IoT-blockchain empowered Trinet: optimized fall detection system for elderly safety
Numerous elderly folks reside alone in their homes. Seniors may find it difficult to ask for assistance if they fall. As the elderly population keeps growing, elderly fall incidents are becoming a critical public health concern. Creating a fall detection system for the elderly using IoT and blockcha...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-09-01
|
| Series: | Frontiers in Bioengineering and Biotechnology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fbioe.2023.1257676/full |
| _version_ | 1797678058112548864 |
|---|---|
| author | Fayez Alfayez Surbhi Bhatia Khan |
| author_facet | Fayez Alfayez Surbhi Bhatia Khan |
| author_sort | Fayez Alfayez |
| collection | DOAJ |
| description | Numerous elderly folks reside alone in their homes. Seniors may find it difficult to ask for assistance if they fall. As the elderly population keeps growing, elderly fall incidents are becoming a critical public health concern. Creating a fall detection system for the elderly using IoT and blockchain is the aim of this study. Data collection, pre-processing, feature extraction, feature selection, fall detection, and emergency response and assistance are the six fundamental aspects of the proposed model. The sensor data is collected from wearable devices using elderly such as accelerometers and gyroscopes. The collected data is pre-processed using missing value removal, null value handling. The features are extracted after pre-processed data using statistical features, autocorrelation, and Principal Component Analysis The proposed approach utilizes a novel hybrid HSSTL combines Teaching-Learning-Based Optimization and Spring Search Algorithm to select the optimal features. The proposed approach employs TriNet, including Long Short-Term Memory, optimized Convolutional Neural Network (CNN), and Recurrent Neural Network for accurate fall detection. To enhance fall detection accuracy, use the optimized Convolutional Neural Network obtained through the hybrid optimization model HSSTL. Securely store fall detection information in the Blockchain network when a fall occurs. Alert neighbours, family members, or those providing immediate assistance about the fall occurrence using Blockchain network. The proposed model is implemented in Python. The effectiveness of the suggested model is evaluated using metrics for accuracy, precision, recall, sensitivity, specificity, f-measure, NPV, FPR, FNR, and MCC. The proposed model outperformed with the maximum accuracy of 0.974015 at an 80% learning rate, whereas the suggested model had the best accuracy score of 0.955679 at a 70% learning rate. |
| first_indexed | 2024-03-11T22:53:54Z |
| format | Article |
| id | doaj.art-e808ffe7910d435aaab72f337041f500 |
| institution | Directory Open Access Journal |
| issn | 2296-4185 |
| language | English |
| last_indexed | 2024-03-11T22:53:54Z |
| publishDate | 2023-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Bioengineering and Biotechnology |
| spelling | doaj.art-e808ffe7910d435aaab72f337041f5002023-09-21T21:55:12ZengFrontiers Media S.A.Frontiers in Bioengineering and Biotechnology2296-41852023-09-011110.3389/fbioe.2023.12576761257676IoT-blockchain empowered Trinet: optimized fall detection system for elderly safetyFayez Alfayez0Surbhi Bhatia Khan1Department of Computer Science and Information, College of Science, Majmaah University, Al-Majmaah, Saudi ArabiaDepartment of Data Science, School of Science, Engineering and Environment, University of Salford, Salford, United KingdomNumerous elderly folks reside alone in their homes. Seniors may find it difficult to ask for assistance if they fall. As the elderly population keeps growing, elderly fall incidents are becoming a critical public health concern. Creating a fall detection system for the elderly using IoT and blockchain is the aim of this study. Data collection, pre-processing, feature extraction, feature selection, fall detection, and emergency response and assistance are the six fundamental aspects of the proposed model. The sensor data is collected from wearable devices using elderly such as accelerometers and gyroscopes. The collected data is pre-processed using missing value removal, null value handling. The features are extracted after pre-processed data using statistical features, autocorrelation, and Principal Component Analysis The proposed approach utilizes a novel hybrid HSSTL combines Teaching-Learning-Based Optimization and Spring Search Algorithm to select the optimal features. The proposed approach employs TriNet, including Long Short-Term Memory, optimized Convolutional Neural Network (CNN), and Recurrent Neural Network for accurate fall detection. To enhance fall detection accuracy, use the optimized Convolutional Neural Network obtained through the hybrid optimization model HSSTL. Securely store fall detection information in the Blockchain network when a fall occurs. Alert neighbours, family members, or those providing immediate assistance about the fall occurrence using Blockchain network. The proposed model is implemented in Python. The effectiveness of the suggested model is evaluated using metrics for accuracy, precision, recall, sensitivity, specificity, f-measure, NPV, FPR, FNR, and MCC. The proposed model outperformed with the maximum accuracy of 0.974015 at an 80% learning rate, whereas the suggested model had the best accuracy score of 0.955679 at a 70% learning rate.https://www.frontiersin.org/articles/10.3389/fbioe.2023.1257676/fullIoTblockchainelderly peopledeep learningSSATLBO |
| spellingShingle | Fayez Alfayez Surbhi Bhatia Khan IoT-blockchain empowered Trinet: optimized fall detection system for elderly safety Frontiers in Bioengineering and Biotechnology IoT blockchain elderly people deep learning SSA TLBO |
| title | IoT-blockchain empowered Trinet: optimized fall detection system for elderly safety |
| title_full | IoT-blockchain empowered Trinet: optimized fall detection system for elderly safety |
| title_fullStr | IoT-blockchain empowered Trinet: optimized fall detection system for elderly safety |
| title_full_unstemmed | IoT-blockchain empowered Trinet: optimized fall detection system for elderly safety |
| title_short | IoT-blockchain empowered Trinet: optimized fall detection system for elderly safety |
| title_sort | iot blockchain empowered trinet optimized fall detection system for elderly safety |
| topic | IoT blockchain elderly people deep learning SSA TLBO |
| url | https://www.frontiersin.org/articles/10.3389/fbioe.2023.1257676/full |
| work_keys_str_mv | AT fayezalfayez iotblockchainempoweredtrinetoptimizedfalldetectionsystemforelderlysafety AT surbhibhatiakhan iotblockchainempoweredtrinetoptimizedfalldetectionsystemforelderlysafety |