具生物反饋機制之智慧型長時睡眠監測系統---子計畫一：應用於智慧型長時睡眠監測系統之感測器與類比前端積體電路設計(I)

Abstract

市面上生理訊號量測系統不但昂貴而且體積龐大，對於睡眠障礙患者或嬰兒等需要使用長時睡眠監控系統的人來說，過多的纜線與過大的機器也造成使用上的不便，妨礙睡眠。解析睡眠當中進行的多重生理監測，包含腦電圖、心電圖、肌電圖與眼電圖，其前端訊號均為類比的電訊號，大小差距僅有五十倍，頻寬分佈也僅相差不到五倍，皆為非侵入體表量測方式，若將這些生理訊號用同一組放大電路讀出，並搭配數位與類比的SOC整合晶片，將可大幅縮小監測系統的體積。 本計畫並採用新型乾式電極作為前端感測器，並將放大器電路與電極整合變成主動式電極，整個系統的放大器將與電極一起分散在不同的量測點，經由導線將訊號傳輸至處理器端，透過類比多功器與類比數位轉換器將不同的生理訊號轉換為數位訊號，進行後續資料探勘與儲存動作。此設計不但能大幅提高訊號雜訊比，並且能自由調配不同生理訊號的通道數，成為通用生理訊號放大器。本計畫並研發新型之光聲感測器作為血氧濃度偵測器，藉由發射紅外線脈衝，能使血紅素吸收光波膨脹並收縮，放出超音波，而藉此超音波訊息能判斷血氧濃度，瞭解睡眠狀況與血氧濃度的關係。 而本計畫中的類比數位轉換器，與其他子計畫的數位電路透過SOC整合後，能達到微型化可攜式智慧型長時睡眠監控系統，能協助臨床上更精確的睡眠障礙診斷，並能有效預防嬰兒猝死的不幸事件。

The commercial products to detect physical signals are usually bulky and expansive. The bulky case and too many cables are inconvenient to be used on infants, and are especially disturbing for patients with sleeping disorder. However, in polysomnographic monitoring, which is routine for sleeping disorder, the signals extracted by EEG, ECG, EMG, and EOG are all analog electrical signals. The amplitude differences are within 50 times, and bandwidth differences are within 10 times. Besides, all of them are noninvasive systems. So it is possible to detect all signals with the same amplifier circuit accompanying with a digital and analog SoC chip. Thus the polysomnographic monitoring can be substantially shrunk. In this subproject, the novel dry electrode is adopted as the front-end sensor. The amplifier will be integrated with the sensor to form the active electrode. Thus the amplifiers of the system will be distributed together with electrodes in individual locations, and then interconnect with the SoC chip via wires. Through analog MUX and ADC, various kinds of physical signals will be converted to digital signals for furthering signal processing, data mining, and storing. This design can enhance SNR and give flexibility to channel format and number to achieve a universal biosignal amplifier. On the other hand, the photoacoustic image (PAI) is also developed in this subproject to detect the oxygen level of patients. The hemoglobin flashed with a laser pulse will go through the procedures of optical absorption, thermoelastic expansion, and ultrasonic wave generation. Thus hemoglobin concentration and oxygen saturation can be detected with PAI. The oxygen level is also an important index in sleeping disorder. The ADC developed by this subproject can be integrated with digital circuits developed by other subprojects on a SoC chip to form a miniaturized and portable intelligent long-time sleep monitoring system. The system can be use to precisely diagnose sleeping disorder, and it can also be used to efficiently avoid the infant tragedy of sudden death.