Yürüyüş Parametrelerinden Poma-G Skorunun Tahmini

Öz

POMA testi, bireylerin fonksiyonel hareket ve aktiviteleri gerçekleştirme yeteneğini değerlendirmek için kullanılan bir test türüdür. Fizik tedavi ve rehabilitasyon alanında bireylerin mobilitesini değerlendirmek ve iyileştirme alanlarını belirlemek için yaygın olarak kullanılır. POMA değerlendirmesi genellikle, bu aracı kullanmayı öğrenmiş uzmanlar tarafından yapılmaktadır. POMA skoru, değerlendirmeyi gerçekleştiren uzmana bağlı olarak farklılık gösterebilir çünkü farklı uzmanlar farklı değerlendirme yaklaşımlarına sahip olabilir ve bir uzmanın değerlendirmesi de zaman içinde değişebilir. Bu değerlendirmelerdeki farklılıklar, POMA skorunda varyasyonlara ve güvenilirliğe etki edebilir. Öte yandan, 3B yürüyüş analizi nesnel bir nitelik taşır ve bireylerin mobilitesini değerlendirmek için daha güvenilir bir yöntem sunar. Bu çalışma, nesnel olarak elde edilen mekansal ve zamansal yürüyüş parametrelerine dayanarak POMA testinin yürüyüş bölümünde elde edilen ölçümleri (POMA-G) tahmin etmeyi amaçlamıştır. Bunun için 44 yetişkinin yürüyüş parametrelerinden oluşan bir veri seti kullanılmıştır. POMA-G skorları, biri uzman diğeri ise test hakkında bilgili ancak daha az deneyime sahip olan iki fizyoterapist tarafından değerlendirilmiştir. Çalışmada tahmin modellerini analiz etmenin yanında fizyoterapistlerin değerlendirme güvenilirliği de incelenmiştir.

Anahtar Kelimeler

• Machine Learning
• POMA-G Prediction
• Reliability Analysis

Prediction of POMA-G Score from Spatiotemporal Gait Parameters

Abstract

POMA test, is a type of test used to evaluate an individual's ability to perform functional movements and activities and is commonly used in the field of rehabilitation and physical therapy to assess an individual's mobility and identify areas for improvement. POMA evaluation is typically conducted by experts who are trained in using this tool to assess mobility and balance in individuals. However, it is possible that the POMA score may vary depending on the expert who conducts the evaluation because different experts may have different approaches to evaluation, and an individual expert's evaluation may also vary over time. These differences in the evaluation may lead to variations in the POMA score and may impact its reliability. Gait analysis, on the other hand, is objective in nature, providing a more reliable way of assessing the mobility of individuals. This study aimed to predict the measurements obtained in the gait portion of the POMA test (POMA-G) based on objectively obtained spatiotemporal gait parameters. To do this, a dataset consisting of gait parameters from 44 older adults was used. The POMA-G scores were rated by two physiotherapists, one of whom was an expert and the other who was familiar with the test but less experienced. The study also included an analysis of the reliability of physiotherapists’ assessment and the proposed prediction models.

Keywords

• Machine Learning
• POMA-G Prediction
• Reliability Analysis
• Tinetti Mobility Test

Kaynakça

1. [1] P. E. Caicedo, C. F. Rengifo, L. E. Rodriguez, W. A. Sierra, and M. C. Gómez, “Dataset for gait analysis and assessment of fall risk for older adults,” Data Br., vol. 33, p. 106550, Dec. 2020, doi: 10.1016/J.DIB.2020.106550.
2. [2] M. E. Tinetti, “Performance-oriented assessment of mobility problems in elderly patients,” J. Am. Geriatr. Soc., vol. 34, no. 2, pp. 119–126, 1986, doi: 10.1111/J.1532-5415.1986.TB05480.X.
3. [3] M. W. Rivolta et al., “Evaluation of the Tinetti score and fall risk assessment via accelerometry-based movement analysis,” Artif. Intell. Med., vol. 95, pp. 38–47, Apr. 2019, doi: 10.1016/j.artmed.2018.08.005.
4. [4] S. Köpke and P. G. Meyer, “The Tinetti test,” Z. Gerontol. Geriatr., vol. 39, no. 4, pp. 288–291, Aug. 2006, doi: 10.1007/S00391-006-0398-Y.
5. [5] D. Sethi, S. Bharti, and C. Prakash, “A comprehensive survey on gait analysis: History, parameters, approaches, pose estimation, and future work,” Artif. Intell. Med., vol. 129, p. 102314, Jul. 2022, doi: 10.1016/J.ARTMED.2022.102314.
6. [6] K. S. Al-Zahrani and M. O. Bakheit, “A historical review of gait analysis,” Neurosci. J., vol. 13, no. 2, pp. 105–108, 2008.
7. [7] İ. Kosesoy, C. Öz, F. Aslan, F. Köroğlu, and M. Yığılıtaş, “Reliability and validity of an innovative method of ROM measurement using Microsoft Kinect V2,” Pamukkale Univ. J. Eng. Sci., vol. 24, no. 5, pp. 915–920, Oct. 2018, doi: 10.5505/pajes.2017.65707.
8. [8] S. Aich et al., “Design of a Machine Learning-Assisted Wearable Accelerometer-Based Automated System for Studying the Effect of Dopaminergic Medicine on Gait Characteristics of Parkinson’s Patients,” J. Healthc. Eng., vol. 2020, 2020, doi: 10.1155/2020/1823268.

