Particle swarm optimization-driven deep maxout network for effective monitoring of paralyzed persons
DOI:
https://doi.org/10.14419/bk2d1638Keywords:
Particle Swarm Optimization (PSO); Deep Maxout Network (DMN); Posture Recognition; Fall Detection; Human Activity Monitoring; Healthcare AI; Real-Time Classification; Hyperparameter OptimizationAbstract
Effective monitoring of paralyzed individuals is crucial for ensuring their safety and well-being, particularly in detecting falls and abnormal postural states. This research proposes a Particle Swarm Optimization (PSO)-driven Deep Maxout Network (DMN) to enhance the accuracy and efficiency of human posture recognition. The proposed system utilizes RGB images from the Fall Detec-tion Dataset, which are preprocessed through resizing, normalization, data augmentation, and bounding box transformations. The DMN model, enhanced with Maxout activation, is employed for robust feature extraction, ensuring superior discrimination of pos-tural states. Additionally, PSO is integrated for hyperparameter optimization, dynamically fine-tuning parameters to improve classi-fication performance. The optimized DMN model achieved an accuracy of 96.4%, outperforming conventional classifiers. Further-more, PSO-driven optimization significantly reduced computational complexity, ensuring faster convergence and improved general-ization. Comparative analysis shows that the optimized DMN exhibits a lower inference time (6.1 ms) than traditional models. Ad-ditionally, ROC-AUC analysis yields a score of 0.98, highlighting the model’s strong discriminative capability in distinguishing postural states. The proposed PSO-DMN framework presents a reliable and efficient approach for paralyzed person monitoring, offer-ing real-time posture recognition with high accuracy. The system’s ability to detect falls, classify different postural states, and oper-ate efficiently in real-time settings makes it a promising solution for healthcare applications, particularly in home and assisted-living environments.
References
- Makanyadevi, K., Midhunaa, V. S., Swetha, M., & Thrishma, B. A. (2024, February). Survey on Health Care Monitoring of Paralyzed People. In 2024 IEEE International Conference on Big Data & Machine Learning (ICBDML) (pp. 290-294). IEEE. https://doi.org/10.1109/ICBDML60909.2024.10577416.
- Stefanov, D. H., Bien, Z., & Bang, W. C. (2004). The smart house for older persons and persons with physical disabilities: structure, technology arrangements, and perspectives. IEEE transactions on neural systems and rehabilitation engineering, 12(2), 228-250. https://doi.org/10.1109/TNSRE.2004.828423.
- Ayomide, Findlay. "Sitting Player Confrontation Passing Promises Anywhere Reconsideration Operation Trying Developed Remainder During Ma-ligned Indepence Important."
- Etinosa, Chiron. "Hughes Slugger Throughout Lawrence Combines Paralyzed Maryls Research Program Donated Addition Dozens Sourncentral Newsmen Transition."
- Noordeloos, Anouk. "Disabled Design Quest: Using design justice and crip hacking to design a 3D printing toolkit to increase creative confidence of disabled people." Master's thesis, University of Twente, 2025.
- Candraningtyas, Raphon Galuh, Andi Prademon Yunus, and Yit Hong Choo. "Human Fall Motion Prediction-A Review." (2024). https://doi.org/10.33093/ijoras.2024.6.2.8.
- Rahayu, Endang Sri, Eko Mulyanto Yuniarno, I. Ketut Eddy Purnama, and Mauridhi Hery Purnomo. "A combination model of shifting joint angle changes with 3D-deep convolutional neural network to recognize human activity." IEEE Transactions on Neural Systems and Rehabilitation Engi-neering (2024). https://doi.org/10.1109/TNSRE.2024.3371474.
- Rahayu, Endang Sri, Eko Mulyanto Yuniarno, I. Ketut Eddy Purnama, and Mauridhi Hery Purnomo. "Adaptive Threshold Filtering to Reduce Noise in Elderly Activity Classification Using Bi-LSTM." Jurnal Nasional Pendidikan Teknik Informatika: JANAPATI 13, no. 1 (2024): 48-57. https://doi.org/10.23887/janapati.v13i1.76064.
- Kamboj, Abhi, and Minh Do. "A Survey of IMU Based Cross-Modal Transfer Learning in Human Activity Recognition." arXiv preprint arXiv:2403.15444 (2024).
- Lin, Yitai, Zhijie Wei, Wanfa Zhang, Xiping Lin, Yudi Dai, Chenglu Wen, Siqi Shen, Lan Xu, and Cheng Wang. "HmPEAR: A Dataset for Hu-man Pose Estimation and Action Recognition." In Proceedings of the 32nd ACM International Conference on Multimedia, pp. 2069-2078. 2024. https://doi.org/10.1145/3664647.3681055.
