2025 Vol.8 Oct.N05

• Title: Research on Intelligent Fall Prevention System for Senile Syndrome
• Name: Ning Zhao
• Company: Shenyang Pude Hospital of Traditional Chinese Medicine Co., Ltd ，shenyang 110000，China
• Abstract:

  Most intelligent fall prevention systems only use wearable sensors or cameras to obtain data, and the collected data can be directly used for simple pattern matching or judgment based on fixed thresholds to realize fall detection. The classification accuracy is poor due to the lack of effective dimensionality reduction processing for the characteristics of body sensing signals of the elderly. In this regard, an intelligent anti-fall system for senile syndrome is proposed. In hardware, VDSuit wearable inertial motion capture device with advantages of temperature compensation and anti-magnetic interference is selected, which contains 17 inertial sensors and can collect the centroid position data of all parts of human body in real time. The plantar pressure signal acquisition module composed of pressure insole sensor and main control chip is used to collect plantar pressure information. Taking the mean value, root mean square value and vector sum on the horizontal plane as characteristic parameters, it provides the basis for machine learning and training. By constructing a random forest model, the Gini index is used to measure the node impurity and calculate the feature importance score. According to the score, the features that contribute little to the fall detection are removed, and the data dimension is reduced. According to the classification principle of K-nearest neighbor method, the sample feature vector is taken as the feature space point, and K nearest neighbor samples are found by calculating the distance, and new sample categories are predicted according to most categories, and an early warning mechanism is constructed. In the experiment, the classification accuracy of the proposed method is verified. Through the test and comparison results, it is clear that when the proposed system is used for fall detection and early warning, the optimal classification coincidence degree of the system for fall behavior is 97.05%, which has ideal classification accuracy.

  
  

• Keyword: senile syndrome; Anti-fall system; System design; Fall detection;
• DOI: 10.12250/jpciams2025091008
• Citation form: Ning Zhao.Research on Intelligent Fall Prevention System for Senile Syndrome[J]. Computer Informatization and Mechanical System,2025,Vol.8,pp.

[1] Zhang Q , Bao X , Lin S F .Lightweight network for small target fall detection based on feature fusion and dynamic convolution[J].Journal of Real-Time Image Processing, 2024, 21(1):17.1-17.17.

[2] Ma Z , Wu X , Zhang X ,et al.Anti-Fixed-Interference Fall Detection Based on Doppler-Time-Range Maps Using Millimeter-Wave Radar[J].Sensors Journal, IEEE, 2025, 25(2):3358-3368.

[3] Zhang L , Liu Y A , Wang Q ,et al.A Fall Detection Device Based on Single Sensor Combined with Joint Features[J].Tsinghua Science and Technology, 2024, 30(2):695-707.

[4] Ranjith R .Towards Safer Aging: A Hybrid KNN Model for Pre-Impact Fall Detection Enhanced by Class Balancing[J].Journal of Information Systems Engineering and Management, 2025, 10(7s):480-493.

[5] Li J , Gao M , Wang P L B .Fall detection algorithm based on pyramid network and feature fusion[J].Evolving Systems, 2024, 15(5):1957-1970.

[6] Khan M Z , Althobaiti T , Almutiry M ,et al.VisiSafe: Real-Time Fall Detection for Visually Impaired People Using RF Sensing[J].Sensors Journal, IEEE, 2025, 25(3):5654-5667.

[7] Hidayat S S ,Fadhil, Mujahidin I , et al.Optimizing Fall Detection System as an Early Warning System for the Elderly to Enhance Quality of Life[J].Scientific Journal of Informatics, 2024, 11(2):255-262.

[8] Gunale K G , Mukherji P , Motade S N .Convolutional Neural Network-Based Fall Detection for the Elderly Person Monitoring[J].Journal of Advances in Information Technology, 2023, 14(6):1169-1176.

[9] Priya A A .Deep Learning Assisted Intelligent Fall Detection Mechanism: An IoT and Healthcare Integration of Elderly People[J].Panamerican Mathematical Journal, 2024, 35(2s):01-11.

[10] Panyin G , Guidong Z , Zhigang Z ,et al.Research on Fall Detection System Based on Commercial Wi-Fi Devices[J].ZTE Communications, 2023, 21(4):60-68.

[11] Weng A L , Nysther M , Haddara M ,et al.User Satisfaction with Fall Detection Systems in Smart Elderly Care[J].Procedia Computer Science, 2025, 256:583-593.

[12] Mithran E , Avinash S , Kumar M R ,et al.Elderly Fall Detection Model for Patient Care Using Improvised CNN[J].Communications in Computer and Information Science, 2025:275-291.

[13] Swarubini P J , Ganapathy N .Radar-based elderly fall detection using SPWVD and ResNet Network[J].current directions in biomedical engineering, 2024, 10(4):498-501.

[14] Eflek B .Radar-based Elderly Fall Detection Using Power Spectral Density Features Obtained by Different Methods[J] .2024 8th International Symposium on Innovative Approaches in Smart Technologies (ISAS), 2024:1-5.

[15] Zoellick J C , Lech S , O'Sullivan J L ,et al.Fall sensors, home emergency system,  and social service for≥75-year-olds living at home- a matched control intervention study[J].BMC Geriatrics, 2025, 25(1):1-14.