Document22

IEEE REVIEWS IN BIOMEDICAL ENGINEERING, VOL. 3, 2010 93

Analog Integrated Circuits Design for ProcessingPhysiological Signals

Yan Li, Carmen C. Y. Poon, Member, IEEE, and Yuan-Ting Zhang, Fellow, IEEE

Methodological Review

Abstract—Analog integrated circuits (ICs) designed for pro-cessing physiological signals are important building blocks ofwearable and implantable medical devices used for health mon-itoring or restoring lost body functions. Due to the nature ofphysiological signals and the corresponding application scenarios,the ICs designed for these applications should have low powerconsumption, low cutoff frequency, and low input-referred noise.In this paper, techniques for designing the analog front-endcircuits with these three characteristics will be reviewed, in-cluding subthreshold circuits, bulk-driven MOSFETs, floatinggate MOSFETs, and log-domain circuits to reduce power con-sumption; methods for designing fully integrated low cutofffrequency circuits; as well as chopper stabilization (CHS) andother techniques that can be used to achieve a high signal-to-noiseperformance. Novel applications using these techniques will alsobe discussed.

Index Terms—Analog integrated circuits, low frequency, lownoise, low power, medical devices.

I. INTRODUCTION

G LOBAL population ageing and prevalence of chronicdiseases have placed substantial pressure on our current

healthcare systems [1]–[3]. Meanwhile, there is a pressing callfor a more proactive healthcare approach, where individual’shealth conditions will be monitored closely from birth for theprevention, prediction, early detection, and timely treatmentof diseases [4]–[7]. Long-term and continuous monitoringof health conditions are made possible by wearable and im-plantable devices, which must be small and unobtrusive enoughso that users of these devices can maintain their normal lifestylewithout interruption. These devices require enabling tech-nologies in six areas: miniaturization, integration, networking,digitalization, smart and standardization (MINDSS) [8], [9].

Manuscript received February 24, 2010; revised August 11, 2010; acceptedSeptember 14, 2010. Date of publication September 30, 2010; date of currentversion December 08, 2010. This work was supported in part by the Hong KongInnovation and Technology Fund (ITF), the National Basic Research Programof China (“973” Program) under Grant 2010CB732606, and the GuangdongInnovation Team Fund in China.

Y. Li and Y.-T. Zhang are with the Joint Research Centre for Biomedical En-gineering, The Chinese University of Hong Kong, Hong Kong, China. They arealso with Key Laboratory for Biomedical Informatics and Health Engineering,Chinese Academy of Science and also with SIAT-Institute of Biomedical andHealth Engineering, Chinese Academy of Science (e-mail: [email protected]).

C. C. Y. Poon is with the Joint Research Centre for Biomedical Engineering,The Chinese University of Hong Kong, Hong Kong, China.

Digital Object Identifier 10.1109/RBME.2010.2082521

Fig. 1. Frequency ranges of some physiological signals, where PCG, PPG,EGG, ECG, ERG, EOG, EEG, and EMG refer to the phonocardiographic, pho-toplethysmographic, electrogastrographic, electrocardiographic, electroretino-graphic, el ectrooculographic, electroencephalographic and electromyographicsignals respectively.

The integration in the above description refers to the design andimplementation of integrated circuits (ICs) for these wearableand implantable devices to perform various functions, includingprocessing physiological signals.

Physiological signals, for example bio-potentials such aselectrocardiogram (ECG), electroencephalogram (EEG), andelectromyogram (EMG), chemical quantities such as ion con-centrations and physical quantities such as body temperature,blood pressure, are often small signals of low frequency [10],as shown in Fig. 1. Therefore, after converting a physiologicalsignal into an electrical signal by the corresponding type ofsensor or transducer, an analog front-end circuit is often neededto filter and amplify the signal before digitizing it for furtherprocessing.

