Elsevier

Computer Communications

Volume 160, 1 July 2020, Pages 521-533
Computer Communications

Fog-inspired smart home environment for domestic animal healthcare

https://doi.org/10.1016/j.comcom.2020.07.004Get rights and content

Abstract

Domestic veterinary care is contemplated as one of the significant domains of the healthcare industry. Conspicuously, this research presents a Smart Home-based healthcare monitoring framework for domesticated animals in real-time. The research work employs the Internet of Things (IoT)-based data acquisition in the ambient environment of the home. Acquired IoT-data is pre-processed for feature extraction over the Fog–Cloud computing platform. Moreover, a temporal data granule is formulated using the Temporal Data mining technique, which is used to quantify healthcare vulnerability in terms of Scale of Health Adversity (SoHA) and Temporal Adversity Estimate (TAE). Based on this, a Multi-scaled Long Short Term Memory (M-LSTM) based vulnerability prediction is performed for preventive veterinary healthcare services. Moreover, a fog-assisted real-time alert generation module is presented in the proposed framework to notify the concerned veterinary doctor in the case of a medical emergency. To validate the proposed framework, the experimental simulations are performed over challenging dataset comprising of nearly 34,120 instances. Results show that the presented framework is able to register enhanced performance in comparison to several state-of-the-art decision-making techniques in terms of Temporal Effectiveness, Classification Efficiency, Prediction Efficacy, and System Stability.

Introduction

Development in Information and Communications Technology (ICT) has uncovered numerous technological advancements in recent times [1]. Internet of Things (IoT), a primer ICT-player, has revolutionized numerous industrial sectors resulting in the notion of Industry 4.0 [2]. With the growth of wireless sensor technologies, smart data preceptors, and intelligent devices, IoT has resulted in several technological innovations that were challenging task for earlier technology. Moreover, the development of Fog Computing has resulted in the time-sensitive decision-making with enhanced accuracy. In the healthcare domain, the trio-logical aspects of IoT–Fog–Cloud (IFC) have provided numerous medical services in terms of mobile health care, intelligent medication recommender system, real-time medical emergency systems, and ubiquitous healthcare monitoring. Fig. 1 shows the layered architecture of IFC computing paradigm [3]. Distant health surveillance, infection determination from a remote site, disease spread information, and wireless health tracking are some of the novel applications innovated by next-generation smart healthcare. Moreover, with the incorporation of the IFC platform, time-sensitive analysis of healthcare data has been effectively enhanced to provide efficient decision-making medical services. Furthermore, with the developments of deep learning techniques like Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN) it has become feasible to automate healthcare systems for provisioning mobile healthcare recommendations. Conspicuously, this study is focused on utilizing the IFC paradigm to monitor and predict domesticated animal’s healthcare vulnerability in the home-centric environment.

An increase in animal care awareness has resulted in the adoption of a large number of animals and pets around the world. The animals include dogs, and cats are among the most preferable domesticated pets by human beings. The survey report of People for the Ethical Treatment of Animals (PETA) has depicted the adoption of nearly 6 million cats and dogs in the United States (US).1 Conspicuously, the financial expenditure on domesticated animal care has risen to 6 billion US dollars. On the contrary, a recent veteran report has depicted an increase in severe diseases like Lyme Disease, and Cushing’s Disease [4]. Factually, a 6-month survey report performed in the US showed a 9% death rate in domesticated animals [5]. Numerous factors contribute to the healthcare vulnerability of animals. Specifically, Heart rate > 180 bpm can lead to Tachycardia, Body temperature > 103 F result in Pyrexia, Blueish skin color can be due to Cyanosis, and restlessness can cause Canine seborrhea [5]. Moreover, the accidental causalities among domesticated animals can result in temporary or permanent health loss in the form of bone-impairment. As a result, it becomes difficult for domesticated pets to survive in the household environment. Furthermore, with minimal human intervention for regularized monitoring, domesticated animals have often been ignored in provisioning effective healthcare. In such a scenario, IFC technology can be effectively incorporated to develop an efficient framework to monitor healthcare conditions of domesticated animals in a time-sensitive manner. Moreover, the systems can be designed to generate medical emergency alert signals for real-time healthcare by a remote veterinary doctor. Efficient monitoring of domesticated animals like dogs and cats at homes has added a new dimension for the next generation veterinary service provisioning. Continuous increment in the utilization of mobile internet, IFC computing has altered the course of health data perception, transmission, and storage [6], [7]. This has been possible as a result of small internet-enabled wireless devices having the capability of sensing and transmitting data in real-time [8]. IFC paradigm has paved the way for the development of smart home-based remote pet monitoring procedure [9]. In addition to this, compiling statistics in a time-sensitive manner to identify precursors of harmful anomalies and remote actions has added a new dimension for diagnostic monitoring for veterinary doctors. Since IoT devices are fairly small and reliable, the accuracy level of information acquisition and transmission is very high [10]. Furthermore, by home-based displacement of such sensing devices, it has become possible to observe both macroscopic and microscopic animal’s health conditions in an efficient way. Inspired by these aspects, the major contribution of the presented research are detailed as follows:

