, it is important to have a fine-grained or personalized model.”

In the paper

“Since federated learning comes with a privacy-preserving attribute, it can play a significant role in various industry domains that involve sensitive personal data, such as IoMT for healthcare.”

In the paper

“FL, despite a different training process compared to traditional ML, still suffers from information leakage issues and, having multiple parties, extends the attack surface.”

In the paper

“For example, along with the data privacy and delocalization problems, FL also solves the problem of having inadequate data by providing a trust factor between heterogeneous domains.”

In the paper

“As shown in Fig. 2, edge computing helps to train the model closer to the endusers and hence may conduct inference much faster than in the cloud [46].”

What the sources say

• Not checked yetcites [46] source

In the paper

“The experimental results show that ResNet-18 has the best performance in training both with FL and without FL approaches, while ResNeXt shows the highest efficiency on images with COVID-19 labels.”

What the sources say

• Not checked yetcites [49] source

In the paper

“In IoMT applications, the intercepted data could be a patient's personal or healthcare information, which presents a severe ethical problem.”

In the paper

“It provides a highly trained ML model without the risk of exposing training data since they are kept where they are generated.”

What the sources say

• Not checked yetcites [7] source

In the paper

“It is an efficient solution to address these issues and reduce the network congestion and latency that occur with the cloud.”

In the paper

“The experimental results on human activity recognition demonstrate the effectiveness of the FedHome framework in terms of accuracy, computation, and communication overhead.”

What the sources say

• Not checked yetcites [46] source

In the paper

“As a result, the recent technique of transfer learning, which avoids learning from scratch and solves the problem of insufficient training data, can be effective in this scenario [42].”

What the sources say

• Not checked yetcites [42] source

In the paper

“The theoretical and experimental comparisons show that the proposed model converges faster and with less communication overhead compared to an alternative distributed algorithm.”

What the sources say

• Not checked yetcites [35] source

In the paper

“Some clients, for example, may deliberately try to extract information from other parties, degrade performance, or bring the system down.”

In the paper

“While standards are necessary for security and privacy purposes, they can make sharing and using health data challenging.”

In the paper

“However, operating FL systems on a larger scale might be impractical to establish collaborative agreements and it is more reasonable not to have trustworthiness assumptions.”

In the paper

“For FL consortia in which all parties are bounded by an enforceable collaboration agreement, we can disregard some of the more nefarious motivations, such as deliberated attempts to extract sensitive information or to intentionally corrupt the model.”

In the paper

“In view of the above arguments, precious data often remain confined to individual institutions and are rarely leveraged for analysis, hindering the application of deep learning algorithms in healthcare.”

In the paper

“Besides, the model can be manipulated by attackers inducing additional memorization through gradient-ascent-style attacks.”

In the paper

“In fact, personal health information is considered to be highly sensitive, containing not only diagnostic and healthcare related information but also identifiable details about individuals.”

In the paper

“Bonawitz et al. [62] introduce a secure aggregation protocol for protecting individual model updates.”

What the sources say

• Not checked yetcites [62] source

In the paper

“The applied secure aggregation is a lossless method that can retain the original accuracy with a very high privacy guarantee.”

What the sources say

• Not checked yetcites [62] source

In the paper

“At the same time, it can be noted how some sophisticated countermeasures can be avoided if all parties are deemed trustworthy.”

In the paper

“Lastly, privacy problems are exacerbated in medical companies by strict laws and regulations designed to prevent the risk of re-identification and data breaches.”

In the paper

“If these standards are respected, FL can foster the application of deep learning in digital health and healthcare informatics, making it possible the use of existing health datasets while protecting user privacy.”

In the paper

“Even from the consumer perspective, data privacy is one of the public's most important concerns because data breaches can result in reduced public trust.”

In the paper

“In this solution, the central server is not able to see any local updates, but it can still observe the exact aggregated results at each round.”

What the sources say

• Not checked yetcites [62] source

In the paper

“Li et al. [65] introduced locally differentially private algorithms in the context of meta-learning while providing provable learning guarantees in convex settings.”

What the sources say

• Not checked yetcites [65] Li, 2019, Differentially private meta-learning

In the paper

“Another approach to establish data security and eliminate the trust issues is the deployment of FL on blockchain technology [71]- [73].”

What the sources say

• Not checked yetcites [71] source
• Not checked yetcites [72] source
• Not checked yetcites [73] source

In the paper

“This type of approach comes with a considerable number of hyperparameters that need to be carefully selected and that affect communication and accuracy.”