The design of analog ICs for wearable and implantable de-vices faces three challenges. Firstly, the design must incorpo-rate low power techniques to reduce the amount of heat dissipa-tion such that the surrounding human tissues will not be dam-aged. A heat flux of 80 mW/cm can already cause necrosisin muscle tissue [11], [12]. Low power IC design is also im-portant to be used along with new battery technologies to avoidthe use of bulky batteries or frequent replacement of batteriesduring long-term operations. At present, zinc-air batteries arecommonly used for button cells to power wearable devices suchas hearing-aids because they have high energy densities and are

1937-3333/$26.00 © 2010 IEEE

94 IEEE REVIEWS IN BIOMEDICAL ENGINEERING, VOL. 3, 2010

Fig. 2. � � � curves of NMOSFET under different aspect ratios.

relatively inexpensive to produce [13]. Nevertheless, zinc-airbatteries have a low current capability and are unsuitable forsome long-term monitoring applications. Rechargeable lithiumion batteries are widely used in mobile phones and laptop com-puters due to their high energy density and light weight com-pared to other rechargeable batteries [13], [14]. New batteryor power technologies, e.g., lithium-air battery technologies orconverting electrical power from ambient sources, are underdevelopment to yield more powerful and lightweight batteries[15].

Secondly, physiological signals are often low frequency sig-nals that span from dc to a few kilohertz [10]. Sometimes, thereis also a large dc component caused by electrode and skin in-terface residing in the signal. Hence, the analog front-end ICsare often implemented with low cutoff frequencies to read outthe signals from the electrodes or sensors. Such designs requirelarge resistances and/or large capacitances, which can be easilyachieved with discrete components but are difficult to be fabri-cated on chip directly due to the large areas they occupy. For ex-ample, an integrated 100 pF capacitance already occupies about0.1 mm .

Thirdly, the amplitude of most physiological signals can berelatively small and in the range of a few microvolts to tens ofmillivolts. The quality of the signals is also largely affected bythe noises from the electrodes or sensors, the power supplies, aswell as the user’s motion. Therefore, the circuits must exhibitlow input-referred noises to process the weak physiological sig-nals precisely.

The paper is organized as follows: In Sections II, III, andIV we will briefly review respectively the low power, low fre-quency, and low noise integrated circuit design techniques forprocessing physiological signals. Section V gives some exam-ples of the related applications.

II. LOW POWER DESIGN METHODS

Low power consumption can be achieved by reducing the op-eration current, lowering the supply voltage, or compressing thesignal from current domain to voltage domain using the expo-nential current versus voltage relation of transistors.

A. Subthreshold Circuits

Ideally, a MOSFET, for example an N type, should turn on asthe gate to source voltage exceeds the threshold voltage

, and turn off as drops below ; however, in re-ality, there exists a small current flowing from drain to sourcewhen is below . This operation region is called sub-threshold or weak inversion region [16], [17], which was firstproposed in the 1970s [18]–[20] and further studied by manygroups subsequently [21]–[25].

In recent years, some designers make use of this character-istic to develop low power circuits. The idea is to bias MOSFETin the subthreshold region, restraining the current to a muchsmaller value than in regions above the threshold voltage andthereby largely reducing the power consumption. In this paper,we will use weak inversion (i.e., subthreshold), moderate inver-sion, and strong inversion regions to describe a MOSFET’s threeoperation states, which are differentiated by the inversion coef-ficient, . These three operation regions are differently definedin different models. For example, in the EKV model [26], theweak inversion/subthreshold, moderate inversion, and strong in-version region usually corresponds to , ,and , respectively. While in Cunha’s model, the weakinversion/subthreshold, moderate inversion, and strong inver-sion region corresponds to , , and

, respectively [27]. Fig. 2 shows the curves ofa NMOSFET when with different aspect ratiosin the UMC 0.13- m CMOS process. As shown in the rightpanel of Fig. 2, the weak inversion current, which is the cur-rent when (in the UMC 0.13 m CMOS process,

), can be as low as 1 nA with a small aspect ratio.In addition, since the transconductance is greatest in the weakinversion region compared to that in the moderate and stronginversion regions for a given drain current, biasing a MOSFETin the weak inversion region also helps to decrease the input-re-ferred noise and increase the gain, which are desirable featuresfor processing the small amplitude physiological signals.

