Denetimsiz Anomali Tespiti Yaklaşımı ile Düşme Algılama

Yaşlı nüfusunun hızla artması ve yaşlılığa bağlı olarak karşılaşılan fiziksel, duyusal ve bilişsel gerilemeler, düşmeyi her geçen gün büyüyen bir problem olarak karşımıza çıkarmakta ve düşme tespiti çalışmalarının hız kazanmasına sebep olmaktadır. Günlük aktivitelerin düşmeden ayırt edilmesinden ibaret olan düşme tespiti probleminde, denetimli öğrenme yaklaşımları kullanılmasına rağmen, düşmenin nadir rastlanan ve çok farklı biçimlerde karşılaşılabilen bir olay olması genel bir model elde edilmesine izin vermemektedir. Bu çalışmada denetimsiz anomali tespiti ile düşmenin belirlenmesi önerilmektedir. Denetimsiz öğrenme modelinin elde edilmesinde ve model vasıtasıyla düşmenin tespitinde 35 tip düşme ve 44 tip günlük aktiviteye sahip kapsamlı bir veri setinden faydalanılmıştır. Denetimsiz öğrenme yöntemi olan Gauss karışım modelinin eğitiminde, günlük aktivitelerden toplanan 3-eksen ivmeölçer sinyallerinden elde edilen öznitelikler kullanılmıştır. Test aşamasında model, düşme ve günlük aktivite verileri ile karşılaşmış, modele göre olasılığı çok düşük olan veriler anomali, dolayısıyla düşme olarak kabul edilmiştir. Testlerde düşmeler %90,5 civarında doğru olarak tespit edilmiştir. Sonuçlar düşmenin anomali tespiti yaklaşımları ile belirlenebileceğini ve makine öğrenmesi modelinin elde edilmesi için yalnız günlük aktivite verilerinin yeterli olduğu yaklaşımını doğrulamaktadır.

Fall Detection Using Unsupervised Anomaly Detection Approach

The rapid increase in the elderly population and the physical, sensory, and cognitive declines encountered due to old age cause falling as a growing problem day by day and induce fall detection studies to accelerate. Although supervised learning approaches are used in the fall detection problem, which consists of distinguishing daily activities from falling, it does not allow obtaining a general model because falling is a rare event that can be encountered in many different ways. This study proposes determining the fall with unsupervised anomaly detection. A comprehensive data set with 35 types of falls and 44 types of daily activities was used to obtain the unsupervised learning model and to detect falls through the model. In the training of the Gaussian mixture model, which is an unsupervised learning method, features obtained from 3-axis accelerometer signals collected from daily activities were used. During the test phase, the model was subject to fall and daily activity data, and the data with very low probabilities according to the model were accepted as anomalies, therefore falling. In the tests, the falls were detected with an accuracy of around 90.5% and the results were compared with other studies. The results confirm that the fall can be detected by anomaly detection approaches and that only daily activity data is sufficient to obtain the machine learning model.