What the sources say

• Not checked yetcites [63] McMahan, 2017, Learning differentially private recurrent language models

In the paper

“The simulation results show that PRCL reduces the local training overhead because outsourcing the middle part to the cloud server while providing similar accuracy to other state-of-the-art approaches.”

What the sources say

• Not checked yetcites [68] source

In the paper

“If lower the respective target values, the lower the trust value of the client, and when such a trust value is lower than 60%, the request from this client will be prohibited.”

What the sources say

• Not checked yetcites [70] source

In the paper

“For stronger privacy guarantees, Bhowmick et al. [64] have proposed a relaxed version of local privacy by limiting the power of potential adversaries.”

What the sources say

• Not checked yetcites [64] Bhowmick, 2018, Protection against reconstruction and its applications in private federated learning

In the paper

“In addition, the blockchain-based verification scheme can also improve the reliability of the federated training process.”

In the paper

“Adopting the homomorphisms of secret sharing, it preserves user privacy and requires less computing and communication resources than alternatives as [62], and is able to provide acceptable fault tolerance in the event of user disconnection.”

What the sources say

• Not checked yetcites [69] source
• Not checked yetcites [62] source

In the paper

“At the same time, confidentiality is preserved since the cloud cannot access the plaintext gradient and an attacker can only get the perturbed data due to the added noise.”

What the sources say

• Not checked yetcites [68] source

In the paper

“Providing stronger guarantees than global privacy, it has better model performance than strict local privacy.”

What the sources say

• Not checked yetcites [64] Bhowmick, 2018, Protection against reconstruction and its applications in private federated learning

In the paper

“Similar results have been achieved by EPPDA [69], an efficient privacy-preserving data aggregation method for FL that, based on secret sharing to resist the reverse attack, can covertly aggregate user-trained models without disclosing the user model.”

What the sources say

• Not checked yetcites [69] source

In the paper

“Wang et al.proposed a defensive strategy based on comparing the uploaded model's accuracy and size [70].”

What the sources say

• Not checked yetcites [70] source

In the paper

“Lu et al. [77] designed a new system, called DITEN, that integrates blockchain and FL in edge networks and uses Deep Neural Networks (DNN) as a strategy scheduler to ensure the privacy of user data and enhance learning security.”

What the sources say

• Not checked yetcites [77] source

In the paper

“In such a way, block generation efficiency is guaranteed, enabling decentralized privacy protection while preventing single point failure.”

What the sources say

• Not checked yetcites [72] source

In the paper

“Such a solution offers several advantages such as decentralization, non-tempering, open autonomy, and anonymous traceability [75].”

What the sources say

• Not checked yetcites [75] source

In the paper

“To empower FL with this approach, Qu et al. [72] propose to store only the pointer of the global updates on-chain, while a distributed hash table is used to save the data.”

What the sources say

• Not checked yetcites [72] source

In the paper

“A blockchain-based FL, however, should mitigate the overhead of frequent communications in the blockchain consensus, which can cause excessive latency.”

In the paper

“The solution can also identify and eliminate outliers in the training distributions before combining the model updates.”

What the sources say

• Not checked yetcites [76] source

In the paper

“Two key aspects must be considered for an effective reduction of the communication overhead: the total number of communication rounds, and the size of messages transmitted in each round.”

In the paper

“The most commonly used method for FL, FedAvg [29], is based on averaging local stochastic gradient descent (SGD) updates for the primal problem.”

What the sources say

• Not checked yetcites [29] McMahan, 2017, Communication-efficient learning of deep networks from decentralized data

In the paper

“However, in practice, these solutions suffer from several issues and challenges that should be considered when designing FL-based approaches for healthcare services.”

In the paper

“Some recent techniques can partially solve these privacy issues [61], however, the more secure the techniques, the lower the accuracy of the final model may be [80].”

What the sources say

• Not checked yetcites [61] source
• Not checked yetcites [80] source

In the paper

“We can mention two main obstacles: (i) a high volume of aggregated data from different parties to be processed by the server; (ii) none of these parties fully trust each others (including the server), thus fearing data leakage.”

In the paper

“Although these asynchronous methods are inherently robust to stragglers, they may suffer from slow convergence due to the use of stale models.”

What the sources say

• Not checked yetcites [82] Dutta, 2018, Slow and stale gradients can win the race: Error-runtime trade-offs in distributed SGD

In the paper

“Moreover, decentralizing can be combined with blockchain technology, where a blockchain-based approach can jointly solve the problem of single-point-of-failure and central trust.”

