Parallel Processing of Massive EEG Data with MapReduce

Wang, Lizhe; Chen, Dan; Ranjan, Rajiv; Khan, Samee U.; Kołodziej, Joanna; Wang, Jun

doi:10.1109/icpads.2012.32

Cited by 22 publications

(13 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With cost effectiveness, performance, flexibility and low power consumption in mind, scheduling is a very important part in cloud computing and it has been given the attention it deserves by researchers and the industry [7,8]. This is why there have been so many different approaches, each trying to solve a particular scheduling problem or to be as general as possible, and take into account multiple cluster parameters.…”

Section: Related Workmentioning

confidence: 98%

Adaptive Resource Allocation in Cloud Computing Based on Agreement Protocols

Pop

Tutueanu

Barbieru

et al. 2015

Studies in Big Data

View full text Add to dashboard Cite

In the last years there has been considerable interest in using distributed systems, especially Cloud Systems, in any domains. Resource management contributes to ensure quality of services for any type of application, especially when a system involves elements of heterogeneity characterized by a variety of resources that may or may not be coupled with specific platforms or environments. A problem very close to the industry is the capability to allocate resources in an efficient way and estimate costs, especially when switching from one provider to another. In this chapter we present an extended work oriented on agreement-based resource allocation and a scheduling algorithm, aimed to bring an adaptive fault tolerant distributed system. For the agreement protocol we describe and analyze a 3-Tier structure of resources (hosts and virtual machines). Then an adaptive mechanism for agreement establishment is described. The allocation method considers workload distribution, resources heterogeneity, transparency, adaptability and also the ease to extend by combining with other scheduling algorithms.

show abstract

Section: Related Workmentioning

confidence: 98%

Adaptive Resource Allocation in Cloud Computing Based on Agreement Protocols

Pop

Tutueanu

Barbieru

et al. 2015

Studies in Big Data

View full text Add to dashboard Cite

show abstract

“…Although MapReduce is an easy-programming parallel framework for big data [35], it is not easy to comprehensively parallelize the whole iterative process of L-BFGS. Considering the iterative nature of overall algorithm L-BFGS, we choose to optimize each step of the iteration.…”

Section: Mapreduce Schemementioning

confidence: 99%

CRFs based parallel biomedical named entity recognition algorithm employing MapReduce framework

et al. 2015

View full text Add to dashboard Cite

As the rapid growth of the biomedical literature, the model training time in biomedical named entity recognition increases sharply when dealing with large-scale training samples. How to increase the efficiency of named entity recognition in biomedical big data becomes one of the key problems in biomedical text mining. For the purposes of improving the recognition performance and reducing the training time, this paper proposes an optimization method for two-phase recognition using conditional random fields. In the first stage, each named entity boundary is detected to distinguish all real entities. In the second stage, we label the semantic class of the entity detected. To expedite the training speed, in these two phases, we implement the model training process on a parallel optimization program framework based on MapReduce. Through dividing the training set into several parts, the iterations in the training algorithm are designed as map tasks which can be executed simultaneously in a cluster, where each map function is designed to complete the calculation of a gradient vector component for each part in the training set. Our experiments show that the proposed method in this paper can achieve high performance with short training time, which has important implications for the current biological big data processing.

show abstract

“…Although we have not found any studies for seizure detection in EEG data using Spark or MapReduce, there are some studies that use MapReduce for processing and storage of EEG data. In study (Wang, 2012), the authors implement a parallel version of Ensemble Empirical Mode Decomposition (EEMD) algorithm using MapReduce. They advocate that although EEMD is an innovative technique for processing neural signals, it is highly compute intensive and data-intensive.…”

Section: Related Workmentioning

confidence: 99%

“…Studies of processing EEG data using MapReduce or other cloud based programming models have been scarce. Though previous studies (Wang, 2012;Dutta, 2011) show commendable results, they were used for general processing and storage of EEG, and not for seizure detection. Furthermore, in most of the studies, EEG data was processed as a batch in an offline setting.…”

Section: Introductionmentioning

confidence: 99%

Parallel Real Time Seizure Detection in Large EEG Data

Ahmed

Edlund

Laure

et al. 2016

Proceedings of the International Conference on Internet of Things and Big Data

View full text Add to dashboard Cite

Electroencephalography (EEG) is one of the main techniques for detecting and diagnosing epileptic seizures. Due to the large size of EEG data in long term clinical monitoring and the complex nature of epileptic seizures, seizure detection is both data-intensive and compute-intensive. Analysing EEG data for detecting seizures in real time has many applications, e.g., in automatic seizure detection or in allowing a timely alarm signal to be presented to the patient. In real time seizure detection, seizures have to be detected with negligible delay, thus requiring lightweight algorithms. MapReduce and its variations have been effectively used for data analysis in large dataset problems on general-purpose machines. In this study, we propose a parallel lightweight algorithm for epileptic seizure detection using Spark Streaming. Our algorithm not only classifies seizures in real time, it also learns an epileptic threshold in real time. We furthermore present "top-k amplitude measure" as a feature for classifying seizures in the EEG, that additionally assists in reducing data size. In a benchmark experiment we show that our algorithm can detect seizures in real time with low latency, while maintaining a good seizure detection rate. In short, our algorithm provides new possibilities in using private cloud infrastructures for real time epileptic seizure detection in EEG data.

show abstract

Parallel Processing of Massive EEG Data with MapReduce

Cited by 22 publications

References 24 publications

Adaptive Resource Allocation in Cloud Computing Based on Agreement Protocols

Adaptive Resource Allocation in Cloud Computing Based on Agreement Protocols

CRFs based parallel biomedical named entity recognition algorithm employing MapReduce framework

Parallel Real Time Seizure Detection in Large EEG Data

Contact Info

Product

Resources

About