Towards a cooperative hierarchical healthcare architecture using the HMAN offloading scenarios and SRT calculation algorithm

Abstract

The Internet of medical things (IoMT) is one of the most promising fields that is expected to rapidly expand in the near future. Consequently, vast amounts of data will be generated, necessitating faster and more intensive processing capabilities. Several healthcare architectures based on Edge/Fog technologies have been created to lower healthcare expenses and provide better and more reliable services. Scalability, availability, capacity, latency, and privacy are some of the most pressing issues to consider when designing such architectures due to the critical and sensitive nature of healthcare data. To contribute to the reliability and robustness of electronic health services, this work proposes Healthcare Metropolitan Area Network (HMAN), a cooperative hierarchical Edge/ Fog computing-based architecture for the urban healthcare systems. The presented architecture suggests HMAN offloading scenarios and system response time calculation (HOSSC) algorithm which is specially designed to provide an abundance of offloading and processing scenarios within the network. The architecture also connects patients to the healthcare system by utilising the existing infrastructure in cities (e.g. medical centres and hospitals). Simulation results revealed that the designed architecture produced a ubiquitous and scalable healthcare system with promising and competitive performance, such as the computing capacity and service availability, by adopting multiple cooperative hierarchical offloading scenarios across the framework units. Moreover, the HMAN system was evaluated for latency and found to be very robust, with a short response time ranging between 6.043 and 31.45 ms in responding to 1 to 300 patients simultaneously sending. In addition to these appealing features, the proposed architecture ensures patients' privacy because the data are locally stored and processed in the most anticipated scenarios. The proposed architecture is a viable solution to providing healthcare services to a large number of patients. | INTRODUCTIONThe Internet of Things (IoT) phrase was coined in 1999 by Kevin Ashton, a British technology pioneer [1]. Nowadays, the phrase 'Internet of Things' refers to any network of devices or things that can be characterised as a system connecting objects, involving humans, animals, and inanimate objects, over the Internet. The number of IoT applications have been exponentially growing [2,3]. According to Gartner's research, the worldwide number of connected devices could increase from 8.4 billion in 2020 to 20.4 billion in 2022 [4]. The McKinsey Global Institute estimated that IoT applications will have an annual global economic impact ranging from $3.9 trillion to $11.1 trillion [4]. According to Facts and Factors, the IoT market was expected to reach $1842 billion by 2028, rising at a Compound Annual Growth Rate of 24.5% from 2021 to 2028 [5].The Internet of medical things (IoMT) is one of the most attractive areas in the recent IoT developments, with a promising business growth forecast of $158.07 billion by 20...

Optimal intelligent edge‐servers placement in the healthcare field

“…However, if it is determined that the MC cannot handle the data of all 100 patients, then it must offload a part of it to other units based on the HOSSC algorithms in ref. [5]. The main task of the proposed priority mechanism is to determine which patients are served by the local MC and which ones should be offloaded.…”

“…Table 2 shows the results obtained from applying the above scenarios and demonstrates the effectiveness of the proposed algorithm in providing services with less delay to the patients with the highest level of urgency. In the case of HMAN [5], it is not possible to determine which patient should be treated first, because the system handles data equally without taking urgency into account. This could potentially have negative consequences on the treatment of patients with severe conditions.…”

“…The SRT is calculated based on [5]: $$$SRT=\frac{\sum\limits _{i=1}^{n}{t}_{pi}}{n}$$$ where: $$${t}_{pi}={T}_{wi}+{E}_{AP}\ast {T}_{AP}+\left({c}_{1}+1\right)\ast {t}_{GP}+{c}_{1}\ast V+{c}_{2}\ast S+{c}_{2}\ast {t}_{GH}+{c}_{3}\ast L+{c}_{3}\ast {t}_{GH}+{c}_{4}\ast B+{c}_{4}\ast {t}_{\mathrm{C}\mathrm{l}\mathrm{o}\mathrm{u}\mathrm{d}}$$$ …”

“…Subject to: $${N}_{es} > =\,3$$: to ensure that at least three edge‐servers are placed at 3 MCs to preserve previous achievements in [5]. $$\mathrm{d}\left({mc}_{a},{mc}_{b}\right)=k$$: to ensure a shorter distance between a patient and an MC and to avoid the colocated problem. $$\sum {E}_{ij}=1\,or\,2$$: to ensure that a patient either connects directly with an edge‐servers site or through only 1 MC to minimise the latency. $$\sum {y}_{ij}=1\ (1\le j\le n)$$: to guarantee that all patients must be served, and each is served from exactly one candidate point. …”

“…$${N}_{es} > =\,3$$: to ensure that at least three edge‐servers are placed at 3 MCs to preserve previous achievements in [5].…”