تحلیل الگوی باینری محلی سیگنال‌های فشار پا جهت تشخیص سکته مغزی

نویسندگان

1 دانشجوی کارشناسی ارشد، دانشکده مهندسی پزشکی، دانشگاه صنعتی سهند، تبریز، ایران

2 دانشیار، دانشکده مهندسی پزشکی، دانشگاه صنعتی سهند، تبریز، ایران

چکیده

در بیماران مبتلا به سکته مغزی به‌صورت عمومی مشکلات حرکتی و راه رفتن قابل مشاهده است که کیفیت زندگی آن‌ها را تحت تاثیر قرار می‌دهد. از این‌رو تشخیص دقیق سکته مغزی برای ارائه یک راهکار درمانی و توانبخشی موثر در این بیماران ضروری به نظر می‌رسد. با این حال، توسعه یک ابزار تشخیصی کم‌هزینه و غیرتهاجمی برای کاربردهای کلینیکی یک چالش بزرگ در این زمینه محسوب می‌شود. به همین جهت، در این مطالعه یک روش تشخیصی جدید سکته ایسکمیک بر پایه ویژگی‌های ساختاری سیگنال فشار کف پا و طبقه‌بند ماشین بردار پشتیبان ارائه شده است. در این روش، یک الگوی باینری محلی یکنواخت که از نمایش زمانی-فرکانسی سیگنال فشار کف پا استخراج شده است، برای اخذ ساختار محلی سیگنال در فضای دوبعدی و کمی‌سازی پایداری این الگو استفاده شده است. روش پیشنهادی به کمک سیگنال‌های ثبت شده از 36 فرد سالم و 46 بیمار مبتلا به سکته ایسکمیک در حین راه رفتن طبیعی فرد مورد ارزیابی قرار گرفته است. جهت ارائه تحلیل ناحیه‌ای، طبقه‌بندی با استفاده از کانال‌های مختلف کف پا انجام شده است. نتایج به‌دست آمده به میانگین صحت 99/66 درصد برای تشخیص سکته مغزی رسیده است. در ادامه، طی یک آزمایش مقایسه‌ای، پایداری و عدم تغییر نتایج روش پیشنهادی در برابر سنسورهای فشار نواحی مختلف کف‌ پا و پارامترهای تکنیکی روش الگوی باینری محلی نشان داده شده است. عملکرد روش پیشنهادی نشان می‌دهد که تحلیل الگوی باینری محلی سیگنال فشار کف ‌پا قادر است افراد سالم و بیماران مبتلا به سکته مغزی را به‌صورت موثری تفکیک نماید.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

Local Binary Pattern Analysis of Foot Pressure Signals for Stroke Detection

نویسندگان [English]

  • Arefeh Yagoubi 1
  • Peyvand Ghaderyan 2
1 M.Sc. Student, Faculty of Biomedical Engineering, Sahand University of Technology, Tabriz, Iran,
2 Associate Professor, Faculty of Biomedical Engineering, Sahand University of Technology, Tabriz, Iran,
چکیده [English]

Stroke patients generally exhibit trouble walking and moving, which affects their quality of life. Hence, an accurate diagnosis of stroke is important for providing an effective treatment and rehabilitation strategy. However, the development of a cost-effective and non-invasive diagnostic tool is a big challenge for clinical applications. To address this challenge, in this study, a new ischemic stroke detection has been proposed based on structural features of foot plantar pressure signals and support vector machine classifier. A local uniform binary pattern extracted from the time-frequency representation of pressure signals has been used to capture the local structure over two-dimensional space and quantify the stability of this pattern. The proposed method has been evaluated using the pressure signals recorded during normal walking tasks from 36 healthy controls and 46 Ischemic stroke patients. The classification has also been performed for different plantar channels to offer regional analysis. The obtained results have achieved a high average accuracy rate of 99.66% for stroke detection. Furthermore, the robustness of the proposed method against different plantar regions as well as technical parameters of the local binary pattern approach has been demonstrated in an experimental comparative study. The performance has confirmed that the local binary pattern analysis discriminates effectively stroke patients and healthy controls when foot plantar pressure signals are used.