9. [9] H. Lee, S. J. Sullivan, and A. G. Schneiders, “The use of the dual-task paradigm in detecting gait performance deficits following a sports-related concussion: A systematic review and meta-analysis,” J. Sci. Med. Sport, vol. 16, no. 1, pp. 2–7, Jan. 2013, doi: 10.1016/j.jsams.2012.03.013.
10. [10] I. Kosesoy and C. Oz, “Acquiring Kinematics of Lower extremity with Kinect,” Avrupa Bilim ve Teknol. Derg., no. 32, pp. 92–100, 2021.
11. [11] R. A. States, J. J. Krzak, Y. Salem, E. M. Godwin, A. W. Bodkin, and M. L. McMulkin, “Instrumented gait analysis for management of gait disorders in children with cerebral palsy: A scoping review,” Gait \& Posture, vol. 90, pp. 1–8, 2021.
12. [12] M. Pistacchi et al., “Gait analysis and clinical correlations in early Parkinson’s disease,” Funct. Neurol., vol. 32, no. 1, p. 28, 2017.
13. [13] J. M. Guralnik, L. Ferrucci, E. M. Simonsick, M. E. Salive, and R. B. Wallace, “Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability,” N. Engl. J. Med., vol. 332, no. 9, pp. 556–562, Mar. 1995, doi: 10.1056/NEJM199503023320902.
14. [14] M. F. Folstein, S. E. Folstein, and P. R. McHugh, “‘Mini-mental state’. A practical method for grading the cognitive state of patients for the clinician,” J. Psychiatr. Res., vol. 12, no. 3, pp. 189–198, 1975, doi: 10.1016/0022-3956(75)90026-6.
15. [15] E. Acar and M. S. Özerdem, “On a yearly basis prediction of soil water content utilizing sar data: a machinelearning and feature selection approach,” Turkish J. Electr. Eng. Comput. Sci., vol. 28, no. 4, pp. 2316–2330, 2020.
16. [16] J. Platt, “Sequential Minimal Optimization: A Fast Algorithm for Training Support Vector Machines,” 1998. [Online]. Available: https://www.microsoft.com/en-us/research/publication/sequential-minimal-optimization-a-fast-algorithm-for-training-support-vector-machines/
17. [17] S. Zhang, D. Cheng, Z. Deng, M. Zong, and X. Deng, “A novel kNN algorithm with data-driven k parameter computation,” Pattern Recognit. Lett., vol. 109, pp. 44–54, 2018.
18. [18] L. Breiman, “Random forests,” Mach. Learn., vol. 45, no. 1, pp. 5–32, 2001, doi: 10.1023/A:1010933404324.
19. [19] G. Biau and E. Scornet, “A random forest guided tour,” Test, vol. 25, no. 2, pp. 197–227, 2016.
20. [20] K. Kira and L. A. Rendell, “A practical approach to feature selection,” in Machine learning proceedings 1992, Elsevier, 1992, pp. 249–256.
21. [21] R. J. Urbanowicz, M. Meeker, W. La Cava, R. S. Olson, and J. H. Moore, “Relief-based feature selection: Introduction and review,” J. Biomed. Inform., vol. 85, pp. 189–203, 2018.
22. [22] A. G. Karegowda, A. S. Manjunath, and M. A. Jayaram, “Comparative study of attribute selection using gain ratio and correlation based feature selection,” Int. J. Inf. Technol. Knowl. Manag., vol. 2, no. 2, pp. 271–277, 2010.
23. [23] M. Hall, E. Frank, G. Holmes, B. Pfahringer, P. Reutemann, and I. H. Witten, “The WEKA data mining software: an update,” ACM SIGKDD Explor. Newsl., vol. 11, no. 1, pp. 10–18, 2009.
24. [24] J. P. Weir, “Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM,” J. Strength \& Cond. Res., vol. 19, no. 1, pp. 231–240, 2005.