本研究旨在運用MIMO FMCW雷達技術與機器學習來預測帕金森氏症。透過MIMO FMCW雷達對受試者手指運動軌跡進行偵測，收集相關的生理數據，再利用機器學習進行數據分析，從而預測帕金森氏症。具體來說，MIMO FMCW雷達會發射FMCW訊號，再接收FMCW訊號遇到手指後的反射信號，並運用信號處理技術來獲得手指微小運動軌跡的數據，藉由這些數據，訓練機器學習模型，進行手指運動特徵分析，以有效識別帕金森氏症的相關跡象。
這種非侵入性的檢測方法具有許多優點，能夠提供及時且具有隱私性的診斷。本研究的機器學習模型實驗結果顯示，在預測帕金森氏症方面具有高度的準確性和精確性。這說明了利用FMCW雷達技術結合機器學習進行帕金森氏症預測具有相當的潛力。

This study aims to predict Parkinson's disease using MIMO FMCW radar technology and machine learning. Through MIMO FMCW radar, the trajectory of participants' finger movements is detected, and relevant physiological data is collected. Subsequently, machine learning is employed for data analysis to predict Parkinson's disease. Specifically, the MIMO FMCW radar emits FMCW signals, receives reflected signals upon encountering participants' fingers, and utilizes signal processing techniques to acquire data on subtle finger movements. These data are then used to train a machine learning model for analyzing finger movement characteristics, effectively identifying indicators of Parkinson's disease.
This non-invasive detection method possesses numerous advantages, offering timely and privacy-preserving diagnoses. The experimental results of the machine learning model in this study demonstrate a high level of accuracy and precision in predicting Parkinson's disease. This indicates the considerable potential of utilizing FMCW radar technology combined with machine learning for Parkinson's disease prediction.

[1] 國家發展委員會(2022, Aug)。高齡化時程。檢自: https://www.ndc.gov.tw/Content_List.aspx?n=D527207EEEF59B9B.
[2] 晟德。認識巴金森氏病。檢自: https://www.centerlab.com.tw/business/guardian_detail/2.
[3] H. Ling, L. A. Massey, A. J. Lees, P. Brown, and B. L. Day, "Hypokinesia without decrement distinguishes progressive supranuclear palsy from Parkinson's disease," Brain, vol. 135, no. Pt 4, pp. 1141-1153, Apr 2012, doi: 10.1093/brain/aws038.
[4] M. Djuric-Jovicic, N. S. Jovicic, A. Roby-Brami, M. B. Popovic, V. S. Kostic, and A. R. Djordjevic, "Quantification of Finger-Tapping Angle Based on Wearable Sensors," Sensors (Basel), vol. 17, no. 2, Jan 25 2017, doi: 10.3390/s17020203.
[5] K. Shima, T. Tsuji, A. Kandori, M. Yokoe, and S. Sakoda, "Measurement and Evaluation of Finger Tapping Movements Using Log-linearized Gaussian Mixture Networks," Sensors (Basel), vol. 9, no. 3, pp. 2187-2201, 2009, doi: 10.3390/s90302187.
[6] microsoft. (2023, Oct 24). anomaly detection. from https://learn.microsoft.com/zh-tw/azure/ai-services/anomaly-detector/overview.
[7] D. M. Pozar, Microwave engineering, 4th Edition. Hoboken, NJ: Wiley, 2012.
[8] C. Iovescu and S. Rao. (May 2017). The fundamentals of millimeter wave radar sensors. from https://www.ti.com/lit/wp/spyy005a/spyy005a.pdf?ts=1701165396951.
[9] T. Huang. (2022, Apr 24). FMCW Radar. from https://chih-sheng-huang821.medium.com/%E8%AA%BF%E9%A0%BB%E9%80%A3%E7%BA%8C%E6%B3%A2%E9%9B%B7%E9%81%94-frequency-modulated-continuous-wave-fmcw-radars-%E6%B8%AC%E8%B7%9D-%E6%B8%AC%E9%80%9F%E5%8E%9F%E7%90%86-2-3-981e4706fc9d.
[10] Texas Instruments. (2017, May). MIMO Radar. from https://www.ti.com/lit/an/swra554a/swra554a.pdf.
[11] Texas Instruments. IWR6843ISK. from https://www.ti.com.cn/tool/cn/IWR6843ISK.
[12] Texas Instruments. (2018, Oct). 60GHz mmWave Sensor EVMs. from https://www.ti.com/lit/ug/swru546e/swru546e.pdf?ts=1698316161826&ref_url=https%253A%252F%252Fwww.google.com%252F.
[13] E. Alpaydin, Introduction to machine learning, 3rd Edition. Cambridge, Mass: MIT Press, 2014.
[14] T. Ergen and S. S. Kozat, "Unsupervised Anomaly Detection With LSTM Neural Networks," IEEE Trans Neural Netw Learn Syst, vol. 31, no. 8, pp. 3127-3141, Aug 2020, doi: 10.1109/TNNLS.2019.2935975.
[15] 柯頌竹(2021, Jan 22)。簡介LSTM與GRU。檢自: https://medium.com/programming-with-data/25-%E7%B0%A1%E4%BB%8Blstm-%E8%88%87gru-3e0eaa100d29.
[16] S. Timmermann et al., "Automated matching of two-time X-ray photon correlation maps from phase-separating proteins with Cahn-Hilliard-type simulations using auto-encoder networks," J Appl Crystallogr, vol. 55, no. Pt 4, pp. 751-757, Aug 1 2022, doi: 10.1107/S1600576722004435.
[17] J. S. Do, A. B. Kareem, and J. W. Hur, "LSTM-Autoencoder for Vibration Anomaly Detection in Vertical Carousel Storage and Retrieval System (VCSRS)," Sensors (Basel), vol. 23, no. 2, Jan 15 2023, doi: 10.3390/s23021009.
[18] 江銘文(2022)。基於毫米波FMCW雷達之非接觸式手指輕敲實驗之研究。 碩士論文，國立臺灣科技大學電子工程研究所，臺北市。
[19] L. Alzubaidi et al., "Review of deep learning: concepts, CNN architectures, challenges, applications, future directions," J Big Data, vol. 8, no. 1, p. 53, 2021, doi: 10.1186/s40537-021-00444-8.