本篇論文針對癲癇抑制之功能性電刺激系統電路，研究其發展趨勢與進行晶片功能設計。考量至電刺激系統電路從晶片設計到實際應用，因此可分三個部份進行探討: 電刺激系統晶片設計、系統整合平台及動物實驗過程。

首先為電刺激系統晶片設計方面，提出一多通道雙波型電流模式且具備推挽式電荷補償機制之電刺激器(Functional Biphasic Electrical Current Stimulatior Unit, FBECSU)。刺激電流之設計採用二進制碼至熱碼數位類比電流轉換器(Binary-to-Thermometer DAC)，刺激輸出電流從10uA至2.56mA，故應用範圍從動物實驗涵蓋至人體使用。為了避免進行單一流向電刺激，電極與組織進行化學反應，並溶解出有毒物質傷害組織，故針對電荷平衡採用雙向電流刺激模式。由於電路晶片經由化學製程導致實際電路元件不匹配，長期電刺激情況下，仍將殘存過多電荷於組織內。故此提出一新創之推挽式放大器進行電荷補償機制，利用高效率、低功耗特性以消除殘存電荷，控制於安全範圍內。刺激波型選擇則根據文獻於電刺激波型特性分析，長期電刺激以衰減指數式波型之能量及電荷效率最佳；而短期電刺激以脈波波型之能量及功耗效率最佳。電路佈局以模組化形式增添通道之方便性，未來若因應癲癇實驗考量，則可隨時擴增通道數量。
為了應用電刺激系統晶片於癲癇病患者腦內，設計上將考量到功率消耗及組織阻值增生問題而提出一阻抗偵測機制，不僅能因應當前設定電刺激電流大小調控晶片供給電壓，進而達到最佳效率電刺激；更能用於長期電刺激之組織阻值增生，而導致晶片內部受到跨壓限制時，可藉由阻抗偵測機制進行提升供給電壓及參考電壓。

其次，系統整合平台可分為第一版量測系統及第二版動物實驗平台。第一版量測系統能率先驗證電刺激系統晶片功能設計，以FPGA作為數位訊號產生來源，並結合商用DAC、Level Shifter 晶片於供給電壓及數位訊號強度轉換。第二版本量測系統以驗證抑制癲癇動物實驗之電刺激系統晶片有效性為目的之動物實驗平台，能以藍芽無線傳輸結合手機GUI介面輸入電刺激參數，透過微處理器MCU產生SPI協定之數位控制訊號後，傳遞至動物實驗平台之電刺激系統晶片設定參數與進行功能性電刺激，另外其好處為攜帶方便、可避免複雜設備架設。

最後，動物實驗以多通道電刺激系統晶片之動物實驗平台進行指數衰減式功能性電刺激，由實驗波型及行為結果可得知於癲癇發作期間，腦波EEG訊號振幅起伏劇烈並出現身體抽搐等行為。然而經由長時間癲癇抑制電刺激後，能大幅穩定癲癇發作之症狀。
後期雖於癲癇發作期間腦波EEG訊號之振幅仍比未發作期間稍大些，實驗動物之行為動作卻已恢復正常。由此可見，能證實此多通道電刺激系統晶片進行之指數衰減式功能性電刺激對於癲癇抑制之有效性。

This thesis concentrates on the functional electrical stimulation for epilepsy suppression to research the development and design functions of chips. Considering stimulator system circuits of the chip designing to practical application, there's three parts for discussion as following: electrical stimulation system chip design, integrated system platform, and animal experiment process.

First of all, a multi-channel Functional Biphasic Electrical Current-Mode Stimulator(FBECSU), with push-pull charge compensation mechanism is proposed in the electrical system chip design. The stimulation current design facilitates methodology of Binary-to-Thermometer Digital-to-Analog(DAC) to output stimulation current from 10$\mu$A to 2.56mA, which is capable to apply from animal rats to human. In order to prevent single direction electrical stimulation that chemical reaction dissolves toxic chemical substance between electrodes and tissue, the stimulator system chips were designed with stimulation current with biphasic waveform for charge balance. Owing to the chemical manufacturing process leading to mismatch existence in realistic circuit components, there's might accumulate excessive residual charges in the tissue with long-term stimulation. Therefore, a innovated push-pull amplifier for charge compensation mechanism, which features of high-slew rate, high-slew efficiency and low power consumption, is proposed to eliminate residual charges below the limit of safety window.
Stimulation waveform selection according to literature of electrical stimulation waveform characteristic analysis mentioned the better energy efficiency and charge efficiency of decaying-exponential waveform in long-term electrical stimulation, and the better energy efficiency and power efficiency is pulsed-shape waveform in short-term electrical stimulation.
Modularized circuits layout takes advantages of conveniency for increasing numbers of stimulated channel. Adding the numbers of stimulated channel could save times considering the demands of epilepsy experiment.
The electrical stimulation system chip could be implanted in the brain of epileptic patients, the major issue are the power consumption and tissue impedance hyperplasia problems. Therefore, a impedance sensing mechanism is proposed to control the electrical stimulation current intensity according to recent chip setting for supply voltage so that electrical stimulation achieves the best efficiency. The tissue impedance hyperplasia in long-term electrical stimulation causes limited across voltage of internal circuit existence which reduce output current intensity, the supply and reference voltage enhancement could rely on the proposed impedance sensing mechanism for adjustment.

