A data-driven machine learning integrated wearable medical sensor framework for elderly care service

Ba, Tao; Li, Shan; Wei, Yangyang

doi:10.1016/j.measurement.2020.108383

Cited by 21 publications

(6 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ML models learn from body signals (ECG, electroencephalogram (EEG), electromyogram (EMG), oxygen saturation (SpO 2 ), blood pressure) and assist in remote monitoring, disease diagnosis, and elderly care [31]. An integrated data-driven framework in addition to ML can perform decision-making tasks [32] that depend on the interpretation developed from the signal features collected from elderly patients. Advanced deep learning techniques such as the recurrent neural network (RNN), long short-term memory (LSTM), and convolution neural network (CNN) can be applied to develop a predictive model that can investigate sleep quality as well as physical activity patterns during awake time [33].…”

Section: Prospect Of ML In Wmdmentioning

confidence: 99%

Visualizing Wearable Medical Device Research Trends: A Co-occurrence Network-Based Bibliometric Analysis

Misra,

Roy,

Dey

et al. 2023

Galician med. j.

View full text Add to dashboard Cite

Background. One of the most crucial aspects of someone’s life is health. Therefore, individuals should be conscious about keeping themselves healthy by regular monitoring their health, which can be done with the help of modern medical technologies. Wearable medical devices using wearable sensors are the popular names of emerging technologies in the modern healthcare domain. Aim. This work presents the results of a systematic investigation of extensive research that has occurred for the last two decades in these research streams to provide a comprehensive mapping and temporal distribution of wireless medical device research. Methods. This study presents a relationship between the bibliographic items, their quality, and the quantity representing the most effective research topics on wearable medical devices. The analysis is performed using two useful parameters, namely a bibliometric network and a co-occurrence matrix. Data collection, data standardization, data mapping, and result analysis are the steps involved in the bibliometric analysis technique. In this study, VOSviewer software for bibliometric analysis is applied to the Scopus database. Results. By analysing bibliometric indicators from the Scopus database and using VOSviewer, we represent their distribution in countries, institutions, top researchers, and top journals. Furthermore, we analyse the co-citation of cited authors and the co-occurrence of keywords. The outcomes of the clustering and keyword analysis indicate that the research domain primarily focuses on the Internet of Things, machine learning, wearable sensors, mobile health, electrocardiogram, etc. Conclusions. Statistical investigation in association with the visual exploration presented in this article provides more substantial information than either of them used separately. In the future, this article can illuminate researchers and practitioners to develop a different theory to look at the factors that influence predictability in the research domain of wearable medical devices.

show abstract

Section: Prospect Of ML In Wmdmentioning

confidence: 99%

Visualizing Wearable Medical Device Research Trends: A Co-occurrence Network-Based Bibliometric Analysis

Misra,

Roy,

Dey

et al. 2023

Galician med. j.

View full text Add to dashboard Cite

show abstract

“…One of the main topics of health management is "in-place remote healthcare assistance" (10 articles), mainly including health monitoring and health guidance. Some researchers utilized AI technology to design and improve the health monitoring system of the elderly, including Cardiovascular disease [39], Dementia [40], physical abilities [41][42][43][44], and depressive symptoms monitoring systems [45], while other researchers developed AIbased systems for elderly health guidance, such as drugs management [46], rehabilitation assistance [47], and depression management [48].…”

Section: Health Managementmentioning

confidence: 99%

A Survey of Artificial-Intelligence Enabled Digital Transformation in Elderly Healthcare Field

Lee

Wang

et al. 2022

Advances in Transdisciplinary Engineering

View full text Add to dashboard Cite

With the rapid growth of the elderly population, traditional elderly healthcare services can hardly meet the dynamic and uncertain demand and higher customer expectations. With the digital transformation with digital technological enablers of the elderly-related industry rising globally, Artificial Intelligence (AI) has begun to play a disruptive and critical role in the elderly healthcare industry. But there is still a lack of a holistic review about AI-enabled digital transformation (DT) in the elderly healthcare field. Therefore, 59 works of literature published from 2000 through 2021 were extracted and analysed from Web of Science. This study focuses on the application scenarios, the AI approach used, and the benefits of AI-enabled DT. And this literature review lays the foundation for future research in the elderly healthcare field.

show abstract

“…To reduce the high computational cost of this calculation, while maintaining the performance of the obtained solution, an adaptive data-driven differential evolution algorithm was proposed. In most of the tested cases, a better or competitive performance was obtained with other algorithms in terms of output strength [25].In the work carried out by [26] research is discussed where an integrated data-driven framework assisted by machine learning algorithms acquires signals based on personalized characteristics of elderly patients. Such work, based on wearable devices with a focus on patients' health, aims to improve the detection performance, based on reduced quantity of measurements, aiming at better model accuracy.…”

Section: Examples Of Data-driven Approaches For Solving Problemsmentioning

confidence: 99%

Data Driven: An Overview and Practical Measures for Organizations

Oliveira¹,

Braga²

2021

IJAERS

View full text Add to dashboard Cite

Data is considered a primary resource for innovation. The existence of a large amount of available data, as well as technological tools capable of explore them, allows companies to extract information that can be used to create and implement new ideas and new projects. To this end, the details regarding the care that organizations should have with data are explored. The difficulties regarding the adoption of datadriven approach and some measures to implement this type of decisionmaking approach are also discussed. Some examples of data-driven approaches for diverse industries and products are shown. A real example of prediction model for decision making that is based on industrial data is also discussed. This example shows the difficulties in the preparation of data for the building of these models, which confirms that most of the time spent in the construction of predictive models it is due to this step. The use of the data-driven approach allows organizations to obtain superior results in their processes, thus becoming a tremendous competitive advantage and a special strategic factor in a highly competitive market, regardless of the field of activity.

show abstract

A data-driven machine learning integrated wearable medical sensor framework for elderly care service

Cited by 21 publications

References 24 publications

Visualizing Wearable Medical Device Research Trends: A Co-occurrence Network-Based Bibliometric Analysis

Visualizing Wearable Medical Device Research Trends: A Co-occurrence Network-Based Bibliometric Analysis

A Survey of Artificial-Intelligence Enabled Digital Transformation in Elderly Healthcare Field

Data Driven: An Overview and Practical Measures for Organizations

Contact Info

Product

Resources

About