A Hadoop-Based Platform for Patient Classification and Disease Diagnosis in Healthcare Applications

Harb, Hassan; Mroue, Hussein; Mansour, Ali; Nasser, Abbass; Cruz, Eduardo Motta

doi:10.3390/s20071931

Cited by 27 publications

(10 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Analyzing the received data from sensors along with other historical data leads to more accurate outcomes using high speed workstations at server level. In [11], the authors propose a live data synthesis patient monitoring system, decision making, classification and disease diagnosis. Traditional classification techniques (Kmeans, naïve bayes, decision tree) suffer from a relatively slow performance when applied with numerous amounts of data like the case in health monitoring.…”

Section: Disease Diagnosingmentioning

confidence: 99%

“…Overcoming this performance issue, an update on Kmeans (SKmeans) is proposed which differs by gathering patients according to their consistency status concluded from the overall score that is computed from measurements instead of using the measurements directly. Also, for every repetition, a patient is allocated to the closet centroid according to the steadiness point [11] for the complete set of measurements and this results in enormous decrease in computation time.…”

Section: Disease Diagnosingmentioning

confidence: 99%

See 1 more Smart Citation

A Comparative Study for Evaluating Recent Advances in Wireless Body Sensor Network

Salika

Zaki

Saux

et al. 2022

2022 21st International Symposium on Communications and Information Technologies (ISCIT)

Self Cite

View full text Add to dashboard Cite

Internet of things (IoT) network helps communities, cities and industries better manage their work efficiently and provides a superior quality of service. This can be applied on Wireless Body Sensor Networks (WBSN) employed in healthcare industry. As any network, WBSN has its challenges that attracts researchers for its potential in serving humanity. Thus, enhancing WBSN could take many forms like enhancing the energy consumption of sensors, the communication or transmission and data analysis. Many researches cater to the above topics individually or combined to promise better outcomes. In this paper, we present a comparative study for some of the techniques and enhancement that researchers bring into WBSN for e-health applications and highlighting the gaps that would be a target for future researches. We provide a detailed overview of WBSN purposes and used techniques. The paper presents a number of the recent advanced techniques which fall under the survey's objective and compare them. For an objective comparison, we focus on a set of criteria related to e-health and involving data reduction, energy conservation, decision-making up to the intervention.

show abstract

Section: Disease Diagnosingmentioning

confidence: 99%

Section: Disease Diagnosingmentioning

confidence: 99%

A Comparative Study for Evaluating Recent Advances in Wireless Body Sensor Network

Salika

Zaki

Saux

et al. 2022

2022 21st International Symposium on Communications and Information Technologies (ISCIT)

Self Cite

View full text Add to dashboard Cite

show abstract

“…The integration of competition between multiple agents, multiple tasks and audiences has been tested in healthcare applications of Hadoop, with respect to outcomes such as patient classification and disease diagnosis. Hadoop manages health monitoring and assessment tasks in combining remote sensing devices and the big data technologies, in real time applications [43] due to Hadoop's "data analytical techniques for data analysis and classification".…”

Section: Integrationmentioning

confidence: 99%

Processing Big Data with Apache Hadoop in the Current Challenging Era of COVID-19

Azeroual

Fabre²

2021

BDCC

View full text Add to dashboard Cite

Big data have become a global strategic issue, as increasingly large amounts of unstructured data challenge the IT infrastructure of global organizations and threaten their capacity for strategic forecasting. As experienced in former massive information issues, big data technologies, such as Hadoop, should efficiently tackle the incoming large amounts of data and provide organizations with relevant processed information that was formerly neither visible nor manageable. After having briefly recalled the strategic advantages of big data solutions in the introductory remarks, in the first part of this paper, we focus on the advantages of big data solutions in the currently difficult time of the COVID-19 pandemic. We characterize it as an endemic heterogeneous data context; we then outline the advantages of technologies such as Hadoop and its IT suitability in this context. In the second part, we identify two specific advantages of Hadoop solutions, globality combined with flexibility, and we notice that they are at work with a “Hadoop Fusion Approach” that we describe as an optimal response to the context. In the third part, we justify selected qualifications of globality and flexibility by the fact that Hadoop solutions enable comparable returns in opposite contexts of models of partial submodels and of models of final exact systems. In part four, we remark that in both these opposite contexts, Hadoop’s solutions allow a large range of needs to be fulfilled, which fits with requirements previously identified as the current heterogeneous data structure of COVID-19 information. In the final part, we propose a framework of strategic data processing conditions. To the best of our knowledge, they appear to be the most suitable to overcome COVID-19 massive information challenges.

show abstract

“…The authors proposed a new model that uses a stability-based k-means (SK-means) algorithm to analyze and train the system to predict and diagnose coronavirus patients using data collected from real-time health sensors (e.g., heart rate, body temperature) (Harb et al, 2020). In their study's architecture, the proposed platform consists of four layers:…”

Section: State Of the Artmentioning

confidence: 99%

The Impact of Healthcare Traffic Over H2H and M2M Networks During the Spread Phase of Pandemic Diseases

Samhan

Fawal²,

Uddin³

et al. 2021

JER is an international, peer-reviewed journal that publishes f

View full text Add to dashboard Cite

Recently, the coronavirus pandemic has caused widespread panic around the world. Modern technologies can be used to monitor and control this highly contagious disease. A plausible solution is to equip each patient who is diagnosed with or suspected of having COVID-19 with sensors that can monitor various healthcare and location parameters and report them to the desired facility to control the spread of the disease. However, the simultaneous communication of numerous sensors installed in the majority of an area’s population results in a huge burden on existing Long-Term Evolution (LTE) networks. The existing network becomes oversaturated because it has to manage two kinds of traffic in addition to normal traffic (text, voice, and video): healthcare traffic generated by a large number of sensors deployed over a huge population, and extra traffic generated by people contacting their family members via video or voice calls. In pandemics, e-healthcare traffic is critical and should not suffer packet loss or latency due to network overload. In this research, we studied the performance of existing networks under various conditions and predicted the severity of network degradation in an emergency. We proposed and evaluated three schemes (doubling bandwidth, combining LTE-A and LTE-M networks, and request queuing) for ensuring quality of service (QoS) of healthcare sensor (HCS) network traffic without perturbation from routine human-to-human or machine-to-machine communications. Finally, we simulated all proposed schemes and compared them with existing network scenarios. The results have showed that when we have doubled the bandwidth the SCR of all traffic was 100% as same as the Queue strategy. However, when we prioritized the HCS traffic the SCR has recorded 100%, while H2H and M2M traffic has recorded 73%. When we used hybrid network LTE-A and LTE-M network, the HCS and H2H traffic has recorded 100% and M2M traffic has recorded 70%. After analyzing the results, we conclude that our proposed queuing schemes performed well in all conditions and provide the best QoS for HCS traffic.

show abstract

A Hadoop-Based Platform for Patient Classification and Disease Diagnosis in Healthcare Applications

Cited by 27 publications

References 24 publications

A Comparative Study for Evaluating Recent Advances in Wireless Body Sensor Network

A Comparative Study for Evaluating Recent Advances in Wireless Body Sensor Network

Processing Big Data with Apache Hadoop in the Current Challenging Era of COVID-19

The Impact of Healthcare Traffic Over H2H and M2M Networks During the Spread Phase of Pandemic Diseases

Contact Info

Product

Resources

About