On the other hand, a MOSFET in the weak inversion regionis known to have a poor frequency response. The transitionalfrequency of a MOSFET in Cunha’s model is given by [27]

(1)

where is the mobility of electrons or holes , isthe thermal voltage, and is the inversion coefficient of the

LI et al.: ANALOG INTEGRATED CIRCUITS DESIGN FOR PROCESSING PHYSIOLOGICAL SIGNALS 95

Fig. 3. (a) Bulk driven MOSFET and its equivalent device—JFET [38]. (b) Schematic symbol and equivalent circuit of multiple-input floating gate MOSFET [28].

MOSFET. According to (1), a small will result in a low transi-tional frequency. Usually, should be at least three to ten timeshigher than the operating frequency of the circuit [28]. As a re-sult, small limits the bandwidths of the circuits. For example,assuming and using the UMC 0.13- m CMOS processwith cm Vs and m, the transitional fre-quency is calculated to be Hz according to (1). Supposethe transitional frequency is three times higher than the oper-ating frequency, the operating frequency has to be limited within4 10 Hz. This may be a problem when designing ICs forother applications but not for processing physiological signals,which are normally within 10 Hz, as shown in Fig. 1. More-over, the source and drain substrate current associated with thereversed moat-substrate junction should not be ignored when theweak inversion current is reduced to a certain level [28]. How-ever, a tradeoff can be made between the bias current and theacceptable leakage current. For the above reasons, subthresholdcircuits are the most commonly used in low power designs forprocessing physiological signals, e.g., a nanopower band-passfilter for detecting acoustic signal [29], a 220 nW neural ampli-fier for a multi-channel neural recording system [30], and a 140nW modulator for processing EEG [31], all of which operatedat nanowatt by using the subthreshold method.

B. Low Threshold Voltage Methods

Lowering supply voltage is another way to reduce powerconsumption. Nevertheless, as supply voltage is scaled downwith feature size, threshold voltage cannot be scaleddown at the same rate since this will increase the off-state orstatic leakage of digital circuits [32], [33]. Analog IC designerswill face big challenges due to the limited voltage headroomunder low supply voltages if the threshold voltages cannotbe scaled down with supply voltages at the same rate. Sometechniques have been used to achieve low equivalent thresholdvoltages corresponding to low supply voltages.

1) Bulk-Driven MOSFETs: Guzinski et al. were the first touse a bulk-driven MOS transistor as an active component in anoperational transconductance amplifier (OTA) differential inputstage to yield a small transconductance and to improve the lin-earity of the circuit [34]. In recent years, their idea has been de-veloped into a technique for lowering threshold voltage in lowpower circuit designs [35]–[37].

In a conventional MOSFET, drain current is controlledby gate to source voltage and the influence of bulk tosource voltage is only considered as a parasitic effect.

Different from the conventional setting, as shown in Fig. 3(a), ina bulk-driven MOSFET, is controlled by with a constant

. A bulk-driven MOSFET which functions as a JFET liketransistor can work with negative, zero, or slightly positive biasvoltages. Designing low power circuits with bulk-driven MOS-FETs should consider the following [38]: 1) bulk-driven MOS-FETs are process dependent and therefore only PMOSFET isavailable in N-well process, and only NMOSFET is available inP-well process; 2) should be less than the turn-on voltageof the bulk-channel PN junction, otherwise the parasitic bipolarjunction transistor (BJT) latch-up may be incurred whenis increasing; 3) bulk-driven MOSFETs have poor frequencycharacteristic and small transconductance compared with thegate driven transistors; and 4) bulk-driven MOSFETs have asmaller transconductance and therefore a larger equivalent inputinferred noise than a normal gate driven MOSFET.

Bulk-driven MOSFETs are useful for designing low-voltageand low power circuits with a biomedical application. Lasanenet al. [39] implemented a 1 V, 0.5 W operational amplifier forbiomedical instrumentations using P-type bulk-driven inputdifferential pair in a 0.35- m N-well CMOS process withthreshold voltages being 0.5 V and 0.65 V for NMOSFETsand PMOSFETs, respectively. The design significantly re-duced the threshold voltage and increased the input commonmode range of the amplifier. To avoid the source to bulk andsource to substrate leakage currents and the parasitic BJTlatch-up, the aspect ratio of the bulk-driven MOSFET was de-signed to be sufficiently large to limit the maximum . Panet al. [40] also proposed a novel OTA with dual bulk-driveninput stage in a 0.35- m CMOS process with a 0.9 V supplyand 9.9 W power consumption. The new scheme achieveda rail-to-rail input range and avoided the leakage current ofconventional bulk-driven circuits. The bulk-driven MOSFET isa potential structure in low power and low voltage biomedicalapplications.