  • 1.

    The presented research focuses on utilizing the IFC computing paradigm to present an intelligent home environment for domesticated animals’ health monitoring and prediction of vulnerability so that emergency alerts can be generated in a time-sensitive manner.

  • 2.

    Fog computing node is used to accumulate IoT-data for classification in 2 classes, namely Vulnerable Health Data (VHD) class and Invulnerable Health Data (IHD) class based on probabilistic parameter of Scale of Health Adversity (SoHA).

  • 3.

    SoHA measure is further assessed over real-time health conditions of the domesticated animals to quantify a probabilistic measure of Temporal Adversity Estimate (TAE). TAE presents a prevalence measure to determine health adversity in a time-sensitive manner.

  • 4.

    Multi-scaled Long Short-Term Memory (M-LSTM) based healthcare prediction system is presented which aids in the generation of early warning medical alerts. This enables time-sensitive actions in the direction of the animal’s wellness.

  • 5.

    The proposed IFC-inspired smart-home framework for veterinary health care is simulated in comparison to numerous state-of-the-art decision models to determine the overall performance enhancement.

Fig. 2 depicts the conceptual framework of the proposed smart home environment for domesticated veterinary healthcare.

The rest of the article is organized in different sections. Section 2 gives review of some of the significant works in the domain of IoT-based health surveillance. The proposed solution concerning the Smart Home for monitoring domesticated animals is presented in Section 3. The evaluation of the proposed framework and related outcomes are discussed in Section 4. Section 5 concludes the article with significant future work pathways.

Section snippets

Literature review

This section presents a brief review of related works in the field of IFC-based veterinary health monitoring frameworks. Additionally, a sub-section is formulated to depict significant contributions in the domain of fog-centric healthcare environments.

Proposed model

The modular architecture of the proposed framework for the monitoring of domesticated animals has been depicted in Fig. 3. The incorporation of the IFC computing platform has enabled effective data sensation in real-time. Specifically, the enhanced efficacy of IoT sensors in acquiring ubiquitous data enables accurate data perception regarding animal environment, behavior, and health parameters. The presented framework is comprised of 4 layers including,

  • 1.

    Data Sensation (DS) layer

  • 2.

    Data

Performance evaluation

This section estimates the performance of the proposed model for smart-home based domesticated animal healthcare. The presented framework incorporates several important layers. In the initial layer, data is acquired in real-time using numerous IoT bio-sensors embedded in the ambient environment of the home. Based on the acquired data values, 2 classes have been formulated based on the vulnerability aspect using BBN model. These data instances are aggregated and quantified in terms of SoHA and

Conclusion

Domesticated Animal healthcare has been a major concern for people around the world. Due to inappropriate care, domestic pets are being engulfed in deadly diseases. Conspicuously, this research proposes a fog computing-inspired smart home methodology for provisioning an effective healthcare environment for domesticated animals. Specifically, in this research (i) IoT technology has been utilized to acquiring real-time data in the ambient environment of smart home; (ii) Acquired data is

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (30)

  • YangGeng et al.

    A health-iot platform based on the integration of intelligent packaging, unobtrusive bio-sensor, and intelligent medicine box

    IEEE Trans. Ind. Inf.

    (2014)
  • FarrellMaxwell J. et al.

    Disease mortality in domesticated animals is predicted by host evolutionary relationships

    Proc. Natl. Acad. Sci.

    (2019)
  • HeChenguang et al.

    Toward ubiquitous healthcare services with a novel efficient cloud platform

    IEEE Trans. Biomed. Eng.

    (2013)
  • BonomiFlavio et al.

    Fog computing and its role in the internet of things

  • KoopC. Everett et al.

    Future delivery of health care: Cybercare

    IEEE Eng. Med. Biol. Mag.

    (2008)
  • Cited by (0)

    View full text