In the paper

“Federated networks of IoMT can indeed comprise a massive number of devices (e.g., sensors), and communication in the network can be many orders of magnitude slower than local computation due to limited resources, such as bandwidth, energy, and power [86].”

What the sources say

• Not checked yetcites [86] source

In the paper

“While FedAvg has been shown to work well for non-convex problems, it does not exhibit convergence guarantees and can diverge in practical settings when data are heterogeneous [32] (see Section V-D for how to address this problem).”

What the sources say

• Not checked yetcites [32] Li, 2020, Federated optimization in heterogeneous networks

In the paper

“With either solution, i.e., peer-to-peer or blockchain, the problem is converted to the pursuit of fault tolerance for the clients.”

In the paper

“Other recent methods aiming to enhance the privacy of federated learning use tools such as secure multiparty computation or differential privacy.These approaches often provide privacy at the cost of reduced model performance or system efficiency [62].”

What the sources say

• Not checked yetcites [62] source

In the paper

“In conclusion, the features of FL make it uniquely suited for sensitive data as in the context of IoMT systems, but the architecture and model sharing must address further challenges compared to other fields.”

In the paper

“However, while this distributed approach can help to make the system highly scalable, it also opens up some issues in gradient synchronization.”

In the paper

“We can mention two main obstacles: (i) a high volume of aggregated data from different parties to be processed by the server; (ii) none of these parties fully trust each others (including the server), thus fearing data leakage.”

In the paper

“While blockchain removes the need for a trusted curator and connects each participant through multiparty data retrieval, it also strengthens the data-sharing scheme and provides some fault-tolerant guarantees [83].”

What the sources say

• Not checked yetcites [83] source

In the paper

“In a similar way, Lu et al. [78] integrated federated learning into the consensus algorithm of the blockchain to save the hash rate in the context of industrial IoT, in which blockchain enables secure data retrieval and ensures accurate model training.”

What the sources say

• Not checked yetcites [78] source

In the paper

“Lastly, decentralized training is a potential alternative in FL to reduce high communication costs on the central server.”

In the paper

“To further improve the transmission of model output, the approach is combined with Federated Augmentation (FAug), a data augmentation scheme that can collectively learn the trade-off between privacy leakage and communication overhead using a Generative Adversarial Network (GAN).”

What the sources say

• Not checked yetcites [91] Jeong, 2018, Communication-efficient on-device machine learning: Federated distillation and augmentation under non-IID private data

In the paper

“Moreover, Jeong et al. [91] presented Federated Distillation (FD), a distributed model training algorithm whose communication payload size is much smaller than that of a benchmark FL scheme, particularly when the model size is large.”

What the sources say

• Not checked yetcites [91] Jeong, 2018, Communication-efficient on-device machine learning: Federated distillation and augmentation under non-IID private data

In the paper

“FedProx [32], for instance, tolerates that any selected device to perform partial work, in compliance with the underlying system's constraints, and can safely incorporate these partial updates via a proximal term.”

What the sources say

• Not checked yetcites [32] Li, 2020, Federated optimization in heterogeneous networks

In the paper

“However, for medical data collected by IoMT simple models can be likely inaccurate.”

In the paper

“On top of conventional periodic communication of FL, the proposed FD exchanges, not the model parameters but the model output, so that large local models can be adopted on the ML device.”

What the sources say

• Not checked yetcites [91] Jeong, 2018, Communication-efficient on-device machine learning: Federated distillation and augmentation under non-IID private data

In the paper

“In addition, while a large number of clients can increase the diversity of the local model across them, this may degrade the performance of FL [93].”

What the sources say

• Not checked yetcites [93] Zhang, 2020, Benchmarking semisupervised federated learning

In the paper

“By selecting the devices with favorable resources, FedCS provides high classification accuracy in a significantly shorter time compared to the classical FL.”

What the sources say

• Not checked yetcites [94] source

In the paper

“Although some recent articles have investigated decentralized training over heterogeneous data with local updating schemes, e.g., [90], they are either restricted to linear models or assume full device participation.”

What the sources say

• Not checked yetcites [90] He, 2018, Cola: Decentralized linear learning

In the paper

“Results confirm that using less complex models, such as softmax regression, is a viable solution for most real-world applications since user behavior phenomena are often simple enough to be captured by relatively simple models.”