2) Floating Gate MOSFETs: Since floating gate MOS(FGMOS) structure was proposed as a nonvolatile memorydevice by Kahng and Sze in 1967 [41], it has been widelyused in digital EPROM and EEPROM. The floating gate issurrounded by SiO without any electrical connection, which iscapacitively coupled to the controlling gate. The input signal isapplied to the controlling gate. The effective threshold voltagecan be reduced to a small value by setting the input voltageproperly. Thus, it has been found in low voltage and low powerrealm recently for the tunable and reduced threshold voltage.


The most commonly used structure is multiple-input floatinggate (MIFG) MOS transistor, as shown in Fig. 3(b). The voltageof the floating gate is given by [42]

(2)

where is the number of the inputs, and are the thinput capacitance and voltage, is the total capacitance seenby the floating gate, , , are the parasitic capaci-tances between floating gate and drain, source and bulk, respec-tively, and is the residential charge trapped in the floatinggate. When (bulk and source terminals aregrounded), , and , would be deter-mined only by the controlling gate [42]. Assume as thethreshold voltage according to the floating gate, which is equalto the threshold voltage of a normal MOSFET, the MOSFETwill be turned on as . Considering ,we have [42]

(3)

where is the voltage of the input terminal,are the voltages of the controlling terminals. The transistor willturn on even with a small if appropriate , arechosen, that is the threshold according to controlling gate islargely reduced.

Using MIFG MOSFETs, Villegas et al. [42] presented a tran-simpedance amplifier that can be used to diagnose diseases bymonitoring a certain type of chemicals. The amplifier consumed82.5, 9.825, and 47.325 W for currents varying from (1 pA,0.25 nA), (0.25 nA, 62.5 nA), and (62.5 nA, 1 A), respec-tively. Mourabit et al. [43] proposed a sub-1.5 V, 2 W OTA-Cfilter based on subthreshold MIFG MOSFETs. The cutoff fre-quency was tunable from 0.5 to 200 Hz, which is the desirablerange for most physiological signals. There are also some otherlow power designs using MIFG MOSFETs with a relativelylow cutoff frequency that can be used for medical applications[44], [45]. It should be emphasized that the trapped charges infloating gates during fabrication will produce large dc offset,which can be solved by methods such as UV cleaning [46], tun-neling, and hot electron injection [47] and layout design tech-niques [48]. Ramirez-Angulo et al. also proposed an approachbased on quasi-floating gate transistors to remove the trappedcharges at a low supply voltage [49].

C. Log Companding Technique

Different from the techniques mentioned above, log com-panding aims to reduce power at the circuit level based on thecurrent-voltage characteristics of semiconductor devices. The

Fig. 4. Main principle of log companding method [50].

principle of the technique is shown in Fig. 4, where standsfor the companding function that usually follows an exponen-tial law. The processing chain includes a y-x (usually I-V) com-pression and an x-y (usually V-I) expansion, which is there-fore named companding [50]. The basic concept of the theoryis to choose the I-domain input and output signals within givendynamic ranges, but to process the signals internally using anequivalent V-domain within compressed dynamic ranges. Thereduction of the internal voltage dynamic range makes it verysuitable for low voltage and low power applications.

Log-domain circuits were originally realized with bipolartransistors for their exponential current versus voltage charac-teristics. However, with the development of CMOS process,realizing log-domain circuits with MOS transistors biased inthe subthreshold region became popular, especially for lowpower design including biomedical applications. For instance,Gerosa et al. [51] presented a log-domain preamplifier and filterfor a pacemaker that dissipated at most 2.8 W. Bartolozzi etal. [52] proposed a current mode filter for neuromorphic systemwith less than 1 nW. Lim et al. [53] designed an amplifier forrecording ECG that consumed less than 20 W.

