IoT Implementation In Healthcare: A Case Study Essay.

Application areas of IoT

The Internet of things is currently one of the most popular technologies and the IoT allows mobility within the sensor network. There are numerous application areas of IoT, including Healthcare, Smart cities, smart office and home networks, waste management, security, retail and logistics as well as an industrial sensor network that leverage the capabilities of wireless communication architecture. The study aims to perform a case study based analysis on the various features, issues and opportunities of IoT technology and its implementation in the Healthcare sector. The study outlines some of the issues of IoT system architecture within the case study. Some mitigation plans that justify the approach by evaluating two case studies related to IoT have also been discussed.

• IoT network of Padova University, Italy, is spread across different buildings provides services like environmental monitoring, managing services depending on the user's roles and authorities, localization to the users.
• The system is formed using the concept of IoT and Web 2.0 and uses the 6LoWPAN technology.
• The 6LoWPAN technology helps to marge older and newer web services.
• All the nodes in the network are compatible with IPv6 protocol, allowing direct access to the Internet-capable devices.
• This network offers three services, (i) Office automation, (ii) Allowing only registered users, and (iii) Guest can only access the basic services.
• The Binary XML services module, along with the UDP/6LoWPAN stack, offers both client and server capabilities.

Figure: Wiring infrastructure at the University of Padova

(Source: Created by author)

• The pilot study implemented an IoT sensor network for narcolepsy, a sleeping irregularity due to an imbalance in brain chemical.
• S-Band sensing method has been used that involves orthogonal frequency division multiplexing (Within 2 to 4 GHz frequency range) has been used.
• In this case, validation and classification of human activities like walking, push-ups, and narcolepsy are being done with a K-nearest neighbor, a support vector machine, and a random forest algorithm has been used.
• These IoT-based systems help monitor and record patient’s activity allowing nurses to provide timely services.
• The method uses S-Band sensing and various wireless devices such as an Omni-directional antenna (Receiver) and a card connected to the antenna.
• Multipath propagation impacts the experiment results.

Figure: Time history of Patient’s activities

(Source: Created by author)

1. The existing Wireless sensor network lacks flexibility, and creating a new application on the existing application is challenging and time-consuming.
2. The web service is of two types including SOAP-based and RESTFULL approach based. The SOAP-based web service may have some difficulties as it is not lightweight.
3. In wireless sensor network involving another network on top of the existing network increase the transmission overhead (Adeel et al. 2019).

1. Detection of patients’ daily activities in the healthcare sector is challenging and a vital issue for healthcare personnel.
2. The project involves a combination of categorical and numerical variables where the issues arise due to standardization among the numerical variables between 1 and 0.
3. The major issue outlined in the classification between the different behaviors of the patients with sleeping irregularity.
4. Invasive sensors can effectively monitor patients’ medical problems. However, involving invasive sensors is not possible as it is unsuitable for patients with Parkinson’s disease, skin disease, narcolepsy or infants (Shah et al. 2018).
5. Ultrasound or more radiation technology (UWB) also having issues with limited range, spectrum licensing, and video-based classification cannot be used as its impact on the security feature.
6. Categorizing different human activities such as walking, sitting, sitting on a chair or sleep episodes are challenging.
7. Another significant issue is the Multipath reflections in the room which reduces the experimental results (Shah et al. 2017).

1. In order to add flexibility to the network, the 6LoWPAN standard has been used along with the IPv6 protocol that connects the Wireless sensor network to allow the sensor nodes to connect to the IP protocol.
2. The web service is of two types, including SOAP-based and RESTFULL approach based where the SOAP-based web service may have some difficulties as it is not lightweight.
3. Using Binary web service for the wireless sensor network and the RESTFULL approach and Binary encoded XML in handling the resources helps to reduce the transmission overhead (Ullah and Mahmoud 2017).

• Mitigating the issues by using the invasive sensors, the system used non-invasive sensors such as RadioSence and ZigBee radio-based activity-based prototype system as proposed by other research.
• This system uses the UWB radar system-based fall detection techniques to eliminate the challenges and also eliminate the spectrum licensing issues; this paper uses the S-Band sensing technique with IEEE 802.11 specifications of wireless communication.
• The study used different machine learning algorithms such as S-Brand sensing, Support vector machine, K-nearest neighbor, and Random forest algorithm to categorize different human behavior (Kuo et al. 2018).
• This experiment used microwave-absorbing materials to avoid the noise due to multipath reflections.

1. Since the network requires an effective and secure authentication service for the users (Teachers), the RFID-based authentication technique is useful as it authenticates the authorized users and records data. Along with that, it also blocks any unauthorized users (Ghani et al. 2019).
2. The network utilizes IEEE 802.15.4 radio signal with IPv6/UDP, where the protocol stack helps to provide interoperability between the internet and the TestBed (García-Martín and Torralba 2021).
3. The implementation of the Remote BWS servers is accessible in this case following as it uses a complementally interface namely BWS Client.
4. The EXL algorithm is very much effective in this case as the binary XML coding is easier during the implementation of different multifaceted web services within the sensor node.
5. REST architecture requires global standards where the existing infrastructure with TCP/HTTP/ XML is much more challenging in the case of a larger network. For this reason, the proposed IETF or 6LoWAPP standard is much more effective by following the SENSEI guidelines (Yang et al. 2017).