کلیدواژه‌ها [English]

  • Machine learning
  • Ischemic stroke
  • Time-frequency plantar pressure features
  • automatic detection

مراجع

[1] Lundström, E., et al., Four-fold increase in direct costs of stroke survivors with spasticity compared with stroke survivors without spasticity: the first year after the event. Stroke, 2010. 41(2): p. 319-324.
[2] Norrving, B. and B. Kissela, The global burden of stroke and need for a continuum of care. Neurology, 2013. 80(3 Supplement 2): p. S5-S12.
[3] Rozanski, G.M., et al., Lower limb muscle activity underlying temporal gait asymmetry post-stroke. Clinical Neurophysiology, 2020. 131(8): p. 1848-1858.
[4] Guzik, A., et al., Relationship between observational Wisconsin gait scale, gait deviation index, and gait variability index in individuals poststroke. Archives of physical medicine and rehabilitation, 2019. 100(9): p. 1680-1687.
[5] Wang, Y., et al., Gait characteristics of post-stroke hemiparetic patients with different walking speeds. International journal of rehabilitation research. Internationale Zeitschrift fur Rehabilitationsforschung. Revue internationale de recherches de readaptation, 2020. 43(1): p. 69.
[6] Little, V.L., et al., Pelvic excursion during walking post-stroke: A novel classification system. Gait & posture, 2018. 62: p. 395-404.
[7] Li, M., et al., Gait analysis for post-stroke hemiparetic patient by multi-features fusion method. Sensors, 2019. 19(7): p. 1737.
[8] Valentini, F., et al., Repeatability and variability of baropodometric and spatio-temporal gait parameters–results in healthy subjects and in stroke patients. Neurophysiologie Clinique/Clinical Neurophysiology, 2011. 41(4): p. 181-189.
[9] Wang, W., et al. Evaluation of postural instability in stroke patient during quiet standing. in 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). 2017. IEEE.
[10] Echigoya, K., et al., Changes to foot pressure pattern in post-stroke individuals who have started to walk independently during the convalescent phase. Gait & Posture, 2021. 90: p. 307-312.
[11] Krishnan, S. and Y. Athavale, Trends in biomedical signal feature extraction. Biomedical Signal Processing and Control, 2018. 43: p. 41-63.
[12] Akan, A. and O.K. Cura, Time–frequency signal processing: Today and future. Digital Signal Processing, 2021. 119: p. 103216.
[13] Ojala, T., M. Pietikainen, and T. Maenpaa, Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. IEEE Transactions on pattern analysis and machine intelligence, 2002. 24(7): p. 971-987.
[14] Ezazi, Y. and P. Ghaderyan, Textural feature of EEG signals as a new biomarker of reward processing in Parkinson’s disease detection. Biocybernetics and Biomedical Engineering, 2022. 42(3): p. 950-962.
[15] Novak, V., et al., Cerebral flow velocities during daily activities depend on blood pressure in patients with chronic ischemic infarctions. Stroke, 2010. 41(1): p. 61-66.
[16] Beyaert, C., R. Vasa, and G.E. Frykberg, Gait post-stroke: Pathophysiology and rehabilitation strategies. Neurophysiologie Clinique/Clinical Neurophysiology, 2015. 45(4-5): p. 335-355.
[17] Feigin, V.L., et al., Global, regional, and national burden of stroke and its risk factors, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. The Lancet Neurology, 2021. 20(10): p. 795-820.
[18] Lakhan, S.E., A. Kirchgessner, and M. Hofer, Inflammatory mechanisms in ischemic stroke: therapeutic approaches. Journal of translational medicine, 2009. 7(1): p. 1-11.
[19] Weimar, C., et al., Complications following acute ischemic stroke. European neurology, 2002. 48(3): p. 133-140.
[20] Langerak, A.J., et al., Externally validated model predicting gait independence after stroke showed fair performance and improved after updating. Journal of clinical epidemiology, 2021. 137: p. 73-82.
[21] Lopez-Meyer, P., G.D. Fulk, and E.S. Sazonov, Automatic detection of temporal gait parameters in poststroke individuals. IEEE Transactions on Information Technology in Biomedicine, 2011. 15(4): p. 594-601.
[22] Yang, S., et al., Estimation of spatio-temporal parameters for post-stroke hemiparetic gait using inertial sensors. Gait & posture, 2013. 37(3): p. 354-358.
[23] Simats, A. and A. Liesz, Systemic inflammation after stroke: implications for post‐stroke comorbidities. EMBO Molecular Medicine, 2022. 14(9): p. e16269.
[24] Srinivas, A. and J.P. Mosiganti, A brain stroke detection model using soft voting based ensemble machine learning classifier. Measurement: Sensors, 2023. 29: p. 100871.
[25] Balaban, B. and F. Tok, Gait disturbances in patients with stroke. Pm&r, 2014. 6(7): p. 635-642.
[26] Gor-García-Fogeda, M.D., et al., Observational gait assessments in people with neurological disorders: a systematic review. Archives of physical medicine and rehabilitation, 2016. 97(1): p. 131-140.