III. LOW FREQUENCY DESIGN METHODS

Most physiological signals are low frequency signals as illus-trated in Fig. 1. Therefore, the circuits often need to be designedwith large time constants ( or ), which implythat large capacitances, large resistances, and/or small transcon-ductances (mostly referring to the transconductances of OTAs)have to be implemented. For example, using the UMC 0.13- mCMOS process where a 1-pF capacitor and a 1- resistor oc-cupy 10 mm and 10 mm , respectively, the minimum areato achieve a cutoff frequency of 1 Hz based on the equation

is 0.8 mm , where a 400 pF capacitor and a 0.4resistor are used. The total area of the resistor and capacitor isconsiderably large. Alternatively, a cutoff frequency of 1 Hz canalso be implemented by a 100 pF capacitor and a 0.628 nA/Vtransconductance based on , assuming that 100 pF isthe largest acceptable capacitor. However, designing OTAs withtransconductances below 1 nA/V is also a challenge especiallyif noise performance, dynamic range and chip area are all con-sidered. Some special techniques have been explored to solvethe above problems.

A. MOS Pseudo-Resistor

A resistance of hundreds of mega ohms can be achievedwhen a MOSFET is biased in the subthreshold region. Thiskind of MOS resistor is commonly used together with a ca-pacitor in neural recording circuits for clamping large dc driftat the recording site. The neural recording amplifier proposed


TABLE IOTA-C FILTERS USED FOR PROCESSING PHYSIOLOGICAL SIGNALS

Fig. 5. MOS pseudo-resistors: (a) common tunable MOS pseudo resistor [57],(b) balanced tunable MOS pseudo resistor [57], and (c) MOS-bipolar pseudoresistor [60].

by Chandran et al. [54] used an NMOSFET biased in sub-threshold region and an electrode capacitance to achieve acutoff frequency below 20 Hz. The amplifier can tolerate a dcinput voltage ranging from 0 to 0.4 V without sacrificing the acperformance. Nevertheless, the design cannot reject negativedc inputs. Mohseni et al. [55] proposed an optimized neuralrecording amplifier using the dc baseline stabilization scheme,where a PMOSFET biased in subthreshold region is usedtogether with the recording probe capacitance of about 22 pFto bring the cutoff frequency below 50 Hz. The tolerable rangeof dc inputs was measured to be at least 0.25 V. In these twodesigns [54], [55], resistors were employed to bias the MOSresistors. These biasing resistors were laser trimmed to makethe cutoff frequencies tunable. Olsson et al. [56] designedtheir neural recording amplifier by using diode connected sub-threshold NMOSFETs that had an equivalent resistance greaterthan 15.9 without using biasing circuit. The lower-bandcutoff frequency was measured to be less than 10 Hz and thetolerable range of dc input voltages was 0.5 V.

Since tuning a resistor by trimming is inconvenient and costly,a tunable MOS pseudo resistor was proposed [57], as shown inFig. 5(a). By changing the gate voltage of the MOSFETs, the cir-cuit can be used as a band-tunable extracellular neural chip forrecording both field potentials and action potentials. The toler-able dc input voltage range was measured to be 0.5 V. Consid-ering this MOS resistor exhibits asymmetric and nonlinear resis-tance when the voltage across it varies, Zou et al. [58] proposed

a balanced tunable MOS pseudo resistor in their programmablebiomedical sensor interface chip, as illustrated in Fig. 5(b). Thetwo transistors can be turned on alternatively to achieve a sym-metric incremental resistance curve. The resistance was claimedalmost constant when the voltage across it changed.

In addition, the MOS-bipolar device, which is a drain-gateshorted MOSFET, has also been proposed for implementinglarge resistance [59]. Take a PMOSFET as an example, withnegative , it functions as a diode connected PMOSFET;with positive , the parasitic BJT is turned on and worksas a diode connected BJT. Harrison et al. [60] employed thistechnique to design circuit with a low cutoff frequency. TwoMOS-bipolar devices connected in series were used to reducethe distortion of large output signals, as shown in Fig. 5(c).The cutoff frequency of the amplifier was 0.025 Hz, in whichthe equivalent resistance of the device was larger than 10 .This structure has been adopted in the neural recording ampli-fier presented by Wattanapanitch et al. [61] and the bioampli-fier proposed by Gosselin et al. [62]. In addition, Lim et al.[53] proposed a high-pass filter using two oppositely connectedMOS-bipolar devices and a 2-pF capacitor to achieve a cutofffrequency of 0.45 Hz, which was claimed to have better linearitythan Harrison’s configuration.