1. The S-band sensing technique is effective in providing a wireless network interface.
2. The network used a RESTFULL-based network which is preferable due to its lightweight character.
3. The network infrastructure uses the IEEE 802.11n standard that offers faster, enhanced security features, is less prone to Interference as well as capable of connecting with the advanced Multimedia (Abane, Muhlethaler and Bouzefrane 2021).
4. Additionally, the use of SVM, RF Classifiers, and KNN provide effective and accurate detection of human activities and detection of sleep episodes.
5. As the network using wireless communication protocol by employing radio frequency, the added microwave-absorbing materials help to reduce the noise in the output due to multipath reflections.

Conclusion 

This paper compares different technologies addressed in the two case studies. The technologies are both related to the IoT infrastructure. However, there are some differences in the security devices and security standards depending upon the network devices used. The paper identified that the hospital patient monitoring system uses various sensors connected to the internet via Wi-Fi connectivity, Omni-directional antenna and IEEE 802.11n Security with different machine learning language standards. However, the Padova University network uses standalone mobile devices, static sensor nodes, RFID technology with 802.15.4 and UDP/6LoWPAN standards. There are also differences in the Data analysis techniques as the patient monitoring system utilizes data visualization techniques through different machine learning and statistical method to interpret the result into user understandable format. On the other hand, the university network does not require such a visualization technique.

The following two case studies related to implementing IoT infrastructure within the University of Padova and in Hospital infrastructure to offer effective monitoring of patient activities having Narcolepsy or sleeping disorder. These papers outline two different IoT infrastructures with specific security standards and sensors (Haider et al. 2019). The papers give detailed calculations of the classification result obtained from SVM and RF signals inpatient monitoring systems to help in further studies. Similarly, the commands used in BWS Server and BWS Client within the university network also help increase the potential of network infrastructure by utilized in future studies.

References

Abane, A., Muhlethaler, P. and Bouzefrane, S., 2021. Modeling and improving named data networking over IEEE 802.15. 4. Annals of Telecommunications, 76(11), pp.839-850.

Adeel, A., Gogate, M., Farooq, S., Ieracitano, C., Dashtipour, K., Larijani, H. and Hussain, A., 2019. A survey on the role of wireless sensor networks and IoT in disaster management. In Geological disaster monitoring based on sensor networks (pp. 57-66). Springer, Singapore.

García-Martín, J.P. and Torralba, A., 2021. Model of a Device-Level Combined Wireless Network Based on NB-IoT and IEEE 802.15. 4 Standards for Low-Power Applications in a Diverse IoT Framework. Sensors, 21(11), p.3718.

Ghani, A., Mansoor, K., Mehmood, S., Chaudhry, S.A., Rahman, A.U. and Najmus Saqib, M., 2019. Security and key management in IoT?based wireless sensor networks: An authentication protocol using symmetric key. International Journal of Communication Systems, 32(16), p.e4139.

Haider, D., Romain, O., Le Kernec, J., Shah, S.Y., Farooq, M.M.U. and Qadus, Z., 2019, August. Monitoring body motions related to huntington disease by exploiting the 5G paradigm. In 2019 UK/China Emerging Technologies (UCET) (pp. 1-4). IEEE.

Kuo, Y.W., Li, C.L., Jhang, J.H. and Lin, S., 2018. Design of a wireless sensor network-based IoT platform for wide area and heterogeneous applications. IEEE Sensors Journal, 18(12), pp.5187-5197.

Shah, S.A., Ren, A., Fan, D., Zhang, Z., Zhao, N., Yang, X., Luo, M., Wang, W., Hu, F., Rehman, M.U. and Badarneh, O.S., 2018. Internet of things for sensing: A case study in the healthcare system. Applied sciences, 8(4), p.508.

Shah, S.A., Zhang, Z., Ren, A., Zhao, N., Yang, X., Zhao, W., Yang, J., Zhao, J., Sun, W. and Hao, Y., 2017. Buried object sensing considering curved pipeline. IEEE antennas and wireless propagation letters, 16, pp.2771-2775.

Ullah, I. and Mahmoud, Q.H., 2020, May. A scheme for generating a dataset for anomalous activity detection in iot networks. In Canadian Conference on Artificial Intelligence (pp. 508-520). Springer, Cham.

Yang, X., Shah, S.A., Ren, A., Zhao, N., Fan, D., Hu, F., Rehman, M.U., von Deneen, K.M. and Tian, J., 2017. Wandering pattern sensing at S-band. IEEE journal of biomedical and health informatics, 22(6), pp.1863-1870.