PDF

___

[1] World Health Organization, “Falls.” [Online]. Available: https://www.who.int/news-room/factsheets/detail/falls. [Accessed: 12-Jun-2021].
[2] R. Rajagopalan, I. Litvan, and T. P. Jung, “Fall prediction and prevention systems: Recent trends, challenges, and future research directions,” Sensors (Switzerland), vol. 17, no. 11, p. 2509, Nov. 2017.
[3] United Nations, “World Population Ageing 2019.” [Online]. Available: https://www.un.org/en/development/desa/population/publications/pdf/ageing/WorldPopulationAgeing 2019-Highlights.pdf. [Accessed: 13-Jun-2021].
[4] Türkiye İstatistik Kurumu, “İstatistiklerle Yaşlılar, 2020.” [Online]. Available: https://data.tuik.gov.tr/Bulten/Index?p=Istatistiklerle-Yaslilar-2020-37227. [Accessed: 13-Jun-2021].
[5] K.-C. Liu, C.-Y. Hsieh, H.-Y. Huang, S. J.-P. Hsu, and C.-T. Chan, “An Analysis of Segmentation Approaches and Window Sizes in Wearable-Based Critical Fall Detection Systems With Machine Learning Models,” IEEE Sens. J., vol. 20, no. 6, pp. 3303–3313, Mar. 2020.
[6] O. Kerdjidj, N. Ramzan, K. Ghanem, A. Amira, and F. Chouireb, “Fall detection and human activity classification using wearable sensors and compressed sensing,” J. Ambient Intell. Humaniz. Comput., vol. 11, no. 1, pp. 349–361, 2020.
[7] Y.-H. Nho, J. G. Lim, and D.-S. Kwon, “Cluster-Analysis-Based User-Adaptive Fall Detection Using Fusion of Heart Rate Sensor and Accelerometer in a Wearable Device,” IEEE Access, vol. 8, pp. 40389– 40401, 2020.
[8] M. Saleh and R. L. B. Jeannes, “Elderly Fall Detection Using Wearable Sensors: A Low Cost Highly Accurate Algorithm,” IEEE Sens. J., vol. 19, no. 8, pp. 3156–3164, 2019.
[9] C. Wang et al., “Low-Power Fall Detector Using Triaxial Accelerometry and Barometric Pressure Sensing,” IEEE Trans. Ind. Informatics, vol. 12, no. 6, pp. 2302–2311, 2016.
[10] X. Wang, J. Ellul, and G. Azzopardi, “Elderly Fall Detection Systems: A Literature Survey,” Front. Robot. AI, vol. 7, Jun. 2020.
[11] Y. M. Galvao, L. Portela, J. Ferreira, P. Barros, O. A. De Araujo Fagundes, and B. J. T. Fernandes, “A Framework for Anomaly Identification Applied on Fall Detection,” IEEE Access, vol. 9, pp. 77264– 77274, 2021.
[12] J. Nogas, S. S. Khan, and A. Mihailidis, “DeepFall: Non-Invasive Fall Detection with Deep SpatioTemporal Convolutional Autoencoders,” J. Healthc. Informatics Res., vol. 4, no. 1, pp. 50–70, 2020.
[13] M. Saleh, M. Abbas, and R. B. Le Jeannes, “FallAllD: An Open Dataset of Human Falls and Activities of Daily Living for Classical and Deep Learning Applications,” IEEE Sens. J., vol. 21, no. 2, pp. 1849– 1858, Jan. 2021.
[14] G. Vavoulas, C. Chatzaki, T. Malliotakis, M. Pediaditis, and M. Tsiknakis, “The MobiAct dataset: Recognition of activities of daily living using smartphones,” ICT4AWE 2016 - 2nd Int. Conf. Inf. Commun. Technol. Ageing Well e-Health, Proc., pp. 143–151, 2016.
[15] A. Sucerquia, J. D. López, and J. F. Vargas-Bonilla, “SisFall: A fall and movement dataset,” Sensors (Switzerland), vol. 17, no. 1, 2017.
[16] A. T. Özdemir, “An analysis on sensor locations of the human body for wearable fall detection devices: Principles and practice,” Sensors (Switzerland), vol. 16, no. 8, 2016.
[17] E. Casilari, J. A. Santoyo-Ramón, and J. M. Cano-García, “UMAFall: A Multisensor Dataset for the Research on Automatic Fall Detection,” Procedia Comput. Sci., vol. 110, pp. 32–39, 2017.
[18] S. Usmani, A. Saboor, M. Haris, M. A. Khan, and H. Park, “Latest Research Trends in Fall Detection and Prevention Using Machine Learning: A Systematic Review,” Sensors, vol. 21, no. 15, p. 5134, Jul. 2021.
[19] W. Xiong et al., “Accurate Fall Detection Algorithm Based on SBPSO-SVM Classifier,” ACM Int. Conf. Proceeding Ser., pp. 83–86, 2018.
[20] F. Hussain, F. Hussain, M. Ehatisham-Ul-Haq, and M. A. Azam, “Activity-Aware Fall Detection and Recognition Based on Wearable Sensors,” IEEE Sens. J., vol. 19, no. 12, pp. 4528–4536, 2019.
[21] E. Casilari-Pérez and F. García-Lagos, “A comprehensive study on the use of artificial neural networks in wearable fall detection systems,” Expert Syst. Appl., vol. 138, 2019.
[22] X. Wu, Y. Zheng, C.-H. Chu, L. Cheng, and J. Kim, “Applying deep learning technology for automatic fall detection using mobile sensors,” Biomed. Signal Process. Control, vol. 72, p. 103355, Feb. 2022.
[23] M. M. Islam et al., “Deep Learning Based Systems Developed for Fall Detection: A Review,” IEEE Access, vol. 8, pp. 166117–166137, 2020.
[24] S. S. Khan and B. Taati, “Detecting unseen falls from wearable devices using channel-wise ensemble of autoencoders,” Expert Syst. Appl., vol. 87, pp. 280–290, 2017.
[25] S. Zhao, W. Li, and J. Cao, “A user-adaptive algorithm for activity recognition based on K-means clustering, local outlier factor, and multivariate gaussian distribution,” Sensors (Switzerland), vol. 18, no. 6, 2018.
[26] J. A. Santoyo-Ramón, E. Casilari, and J. M. Cano-García, “A study of one-class classification algorithms for wearable fall sensors,” Biosensors, vol. 11, no. 8, 2021.
[27] L. Li, R. J. Hansman, R. Palacios, and R. Welsch, “Anomaly detection via a Gaussian Mixture Model for flight operation and safety monitoring,” Transp. Res. Part C Emerg. Technol., vol. 64, pp. 45–57, Mar. 2016.
[28] L. Chen, R. Li, H. Zhang, L. Tian, and N. Chen, “Intelligent fall detection method based on accelerometer data from a wrist-worn smart watch,” Measurement, vol. 140, pp. 215–226, Jul. 2019.