Secondly, the integrated platform divided into first version measured system and second version animal experiment platform, The first version measured system is the first step to verify the functions of stimulation chip design. The measured system is based on FPGA for digital signal generation and commercial DAC, Level Shifter chip for supply voltage and digital signal domain transition.
The second version animal experiment platform targets effectiveness of the stimulator system chip in animal experiment of epilepsy suppression. Both of stimulation parameter setting of mobile GUI with bluetooth transmission, and digital control signals generation according to SPI protocol through Microprocessor to stimulation system chip of animal experiment platform for stimulation parameter setting benefit from conveniency for carrying and avoiding the complicated setup of equipment.

Lastly, the animal experiment introduced decaying-exponential waveform into the platform of multi-channel electrical stimulation system chip. As a result of the experimental results of waveform and behavior in the ictal period of epilepsy seizure, the amplitude of EEG signal of brain undulates extremely and behaves body trembling symptoms. Nevertheless, long-term electrical stimulation for epilepsy suppression stabilizes the symptoms substantially. Although the ictal period EEG signal amplitude is a little violent than pre-ictal period, these experimental animal behavior recover to normal status. Thus, the multi-channel stimulation system chip with decaying-exponential shape proves the effectiveness of epilepsy suppression.

[1] H.M. Lee, H. Park, and M. Ghovanloo, “A powerefficient wireless system with adaptive supply control for deep brain stimulation,” IEEE Journal of SolidState Circuits, vol. 48, no. 9, pp. 2203–2216, 2013.
[2] J. Niparko, Cochlear implants, principles and practices. Philadelphia: Lippincott Williams and Wilkins, 2009.
[3] G. Dagnelie, “Retinal implants –emergence of a multidisciplinary field,” Current Opinion In Neurology, vol. 25, no. 1, pp. 67–75, 2012.
[4] H.W. Chiu, M.L. Lin, C.W. Lin, I.H. Ho, W.T. Lin, P.H. Fang, Y.C. Li, Y.R. Wen, and S.S. Lu, “Pain control on demand based on pulsed radiofrequency stimulation of the dorsal root ganglion using a batteryless implantable CMOS SoC,” IEEE Transactions on Biomedical Circuits and Systems, vol. 4, no. 6, pp. 350–359, 2010.
[5] M. Capogrosso, T. Milekovic, D. Borton, F. Wagner, E. Moraud, J. Mignardot, N. Buse, J. Gandar, Q. Barraud, D. Xing, E. Rey, S. Duis, J. Yang, W. Ko, Q. Li, P. Detemple, T. Denison, S. Micera, E. Bezard, J. Bloch, and G. Courtine, “A brain–spine interface alleviating gait deficits after spinal cord injury in primates,” Nature, vol. 539, pp. 284–288, 2016.
[6] V.S. Polikova, P.A. Trescob, and W.M. Reicherta, “Response of brain tissue to chronically implanted neural electrodes,” Journal of Neuroscience Methods, vol. 148, pp. 1–18, 2005. 103
[7] C.Y. Lin, W.L. Chen, and M.D. Ker, “Implantable stimulator for epileptic seizure suppression with loading impedance adaptability,” IEEE Transactions on Biomedical Circuits and Systems, vol. 7, no. 2, pp. 196–203, 2013.
[8] E. Noorsal, K. Sooksood, H. Xu, R. Hornig, J. Becker, and M. Ortmanns, “A neural stimulator frontend with highvoltage compliance and programmable pulse shape for epiretinal implants,” IEEE Journal of SolidState Circuits, vol. 47, no. 1, pp. 244– 256, 2012.
[9] S. Arfin and R. Sarpeshkar, “An energyefficient, adiabatic electrode stimulator with inductive energy recycling and feedback current regulation,” IEEE Transactions on Biomedical Circuits and Systems, vol. 6, no. 1, pp. 1 14, 2012.
[10] M.N. van Dongen and W.A. Serdijn, “A powerefficient multichannel neural stimulator using highfrequency pulsed excitation from an unfiltered dynamic supply,” IEEE Transactions on Biomedical Circuits and Systems, vol. 10, no. 1, pp. 61–71, 2016.
[11] Y.K. Lo, K. Chen, P. Gad, and W. Liu, “A fully integrated high compliance
