“…In machine learning methods, firstly, the algorithm is trained on a feature set extracted from the dataset then it is used for testing real-time data. Some famous machine learning algorithms used for detection or prediction of falls are: Support Vector Machine [23] Multilayer Perceptron [24], K-Nearest Neighbors [25], Naive Bayes [26], and others. These methods are used to gain insights of data for detection and even prediction of future falls.…”
AbstractFalling is a commonly occurring mishap with elderly people, which may cause serious injuries. Thus, rapid fall detection is very important in order to mitigate the severe effects of fall among the elderly people. Many fall monitoring systems based on the accelerometer have been proposed for the fall detection. However, many of them mistakenly identify the daily life activities as fall or fall as daily life activity. To this aim, an efficient machine learning-based fall detection algorithm has been proposed in this paper. The proposed algorithm detects fall with efficient sensitivity, specificity, and accuracy as compared to the state-of-the-art techniques. A publicly available dataset with a very simple and computationally efficient set of features is used to accurately detect the fall incident. The proposed algorithm reports and accuracy of 99.98% with the Support Vector Machine(SVM) classifier.
“…In machine learning methods, firstly, the algorithm is trained on a feature set extracted from the dataset then it is used for testing real-time data. Some famous machine learning algorithms used for detection or prediction of falls are: Support Vector Machine [23] Multilayer Perceptron [24], K-Nearest Neighbors [25], Naive Bayes [26], and others. These methods are used to gain insights of data for detection and even prediction of future falls.…”
AbstractFalling is a commonly occurring mishap with elderly people, which may cause serious injuries. Thus, rapid fall detection is very important in order to mitigate the severe effects of fall among the elderly people. Many fall monitoring systems based on the accelerometer have been proposed for the fall detection. However, many of them mistakenly identify the daily life activities as fall or fall as daily life activity. To this aim, an efficient machine learning-based fall detection algorithm has been proposed in this paper. The proposed algorithm detects fall with efficient sensitivity, specificity, and accuracy as compared to the state-of-the-art techniques. A publicly available dataset with a very simple and computationally efficient set of features is used to accurately detect the fall incident. The proposed algorithm reports and accuracy of 99.98% with the Support Vector Machine(SVM) classifier.
“…Several approaches are presented to help health providers offer a better quality of care than many in-person treatment modalities and improve telemedicine through IoT [53,54]. In addition, the topology of the telemedicine network is considered a subset of the overall healthcare and includes a required network structure to support the streaming of patients' vital signs from Tier 1 until it reaches the final destination at Tier 3 and vice versa [55,56]. Healthcare services from distributed hospitals are managed and distributed by a medical centre (Tier 3) in a normal scenario [57].…”
A newly distributed fault-tolerant mHealth framework-based Internet of things (IoT) is proposed in this study to resolve the essential problems of healthcare service provision during the occurrence of frequent failures in a telemedicine architecture. Two models are presented to support the telehealth development of chronic heart disease (CHD) in a distant environment. In model-1, a new local multisensor fusion triage algorithm known as three-level localisation triage (3LLT) is proposed. In 3LLT, a group of heterogeneous sources is applied to triage patients as a clinical process, and the emergency levels inside/ outside the home of a patient with CHD are determined. Failures related to sensor fusion can also be detected. In model-2, a centralised IoT connection towards distributed smart hospitals is employed by mHealth based on two attributes: (1) healthcare service packages and (2) time of arrival of a patient at a hospital. Three decision matrices have been used to overcome several issues on hospital selection based on multi-criteria decision-making by using an analytic hierarchy process. Two datasets are utilised: (1) a clinical CHD dataset, which includes 572 patients for testing model-1, and (2) a nonclinical dataset, which includes hospital healthcare service packages for testing model-2. Consequently, patients with CHD can be triaged into different emergency levels (risk, urgent and sick) with mHealth, and a final decision is made by selecting an appropriate hospital. Results are obtained through the clinical triage of patients, and different scenarios are provided for hospital selection. The verification of statistical results indicates that the proposed mHealth framework is systematically valid. The contribution of the mHealth framework is presented to provide an improved triage process, afford timely services and treatment for CVD patients and minimise the chances of error. These health sectors and policymakers can also recognise the evaluation benefits of smart hospitals by using the presented framework and move forward to fully automated mHealth applications.
“…With circulated Fogbased computational capacities, applications require a better capacity than adjust to the consistent changes that happen inside the elements of a cutting edge city: the best way to give the required adaptability will be using progressed Artificial Intelligence (AI) procedures that assist frameworks to learn and demonstrate the conduct of data in close continuous. When using the cloud-only solutions, the data retrieval times are too high for a real-time scenario, such as fall detection or stroke mitigation, that require high response times from medical professionals [10]. Frequently sending information to the cloud for computation accounts for higher power consumption and costs associated, even more so today, when the amount of data generated by sensors is huge [11].…”
In this paper, we propose a framework for the management of the Internet of Things (IoT) devices in a smart building to model services based on the serverless computing paradigm. The deployment of an IoT compatible serverless paradigm consists of a hierarchical structural design across the edge, fog, and cloud computing layers. The fog/edge nodes collect the data generated from various sensors, process the data in the intermediate nodes, and then forward certain data to a cloud for future analysis. The framework consists of a heterogeneous IoT network. We proposed a data distribution algorithm in the framework to make sure management, maintenance and availability of heterogeneous IoT network in the serverless computing paradigm are effective and efficient. The experiments conducted are validated at the developed fog and edge gateways using API mechanism. The response times for an application doing the computation at fog level and at the cloud level are compared. The experimentation shows that latency is less for the fog model as compared to the data sent to the cloud model.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.