Selin AYDIN FANDAKLI, Halil Ibrahim OKUMUS

Ataletsel Ölçüm Sensörü ile Ayak Protezi için Ayak Hareketlerinin Sınıflandırılması

Günümüzde Atalet Ölçüm Birimleri alt ekstremite protez çalışmalarında kontrol amaçlı kullanılmaktadır. Bu yazıda dorsifleksiyon, plantarfleksiyon, inversiyon ve eversiyon gibi ayak hareketlerinin analizi ve sınıflandırılması ile ilgili bir uygulama sunulmuştur. Bu çalışma, robotik ayak sistemi için hareket modelini tanımak ve anormal yürüme koşullarına uyum sağlamak için ayak hareketlerinin sınıflandırılmasını amaçlamaktadır. 20-34 yaşları arasındaki on bir gönüllünün ayağının metatarsalına bağlı IMU sensöründen gelen hareket verileriyle dokuz parametre ölçülür. Bu parametrelerden toplam, ortalama, standart sapma, kovaryans, çarpıklık ve basıklık gibi istatistiksel özellikler çıkartılarak boyut küçültülür. Karar Ağacı, Doğrusal Ayrım Analizi, Naïve Bayes Sınıflandırıcı, K-En Yakın Komşu ve Destek Vektör Makinesi gibi sınıflandırıcılar ile her kişi için ayrı ayrı sınıflandırma işlemi yapılır. 11 gönüllü için elde edilen sınıflandırma doğruluklarının ortalaması alınmış ve en yüksek doğruluk KNN için % 97.2 ile elde edilmiştir.

Anahtar Kelimeler:

protez, jiroskop, destek vektör makinesi, k-en yakın komşu, ivmeölçer, manyetometre

Classification of the Foot Movements with Inertial Measurement Sensor for Ankle-Foot Prosthesis

Today, Inertial Measurement Units is used for control in lower extremity prosthesis studies. In this article, an application related to the analysis and classification of foot movements such as dorsiflexion, plantarflexion, inversion and eversion is presented. This study aims to perform the classification of foot movements to recognize the movement pattern and to adapt to abnormal walking conditions for the robotic foot system. Nine parameters are measured with motion data from the IMU sensor connected to the metatarsal of the foot from eleven volunteers aged 20-34 years. Size is reduced by extracting statistical properties such as sum, mean, standard deviation, covariance, skewness and kurtosis from these parameters. Classification process is performed with classifiers such as Decision Tree, Linear Discriminant Analysis, Naïve Bayes Classifier, K-Nearest Neighbor and Support Vector Machine separately for each person. The classification accuracies obtained for 11 volunteers are averaged and the highest accuracy is obtained with 97.2% for KNN.

Keywords:

prosthesis, gyroscope, support vector machine, k-nearest neighbor, accelerometer, magnetometer,