Voltage SoC for epiretinal and neural prostheses,” IEEE Transactions on Biomedical Circuits and Systems, vol. 7, no. 6, pp. 761–772, 2013.
[12] Z. Luo and M.D. Ker, “A highvoltagetolerant and precise chargebalanced neurostimulator in low voltage CMOS process,” IEEE Transactions on Biomedical Circuits and Systems, vol. 10, no. 6, pp. 1087–1099, 2016.
[13] C. Liu, P. H. Chen, M. D. Ker, and C. Y. W. et al., “An 82.9TripleOutput Battery Management Unit for Implantable Neuron Stimulator in 180nm Standard CMOS,” 104 IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing, vol. 66, pp. 788–792, May 2019.
[14] D. Rozgi´c, V. Hokhikyan, W. Jiang, I. Akita, M. S. BasirKazeruni, H. Chandrakumar, and D.Markovi´c, “A 0.338, artifact free, 64 contact neuromodulation platform for simultaneous stimulation and sensing,” IEEE Transactions on Biomedical Circuits and Systems, vol. 13, pp. 38–55, February 2019.
[15] X. Liu, A. Demosthenous, and N. Donaldson, “An integrated implantable stimulator that is failsafe without offchip blocking capacitors,”
IEEE Transactions on Biomedical Circuits and Systems, vol. 2, pp. 231–244, January 2008.
[16] S. Ha, C. Kim, J. Park, G. Cauwenberghs, and P. P. Mercier, “A fully integrated RF powered energy-replenishing current-controlled stimulator,” IEEE Transactions on Biomedical Circuits and Systems, vol. 13, pp. 191–202, Febuary 2019.
[17] A. Demosthenous, “Advances in microelectronics for implantable medical devices,” Advances in Electronics, vol. 2014, no. 981295, 2014.
[18] K. Sooksood, T. Stieglitz, and M. Ortmanns, “An active approach for charge balancing in functional electrical stimulation,” IEEE Transactions on Biomedical Circuits and Systems, vol. 4, no. 3, pp. 162–170, 2010.
[19] N. Butz, A. Taschwer, S. Nessler, Y. Manoli, and M. Kuhl, “A 22 V compliant 56μW twin-track active charge balancing enabling 100even in monophasic and 36stimulators,” IEEE Journal of Solid-State Circuits, vol. 53, pp. 2298–2310, August 2018.
[20] E. Greenwald, C. Maier, Q. Wang, R. Beaulieu, R. E. Cummings, G. Cauwenberghs, and N. Thakor, “A CMOS current steering neurostimulation array with integrated 105 DAC calibration and charge balancing,” IEEE Transactions on Biomedical Circuits and Systems, vol. 11, pp. 324–335, April 2017.
[21] Y. J. Jeon, L. Yao, Y. Gao, and M. A. Arasu, “A 0.034neural stimulation frontend (SFE) IC with onchip voltage compliance monitoring circuit and analysis on resting potential by utilizing the SFE IC,” IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications, vol. 66, pp. 3797–3810, Octorber 2019.
[22] M. H. Maghami, A. Sodagar, and M. Sawan, “Analysis and design of a high-compliance ultrahigh output resistance current mirror employing positive shunt feedback,” International Journal of Circuit Theory and Applications, vol. 43, 2014.
[23] H. M. Lee, K. Y. Kwon, W. Li, and M. Ghovanloo, “A powerefficient switched-capacitor stimulating system for electrical/optical deep brain stimulation,” IEEE
Journal of Solid-State Circuits, vol. 50, pp. 360–374, January 2015.
[24] J. Charthad, T. C. Chang, Z. Liu, A. Sawaby, M. J. Weber, S. Baker, F. Gore, S. A. Felt, and A. Arbabian, “A mmsized wireless implantable device for electrical stimulation of peripheral nerves,” IEEE Transactions on Biomedical Circuits and Systems, vol. 12, p. 207–222, April 2018.
[25] A. Urso, V. Giagka, M. van Dongen, and W. A. Serdijn, “An Ultra High-Frequency 8-Channel Neurostimulator Circuit With 68Efficiency,” IEEE Transactions on Biomedical Circuits and Systems, vol. 13, pp. 882–892, Octorber 2019.
[26] P.L. Yi, S.B. Jou, Y.J. Wu, and F.C. Chang, “Manipulation of Epileptiform Electrocorticograms (ECoGs) and Sleep in Rats and Mice by Acupuncture,” Journal of Visualized Experiments: JoVE, December 2016.
[27] H.T. Tseng, Y.T. Hsiao, P.L. Yi, and F.C. Chang, “Deep brain stimulation increases seizure threshold by altering rem sleep and delta powers during nrem sleep,” Frontiers in Neurology, vol. 11, p. 752, 2020.