[27] Yavuzer, G., et al., Rehabilitation of stroke patients: clinical profile and functional outcome. American journal of physical medicine & rehabilitation, 2001. 80(4): p. 250-255.
[28] Hendricks, H.T., et al., Motor recovery after stroke: a systematic review of the literature. Archives of physical medicine and rehabilitation, 2002. 83(11): p. 1629-1637.
[29] Firouzi, M., et al., Immediate effects of the honda walking assist on spatiotemporal gait characteristics in individuals after stroke. Medicine in Novel Technology and Devices, 2022. 16: p. 100173.
[30] Duncan, P.W., et al., Body-weight–supported treadmill rehabilitation after stroke. New England Journal of Medicine, 2011. 364(21): p. 2026-2036.
[31] Saljuqi, M. and P. Ghaderyan, A novel method based on matching pursuit decomposition of gait signals for Parkinson’s disease, Amyotrophic lateral sclerosis and Huntington’s disease detection. Neuroscience Letters, 2021. 761: p. 136107.
[32] Burnfield, M., Gait analysis: normal and pathological function. Journal of Sports Science and Medicine, 2010. 9(2): p. 353.
[33] Rogers, A., et al., Repeatability of plantar pressure assessment during barefoot walking in people with stroke. Journal of Foot and Ankle Research, 2020. 13: p. 1-7.
[34] Meyring, S., et al., Dynamic plantar pressure distribution measurements in hemiparetic patients. Clinical Biomechanics, 1997. 12(1): p. 60-65.
[35] Meignen, S., T. Oberlin, and D.-H. Pham, Synchrosqueezing transforms: From low-to high-frequency modulations and perspectives. Comptes Rendus Physique, 2019. 20(5): p. 449-460.
[36] Kanitthika, K. and K.S. Chan. Pressure sensor positions on insole used for walking analysis. in The 18th IEEE International Symposium on Consumer Electronics (ISCE 2014). 2014. IEEE.
[37] Saljuqi, M. and P. Ghaderyan, Combining homomorphic filtering and recurrent neural network in gait signal analysis for neurodegenerative diseases detection. Biocybernetics and Biomedical Engineering, 2023. 43(2): p. 476-493.
[38] Daubechies, I. and S. Maes, A nonlinear squeezing of the continuous wavelet transform based on auditory nerve models, Wavelets Med. 1996, Biol.
[39] Thakur, G. and H.-T. Wu, Synchrosqueezing-based recovery of instantaneous frequency from nonuniform samples. SIAM Journal on Mathematical Analysis, 2011. 43(5): p. 2078-2095.
[40] Oberlin, T., S. Meignen, and V. Perrier. The Fourier-based synchrosqueezing transform. in 2014 IEEE international conference on acoustics, speech and signal processing (ICASSP). 2014. IEEE.
[41] Degirmenci, D., et al. Synchrosqueezing transform in biomedical applications: A mini review. in 2020 Medical Technologies Congress (TIPTEKNO). 2020. IEEE.
[42] Ozdemir, M.A., O.K. Cura, and A. Akan, Epileptic eeg classification by using time-frequency images for deep learning. International journal of neural systems, 2021. 31(08): p. 2150026.
[43] Pietikäinen, M., Local binary patterns. Scholarpedia, 2010. 5(3): p. 9775.
[44] Ojala, T., M. Pietikäinen, and D. Harwood, A comparative study of texture measures with classification based on featured distributions. Pattern recognition, 1996. 29(1): p. 51-59.
[45] Wang, L. and D.-C. He, Texture classification using texture spectrum. Pattern recognition, 1990. 23(8): p. 905-910.
[46] Ganjali, M. and V. Shalchyan, Extracting Spatial Spectral Patterns from EEG Signals for Diagnosis of Mild Cognitive Impairment. TABRIZ JOURNAL OF ELECTRICAL ENGINEERING, 2019. 48(4): p. 1741-1752.
[47] Burges, C.J., A tutorial on support vector machines for pattern recognition. Data mining and knowledge discovery, 1998. 2(2): p. 121-167.
[48] Mirzaeian, R. and P. Ghaderyan, Gray-level co-occurrence matrix of Smooth Pseudo Wigner-Ville distribution for cognitive workload estimation. Biocybernetics and Biomedical Engineering, 2023. 43(1): p. 261-278.
[49] Boubchir, L., et al. Classification of EEG signals for detection of epileptic seizure activities based on LBP descriptor of time-frequency images. in 2015 IEEE International Conference on Image Processing (ICIP). 2015. IEEE.
[50] Gao, S., Gray level co-occurrence matrix and extreme learning machine for Alzheimer's disease diagnosis. International Journal of Cognitive Computing in Engineering, 2021. 2: p. 116-129.
[51] Quinlan, J.R., Program for machine learning. C4. 5, 1993.
[52] Moghaddam, F., P. Ghaderyan, and M. Shamsi, Diagnosis of Attention Deficit/Hyperactivity Disorder using the analysis of different brain regions connectivity and Dynamic Time Warping method. TABRIZ JOURNAL OF ELECTRICAL ENGINEERING, 2023. 53(3): p. 223-233.

فایل ها

هم رسانی

ارجاع به این مقاله

آمار