Detecting Elderly Behaviors Based on Deep Learning for Healthcare: Recent Advances, Methods, Real-World Applications and Challenges

Almutairi, Mubarak; Chiroma, Haruna; Abubakar, Sule

doi:10.1109/access.2022.3186701

Cited by 15 publications

(19 citation statements)

References 89 publications

(81 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Six primary criteria were identified in this research as possible explanations for the peculiar actions of the elderly. Almutari et al [7] derived these variables from features associated with the various actions taken by the elderly. In this context, ''abnormal behavior,'' ''falls,'' ''hand gesture,'' ''face recognition,'' ''depressive condition,'' and ''body gesture'' are all relevant factors to consider.…”

Section: B the Social Inclusion Dimension Of 5g Adoptionmentioning

confidence: 99%

“…Based on the fact that while gathering the requirement of the features necessary to build an application that will aid in monitoring elderly people behavior in general, in which Almutari et al [7] provided 6 of those features to include ''Abnormal Behavior Detection (AB)'', ''Falls Detection (FD)'', ''Hand Gesture Detection (HG)'', ''Facial Recognition Detection (FR)'', ''Depressed Disorder Detection (DD)'', ''Body Gesture Detection (BG)'', from these features and assessment of their interrelations which was examined. The availability of a reliable network is one of the resources that is required to recognize older behavior.…”

Section: Conceptualization Of the Research Variablesmentioning

confidence: 99%

“…However, the implementation of such a system must take into account the well-being of the elderly people and the unusual behavior that might arise. For this reason, the current study draws on the features of the elderly people gathered from Almutari et al [7] from the perspective of designing a system for detecting elderly people behavior, such as ''Abnormal Behavior Detection (AB)'', ''Falls Detection (FD)'', ''Hand Gesture Detection (HG)'', ''Facial Recognition Detection (FR)'', ''Depressed Disorder Detection (DD)'', and ''Body Gesture''. On the negative side, there are numerous causes of unusual behavior that are related with medical and psychiatric diseases.…”

Section: Introductionmentioning

confidence: 99%

“…• The study, was able to mapped activities from Alshammari et al [8] dataset with Abnormal behavior'' dimension from Almutari et al [7] in order to conceptualized and examine the key criteria that influence an unusual behavior to further use deep learning for detecting these attributes.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The Key Criteria for Predicting Unusual Behavior in the Elderly With Deep Learning Models Under 5G Technology

Zamzami

2023

IEEE Access

View full text Add to dashboard Cite

Deep learning algorithms and technology based on 5G networks may be able to help identify unusual behavior in elderly people. Because 5G networks have a lower latency and a greater bandwidth, it is possible to use more complex algorithms and larger data sets for training and detection in a real-time. On top of that, real-time analysis of the data gathered through in-home monitoring of the elderly can become much simpler to carry out with the help of 5G's potential to simplify the process. However, the system needs to be developed in a way that it considers the preferences of the most important criteria of elderly people, who have more requirements in terms of their living situation and the environment in which they live. Therefore, this study advocated using ''Decision Making Trial and Evaluation Laboratory'' (DEMATEL) to analyze the most crucial criteria (feature) required for creating a model for identifying odd behavior in elderly persons. Convolutional Neural Networks (CNNs) and Long Short-Term Memories (LSTMs) are adopted in detecting unusual behavior in the elderly after the analysis key criteria for predicting unusual behavior in the elderly by DEMATEL. The research established a concept by linking the SIMADL dataset with the dimension of elderly people's behavior and performed an experimental analysis using both CNN and LSTM. Performance evaluations show that the LSTM performs better in detecting unusual behavior in elderly persons with 96% accuracy. Depressive disorder is the most significant aspect of ageing that might lead to a typical unusual behavior in elderly persons, according to DEMATEL analysis.

show abstract

Section: B the Social Inclusion Dimension Of 5g Adoptionmentioning

confidence: 99%

Section: Conceptualization Of the Research Variablesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The Key Criteria for Predicting Unusual Behavior in the Elderly With Deep Learning Models Under 5G Technology

Zamzami

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…A common approach to meet computing requirements for such applications is to deploy DL models on cloud computing platforms [5]. However, some of these applications require secure real-time analysis of generated medical data [6][7][8], which subjects them to security [9] and latency [10] issues and makes them unsuitable for deployment. Although they still provide a viable approach to meet the low-latency [11], privacy-preserving [10], and security [12] requirements [13], edge computing devices find it challenging to meet the computation, memory, and power requirements of deep neural networks (DNNs) [14].…”

Section: Introductionmentioning

confidence: 99%

Spike-Event X-ray Image Classification for 3D-NoC-Based Neuromorphic Pneumonia Detection

Wang¹,

Ikechukwu²,

Dang³

et al. 2022

Electronics

View full text Add to dashboard Cite

The success of deep learning in extending the frontiers of artificial intelligence has accelerated the application of AI-enabled systems in addressing various challenges in different fields. In healthcare, deep learning is deployed on edge computing platforms to address security and latency challenges, even though these platforms are often resource-constrained. Deep learning systems are based on conventional artificial neural networks, which are computationally complex, require high power, and have low energy efficiency, making them unsuitable for edge computing platforms. Since these systems are also used in critical applications such as bio-medicine, it is expedient that their reliability is considered when designing them. For biomedical applications, the spatio-temporal nature of information processing of spiking neural networks could be merged with a fault-tolerant 3-dimensional network on chip (3D-NoC) hardware to obtain an excellent multi-objective performance accuracy while maintaining low latency and low power consumption. In this work, we propose a reconfigurable 3D-NoC-based neuromorphic system for biomedical applications based on a fault-tolerant spike routing scheme. The performance evaluation results over X-ray images for pneumonia (i.e., COVID-19) detection show that the proposed system achieves 88.43% detection accuracy over the collected test data and could be accelerated to achieve 4.6% better inference latency than the ANN-based system while consuming 32% less power. Furthermore, the proposed system maintains high accuracy for up to 30% inter-neuron communication faults with increased latency.

show abstract

Neuronale Netze und Deep Learning

Puppe,

Krug,

Kempf

2024

Digitale Innovationen in Der Pflege

View full text Add to dashboard Cite

Detecting Elderly Behaviors Based on Deep Learning for Healthcare: Recent Advances, Methods, Real-World Applications and Challenges

Cited by 15 publications

References 89 publications

The Key Criteria for Predicting Unusual Behavior in the Elderly With Deep Learning Models Under 5G Technology

The Key Criteria for Predicting Unusual Behavior in the Elderly With Deep Learning Models Under 5G Technology

Spike-Event X-ray Image Classification for 3D-NoC-Based Neuromorphic Pneumonia Detection

Neuronale Netze und Deep Learning

Contact Info

Product

Resources

About