B. Low Cutoff Frequency Filter Design

1) OTA-C Filter: OTA-C filters have been commonly used inprocessing physiological signals as shown in Table I, which listssome representative designs published recently. Table II sum-marizes several techniques that have been proposed to imple-ment small transconductances for designing these filters withlow cutoff frequency. Details of these techniques are presentedas follows.

a) Small transconductance OTA design: Current divi-sion technique has been proposed to reduce the small signaltransconductances in voltage to current converters and Fig. 6(a)


Fig. 6. OTAs based on (a) current division [69], (b) current cancellation [69], and (c) current division and current cancellation [63] techniques.

Fig. 7. (a) Generic series-parallel current mirror. Unit transistors are identical � � � � � to achieve effective copy factor � � �� [71] and(b) OTA based on series-parallel current division technique [71].

TABLE IITECHNIQUES FOR DESIGNING OTAS WITH SMALL TRANSCONDUCTANCE

shows an OTA using this method. The drain current of MR isdivided by MM and M1, most of which flows to the groundthrough MM when . The small signaltransconductance of the OTA is given by [69]

(4)

where is the small signal drain-source conductance ofMR, , and are thetransconductances of MM and M1, respectively. When in-creases, the dc current of M1 and M2 decreases at the same rate,which will induce an increasing leakage current. As a result, theoffset voltage caused by the leakage current cannot be omitted.Thus, there will be a tradeoff between transconductance andoffset voltage when choosing .

Current cancellation was first presented by Garde [70] inbipolar OTAs design. Jose et al. [69] applied it into MOS am-plifiers. As shown in Fig. 6(b), the currents of M1 and MN are

partially cancelled at the output. The small signal transconduc-tance of the OTA can be expressed as [69]

(5)

where , and , are transconductances ofMN and M1 respectively. In order to reduce , is usuallyset to approach to 1. In fact, it is limited between 0.5 and 0.9[69]. can be reduced to 10 A/V for in the order of10 A/V.

The above methods are insufficient to implement transcon-ductances at the nV/A level. In order to get an even smallertransconductance, the above methods can be combined togetheras illustrated in Fig. 6(c). The current of MR, operating in trioderegion, is divided by MM, M1, and MN, most of which flows toground when , . The small signal transconduc-tance of the OTA can be written as [63]

(6)

This method has been used frequently when designing filters,e.g., the low-pass filter for medical applications [63], the low-pass notch filter for EEG system [65] and the filter for portableECG detection [68], all of which achieved transconductances ofa few nA/V.

Another technique is Series-Parallel (SP) current divisionstructure, which reduces the transconductance based on SP cur-rent mirror. The generic structure is shown in Fig. 7(a), wherethe copy factor is . Assumeand for the amplifier shown in Fig. 7(b), we obtain

. In order to get a transconductance of pico level,a high division factor must be chosen [71].

This technique requires choosing a bias current to tradeoffbetween the transconductance and input linear range. On


Fig. 8. Schematic of capacitor scaler [74].

one hand, reducing the biased current is needed for a smalltransconductance. On the other hand, increasing the biasedcurrent widens the input linear range, which is approximated as

[72], where is the subthresholdslope factor, is the inversion coefficient of MOSFET propor-tional to the bias current, is the thermal voltage, and isthe maximum acceptable relative error of the input differentialcurrent within the linear range, i.e., [72]

for (7)

where , is the transconductance of M1, asshown in Fig. 7(b). By using this method, OTAs with transcon-ductances varying from 33 pA/V to a few nA/V can be obtained[67], [71].