PDF

___

Dev, V. A., Eden, M.R. (2019). Computer Aided Chemical Engineering, Gradient Boosted Decision Trees for Lithology Classification, Vol. 47, 113-118. doi: 10.1016/ B978-0-12-818597-1.50019-9.
Fleck, J. J., Peters, R. A., Zelik, K. E. (2018). IMU-Based gait analysis in lower limb prosthesis users: comparison of step demarcation algorithms. Gait&Posture, 64, 30-37. doi: 10.1016/j.gaitpost.2018.05.025
Gao, F., Liu, G., Liang, F., Liao, W. H. (2020). IMU-Based locomotion mode identification for transtibial prosthese, orthoses, and exoskeletons. IEEE Transactions on Neural Systems and Rehabilitation Engineering, Vol. 28, No.6, pp.1334-1343.
Jiang, X., Chu, H. T., Xiao, Z. G., Merhi, L. K., Menon, C. (2016) Ankle Positions Using Force Myography: an Exploratory Investigation, IEEE Healthcare Innovation Point-Of-Care Technologies Conference, 29-32. doi: 10.1109/HIC.2016.7797689.
Kotu, V., Deshpande, B. (2019). Data Mining (Second Edition), Chaapter 4: Classification, 65-163. doi: 10.1016/B978-0-12-814761-0.00004-6.
Li, H., Derrode, S., Pieczynski, W. (2019) An adaptive and on-line IMU-based locomotion activity classification using a triplet Markov mode, Neurocomputing, 362, 94-105.
Malek, S., Hui, C., Aziida, N., Cheen, S., Toh, S., Millow, P. (2019). Encyclopedia of Bioinformatics and Computational Biology, Ecosystem monitoring through predictive modeling, Volume 3, 1-8. doi: 10.1016/B978-0-12-809633-8.20060-5.
McClean, S. I. (2003). Encyclopedia of Physical Science and Technology (Third Edition), Data Mining and Knowledge Discovery, 229-246. doi: 10.1016-B0-12-227410-5/00845-0.
McDermott, M. M., Greenland, P., Liu, K. (2001). Leg symptoms in peripheral arterial disease: associated clinical characteristics and functional impairment. JAMA 2001, 286(13), 1599-606. doi: 10.1001/jama.286.13.1599.
Meffen, A., Pepper, C. J., Sayers, R. D., Gray, L. J. (2020). Epidemiology of major lower limb amputation amputation using routinely collected electronic health data in the UK: a systematic review protocol. BMJ Open 2020, 10(6):e037053. doi:10.1136/bmjopen-2020-037053.
Misra, S., Li, H. (2020). Machine Learning for Subsurface Characterization, Chapter 9- Noninvasive fracture characterization based on the classification on sonic wave travel times, 243-287. doi: 10.1016/B978-0-12-817736-5.00009-0.
MTech, S. M., Rajput, D. S. (2019). Deep Learning and Parallel Computing Environment for Bioengineering Systems, Nonlinear Decision Tree Regression, 153-164. doi:10.1016/B978-0-12-816718-2.00016-6.
Mushtaq, M. S., Mellouk, A. (2017). Quality of Experience Paradigm in Multimedia Services, 2-Methodologies for Subjective Video Streaming QoE Assessment, 27-57. doi: 10.1016/B978-1-78548-109-3.500002-3.
Parkka, J., Ermes, M., Korpipaa, P., Peltola, J., Korhonen, I. (2006). Activity classification using realistic data from wearable sensors. IEEE Transaction on Information Technology in Biomedicine, Vol. 10, 119-128.
Pisner, D. A., Schnyer, D. M. (2020). Machine Learning Chapter 6- Support vector machine, 101-121. doi: 10.1016/ B978-0-12-815739-8.00006-7.
Seel, T., Raisch, J., Schauer, T. (2014). IMU-Based Joint Angle Measurement for Gait Analysis. Sensors, 14, 6891-6909. doi: 10.3390/s140406891.
Spoden, M., Nimptsch, U., Mansky, T. (2019). Amputation rates of the lower limb by amputation level - observational study using German national hospital discharge data from 2005-2015. BMC Health Services Research, 19(8), 1-9. doi.org/10.1186/s12913-018-3759-5.
Stanimirova, I., Daszyowski, M., Walczak B. (2013). Data Handling in Science and Technology, Chapter 8-Robust methods in analysis of multivariate food chemistry data, Volume 28, 315-340. doi:10/1016/B978-0-444-59528-7.00008-9.
Steffen, L. M., Duprez, D. A., Boucher, J. L. et al. (2008) Management of peripheral arterial disease. Diabetes Spectrum, 21:171-7.
Subasi, A. (2020). Practical Machine Learning for Data Analysis Using Python, Chapter 3-Machine learning techniques. 91-202. doi: 10/1016/B978-0-12-821379.7.00003-5.
Subasi, A. (2020). Artificial Intelligence in Precision Health, Chapter 11- Use of artificial intelligence in Alzheimer’s disease detection, 257-278. doi:10/1016/B978-0-12-817133-2.00011-2.
Subasi, A., Khateeb, K., Brahimi, T., Sarriete, A. (2020). Innovation in Health Informatics, Chapter-5 Human activity recognition using machine learning methods in a smart healthcare environment, 123-144. doi: 10.1016/B978-0-12-819043-2.00005-8.
Quraishi, M. A., Ishak, A. J., Ahmad, S., Hassan, M. K., Qurishi, M. A., Ghapanchizadeh, H., Alamri, A. (2016) Classification of ankle joint movements based on surface electromyography signals for rehabilitatin robot applications. Medical and Biological Engineering, 55(5). doi: 10.1007/s11517-016-1551-4.
Vaibhaw, J. S., Pattnaik, P. K. (2020), An Industrial IoT Approach for Pharmaceutical Industry Growth, Chapter 2 Brain- computer interfaces and their applications, 31-54. doi:10.1016/B978-0-12-821326-1.00002-4.
Witten, I. H., Frank, E., Hall, M. A., Pal, C. J. (2017) Data Mining (Fourth Edition), Chapter 8-Data transformations, 285-334. doi: 10.1016/B978-0-12-804291-5.00008-8.
Yuliani, S., Saputra M. (2016). Collaboration of Kalman Filter with Complementary Filter to Optimize the Results of Gyroscope and Accelerometer Sensors (In Bahasa Indonesia) in Prosiding Seminar Nasional Rekayasa dan Desain Itenas, 1, 1-6.
Ziegler-Graham, K., MacKenzie, E. J., Ephraim, P. L., Travison, T. G., and Brookmeyer, R. (2008). Estimating the prevalence of limb loss in the United States: 2005 to 2050. Arch. Phys. Med. Rehabil., 89(3), 422-429. doi:10.1016/j.apmr.2007.11.005.
Figure-1: https://www.robotistan.com/sparkfun-9-dof-imu-9-degrees-of-freedom-imu-breakout-lsm9ds1, (Date Accessed: 25 August 2021)
Figure-2: https://cdn.sparkfun.com/assets/learn_tutorials/3/7/3/LSM9DS1_Datasheet.pdf, (Date Accessed:25 August 2021)