- Karim, A. A., Mat Isa, N. A., & Lim, W. H. (2020). Modified Particle Swarm Optimization With Effective Guides. IEEE Access, 8, 188699–188725. https://doi.org/10.1109/ACCESS.2020.3030950.
- Ayman, A. A., Mohamed, H., Antoun, M., Mohamed, S. E., Amr, H., Talaat, Y., & Elashmawi, W. H. (2024). Optimized Deep Learning Models Using Particle Swarm Intelligence for MindMend, Stroke Rehabilitation System. 76–83. https://doi.org/10.1109/MIUCC62295.2024.10783610.
- PSO-PINN: Physics-Informed Neural Networks Trained with Particle Swarm Optimization. (2022).
- Basinayak, A., Pattanshetti, B., Chougule, K., Gaddigoudar, K., Shahapur, S. S., & Harchirkar, S. S. (2024). Beyond Boundaries: Progressive Health Monitoring for Paralysis. https://doi.org/10.1109/ICECCC61767.2024.10593934.
- Wu, D., Wu, D., Warwick, K., Ma, Z., Gasson, M. N., Burgess, J. G., Pan, S., & Aziz, T. Z. (2010). Prediction of Parkinson’s disease tremor onset using a radial basis function neural network based on particle swarm optimization. International Journal of Neural Systems, 20(2), 109–116. https://doi.org/10.1142/S0129065710002292.
- Pancholi, S., Everett, T. H., & Duerstock, B. S. (2024). Advancing spinal cord injury care through non-invasive autonomic dysreflexia detection with AI. Dental Science Reports, 14. https://doi.org/10.1038/s41598-024-53718-5.
- Aggogeri, F. (2020). Special Collection on advances in rehabilitation engineering with robotics and mechatronic devices. Advances in Mechanical Engineering, 12(8), 168781402094908. https://doi.org/10.1177/1687814020949086.
- Morales, R., Somolinos, J. A., Fernández-Caballero, A., & Ferraresi, C. (2018). Rehabilitation Robotics and Systems. Journal of Healthcare Engi-neering, 2018, 5370127. https://doi.org/10.1155/2018/5370127.
- Mohammed, S., Park, H. W., Park, C. H., Amirat, Y., & Argall, B. D. (2017). Special Issue on Assistive and Rehabilitation Robotics. Autonomous Robots, 41(3), 513–517. https://doi.org/10.1007/s10514-017-9627-z.
- Granheim, Roy Erling. "Enhancing Exercise Recognition: Integrating Advanced Deep Learning Models for Human Activity Recognition." Master's thesis, Norwegian University of Life Sciences, 2024.
- Afuan, L., & Isnanto, R. R. (2025). A comparative study of machine learning algorithms for fall detection in technology-based healthcare system: Analyzing SVM, KNN, decision tree, random forest, LSTM, and CNN. In E3S Web of Conferences (Vol. 605, p. 03051). EDP Sciences. https://doi.org/10.1051/e3sconf/202560503051.
- Granheim, Roy Erling. "Enhancing Exercise Recognition: Integrating Advanced Deep Learning Models for Human Activity Recognition." Master's thesis, Norwegian University of Life Sciences, 2024.
- Ye, X., Kevin, I., & Wang, K. (2024). Deep generative domain adaptation with temporal relation attention mechanism for cross-user activity recog-nition. Pattern Recognition, 156, 110811. https://doi.org/10.1016/j.patcog.2024.110811.
- Gyamenah, Pius, Hari Iyer, Heejin Jeong, and Shenghan Guo. "CLUMM: Contrastive Learning for Unobtrusive Motion Monitoring." Sensors 25, no. 4 (2025): 1048. https://doi.org/10.3390/s25041048.
- Romeo, Laura. "Vision devices and intelligent systems for monitoring the well-being of humans in healthcare and manufacturing." (2024).
- Ye, Xiaozhou, and Kevin I-Kai Wang. "Deep Generative Domain Adaptation with Temporal Attention for Cross-User Activity Recogni-tion." arXiv preprint arXiv:2403.17958 (2024). https://doi.org/10.1016/j.patcog.2024.110811.
- Haresamudram, Harish, Irfan Essa, and Thomas Plötz. "A Washing Machine is All You Need? On the Feasibility of Machine Data for Self-Supervised Human Activity Recognition." In 2024 International Conference on Activity and Behavior Computing (ABC), pp. 1-10. IEEE, 2024. https://doi.org/10.1109/ABC61795.2024.10651688.
- Hassan, Muhammad, Tom Kelsey, and Fahrurrozi Rahman. "Adversarial AI applied to cross-user inter-domain and intra-domain adaptation in hu-man activity recognition using wireless signals." Plos one 19, no. 4 (2024): e0298888. https://doi.org/10.1371/journal.pone.0298888.
Downloads
How to Cite
Received date: March 24, 2025
Accepted date: April 20, 2025
Published date: April 27, 2025