1. Efficient Large-scale Medical Data (eHealth Big Data) Analytics in Internet of Things
Authors:
Christos Stergiou
Andreas P. Plageras
Kostas E. Psannis
George Kokkonis
Yutaka Ishibashi
Byung-Gyu Kim
B. Brij Gupta
International Workshop on the Internet of Things and Smart Services (ITSS2017)
Thessaloniki
24 – 26 July 2017

2. Purpose & Objectives Purpose: Goals:
The collection of medical/e-health big data in real time.
The transfer of these data through a network to a cloud server.
These data will be processed in the cloud which makes its analysis so as to become meaningless.
By the analysis of these data is done the data mining.
The transfer of the analyzed health data will be held by the devices of the relevant persons.
Goals:
An analytical study of the technologies IoT, Cloud Computing (CC), and Big Data to resolve various issues facing the health sector in relation to these technologies.
We deal with the security of medical data which are personal data & must be protected.

3. Contents Problems in the Healthcare Sector Big Data Challenges
Literature Review
Proposed Architecture
Experimental Results
Contribution to Theoretical & Applied Scientific Knowledge
Innovation & Future Work
Conclusions
Big Data appears as a technology created and developed through communication systems.
Due to data usage and even large-scale amounts of data, the use of storage space without restrictions on its use becomes necessary.
Cloud Computing mentions to a substructure in which data storage and data processing occur in real time outside of the user’s device.

4. Problems in the Healthcare Sector
Issues:
The rapidly aging (demography problem)
The chronic diseases
Hypertension, Heart failure, Diabetes mellitus, etc.
The rate diseases
(e.g. Alzheimer’s disease)
The hereditary diseases,
The lack of health personnel and the health infrastructure,
The difficult treatment of emergency cases
(e.g. the accidents, the emergency obstetric care etc.),
The organizational problem,
The patients with mild disease (avian, etc.) that they need no monitoring and binding site on the already congested hospitals’ infrastructure,
The corruption of information over time.

5. Big Data Challenges Challenges of large-scale data
their representation,
reducing the redundancy that exists in them,
the quality and the variety,
the management of the life cycle,
the confidentiality,
their expendability,
the energy management,
the heterogeneity,
the speed and the accuracy,
the privacy and the security,
the storing of them,
the extracted knowledge from them,
the creation or development of their analysis tools and algorithms or techniques as well as, other serious issues that need total improvement.

6. Literature Review The latest findings
there are some “gaps” in the way in which such data are transmitted through the levels of management, analysis, and transportation,
but also, some problems that arise from their use, and which we will try to optimize by proposing new techniques and new algorithmic solutions.

7. Proposed Architecture
The system:
Provides to the relevant people and in real time, medical information related to the health of a patient,
Monitoring the patients’ health in & out of the hospital,
Freeing thereby places
such as a hospital bed
Freeing resources of the hospital
such as food & further savings
Provides more comfortable environment for the patient.

8. Proposed Architecture

9. Used & Future Tools
The tools we will use for the development and implementation of this research will be:
the use of Big Data software technologies, IoT, and CC,
as well as algorithms for the creation of platforms, both in code form and in flow chart form
Contiki OS & Cooja Emulator

10. The Cooja Emulator

11. Sensors’ Average Temperature

12. Sensors’ Temperature

13. Sensors’ Battery Voltage

14. Sensors’ Battery Indicator

15. Sensors’ Relative Humidity

16. Network’s Latency

17. Network’s Packets Received (over time)

18. Network’s Packets Lost (over time)

19. Network’s Packet Received (per Node)

20. Network’s Hops per Node

21. Average Power Consumption

22. Average Radio Duty Cycle

23. Instantaneous Power Consumption

24. History of Power Consumption

25. Contribution to Theoretical Scientific Knowledge
In the future, we will propose new medical data transport protocols for real time communications via the Internet.
In the future, we will propose the best way to protect the transmitted medical data.
In the future, we will investigate the smooth transition in real time of one wireless data transmission technology to another.
e.g. Bluetooth Low Energy to LoRaWAN.

26. Contribution to Applied Scientific Knowledge
Hybrid Transmission Data System
combines the main and the modern technologies of wireless data transfer

27. Innovation & Future Work
Aim
Field of bio-informatics and IoT
Security of medical data
Novelty
Combination of bio-informatics and IoT
Security based on development & implementation of a communication protocol

28. Innovation & Future Work
Transfer sensitive medical data
Propose new data transfer protocols
Investigate maximum upload rate in each protocol
Propose new data compression algorithms

29. Conclusions
Purpose
Collection of medical (e-health) big data in real time
Transfer medical data to a cloud server
Data mining of medical data in cloud server
Deal security issues
Future Work
Develop & implement a communication protocol
Propose innovative & optimized algorithms

30. Thank you for your attention!
Questions?