What the sources say

• Not checked yetcites [37] source

In the paper

“Although a high number of local updating methods can effectively reduce the total number of communication rounds, it can be convenient to reduce the size of messages exchanged at each round via compression schemes model, e.g., sparsification, subsampling, and quantization.”

In the paper

“The authors showed how using CNNs and LSTMs, the communication costs can be reduced by two orders of magnitude compared to the original federated learning algorithm.”

What the sources say

• Not checked yetcites [88] Konen, 2016, Federated learning: Strategies for improving communication efficiency

In the paper

“IoMT devices from remote areas may be proner to drop due to poor network connectivity or battery failure and, therefore, the trained federated model will be biased towards devices with favorable network conditions.”

In the paper

“Recent studies present a jointannouncement protocol that can randomly select some devices/ clients from a large number of participants in a given training round [85].Such a joint-announcement protocol is useful to alleviate the out-of-sync issue and tolerate the dropout of some clients at some points before the completion of each training iteration.”

What the sources say

• Not checked yetcites [85] Bonawitz, 2019, Toward federated learning at scale: System design

In the paper

“However, while several recent studies have investigated convergence guarantees of variants of FL methods [95]- [97], little work has evaluated the impact of failures over the FL algorithm or studied directly the effect of dropped devices.”

What the sources say

• Not checked yetcites [95] Jiang, 2018, A linear speedup analysis of distributed deep learning with sparse and quantized communication
• Not checked yetcites [96] source
• Not checked yetcites [97] Yu, 2019, On the linear speedup analysis of communication efficient momentum SGD for distributed non-convex optimization

In the paper

“Both approaches (multi-task and meta-learning) enable personalized or device-specific modeling, which is considered a valuable attempt to handle the statistical heterogeneity of the data.”

In the paper

“Once the trained generator of the GAN is downloaded, each device can replenish the target labels until an IID training dataset is achieved”

What the sources say

• Not checked yetcites [91] Jeong, 2018, Communication-efficient on-device machine learning: Federated distillation and augmentation under non-IID private data

In the paper

“This paradigm of data generation violates the frequently-used assumptions of independent and identically distributed (IID) data and may cause problems in modeling, theoretical analysis, and empirical evaluation of solutions.”

In the paper

“Therefore, new schemes ensuring fault tolerance would improve the deployment of FL in healthcare where IoMT devices are likely to a fault, encounter challenging network conditions, and exhaust their batteries.”

In the paper

“In addition to the variety of data, the number of data samples can vary significantly from device to device, and there may be an underlying statistical structure that captures the relationship among devices and their associated distributions.”

In the paper

“However, despite introducing a trade-off between accuracy and centralization, this could lead to significant communication overhead, especially for medical data.”

In the paper

“To rectify the non-IID training dataset while addressing the communication overhead, Jeong et al. [91] proposed FAug, in which each device can generate the missing data samples locally using a generative model, i.e., a conditional GAN.”

What the sources say

• Not checked yetcites [91] Jeong, 2018, Communication-efficient on-device machine learning: Federated distillation and augmentation under non-IID private data

In the paper

“It addresses the statistical challenge of FL when local data is non-IID by distributing a small amount of globally shared data containing examples from each class.”

What the sources say

• Not checked yetcites [31] Zhao, 2018, Federated learning with non-IID data

In the paper

“With ELFISH, the neurons that have less significant weight parameter updates will be randomly masked.”

What the sources say

• Not checked yetcites [107] Xu, 2019, ELFISH: Resource-aware federated learning on heterogeneous edge devices, Ratio

In the paper

“Furthermore, Anh et al. [109] applied a Deep Q-learning algorithm based on the Double Deep Q-Network (DDQN) to optimize resource allocation for model training.”

What the sources say

• Not checked yetcites [109] source

In the paper

“For example, training CNN-based models for COVID-19 detection (e.g., ResNet-50) requires storage and memory resources that may not be available on all devices.”

In the paper

“The simulation results show that the proposed approach can minimize energy consumption and improve the training latency.”

What the sources say

• Not checked yetcites [109] source

In the paper

“While machine learning approaches, and especially deep learning, are becoming the de-facto knowledge discovery approach in digital healthcare, it appears that data-driven medicine aiming to improve patient care globally requires federated efforts.”

In the paper

“Similarly, Xu et al. [107] proposed a resource-aware federated learning framework, called ELFISH.”

What the sources say

• Not checked yetcites [107] Xu, 2019, ELFISH: Resource-aware federated learning on heterogeneous edge devices, Ratio