In addition to the above techniques, the OTA based on sub-threshold MIFG MOS transistors in [43] achieved 15 p–150pA/V transconductances when the bias current ranged from 1to 100 nA. The corresponding cutoff frequency of the OTA-Cfilter spanned from 0.2 to 15 Hz with an ultra power consump-tion of no more than 2 W. This work provides us a promisingway in low frequency and low power designs.

b) Large capacitor on chip: According to the Miller effect,the effective impedance can be reduced by increasing the inputcurrent while keeping the same input voltage. Specifically, if theinput current is sampled, amplified, and fed back into the input,the equivalent impedance will be scaled down with the currentamplification factor [73], [74]. Impedance scaler, as shown inFig. 8, is developed based on this concept. The small signal ad-mittance is given by [63]

(8)where and are the small signal drain source con-ductances of MSN and MSPN respectively, is the ratio of thesmall signal transconductances of MSN and MS1 respectively,

, and , are parasitic capacitors of thecorresponding node. By putting the pole muchhigher than the pass-band frequency, reducing and ,and omitting , (8) can be simplified as

(9)

A large equivalent capacitance can be achieved in a small chiparea by adopting a large and small basic capacitor . Forexample, basic capacitors of 5 and 25 pF were used to achieveequivalent capacitances of 18–200 and 125 pF, respectively

Fig. 9. Current steering filter [75].

Fig. 10. Simple MOSFET switch modulator.

[63], [67]. A disadvantage of such capacitance multiplicationtechniques is the limited linear range it possessed, which isinversely proportional to the multiplication ratio.

2) Current Steering Filter: An R-MOSFET-C filter usingcurrent steering method is shown in Fig. 9, which was designedaiming to achieve a low cutoff frequency [75]

(10)The cutoff frequency can be tuned by changing the gate voltageof M1 , which in turn steered the current that flow from thecapacitor to virtual ground. The gain of the filter was determinedby . The filter not only achieved a cutoff frequency of aslow as 1.8 Hz without off-chip components but also exhibitedlow distortion [75].

IV. LOW NOISE DESIGN METHODS

Physiological signals are prone to be interfered by the noisescaused by sensors, electrodes, environment, power supplies andpower frequency, etc. Therefore, techniques must be employedto optimize the noise performance of the circuits.

Chopper stabilization (CHS) is commonly used for pro-cessing bioelectrical signals, like EEG, ECG, and EMG. InCHS method, the signal is firstly modulated to a higher fre-quency where there is no noise and then demodulated backto the baseband after amplification [76]. A simple MOSFETswitch modulator is shown in Fig. 10.

The differential difference chopper stabilization amplifier(CHSDDA) is popular as an alternative instrumentation ampli-fier. Chan et al. [77] proposed this topology for EEG recordingto reduce the mismatches, noises, and offsets. The measuredcommon mode rejection ratio (CMRR) with practical mismatchwas relatively higher than three operational amplifiers instru-mentation amplifier (3OPIA) and current mode instrumentationamplifier (CMIA), which were simulated with 0.2% resistormismatch and 0.2% transistor mismatch, respectively. In addi-tion, some other amplifiers designed for recording multichannelcortical signals [78], ENG [79], neural field potentials [80],


EEG [81] achieved noise efficiency factors (NEFs) of 3.35, 5.3,4.6, 4.9 respectively when using the CHS method.

Attention must be paid to the charge injection and clockfeedthrough effects in simple MOS switches [17]. Chargeinjection usually introduces three types of errors in the outputvoltage of a MOS switch, the gain error, nonlinearity and dcoffset [17]

(11)

where is the Fermi potential in the semiconductor, isthe body effect coefficient, is the gate oxide capaci-tance per unit area, and is the hold capacitor. Effortshave been made to cancel the charge injection and improvethe performance of the MOS switch modulator, such asadopting large shunting capacitor and slow transition time,using symmetrical capacitances at drain and source of theMOS switch with half-sized dummy switches, etc. [76]. Clockfeedthrough usually introduces a constant offset voltage, whichequals [16], where is thegate-source or gate-drain overlap capacitance per unit width,and is the voltage during the high state of the clock signal.

In addition to CHS, Chan et al. employed auto-zeroing (AZ)in the amplifier for cortical neural prostheses. The basic idea ofAZ is sampling the noise and offset (in the sampling phase) andthen subtracting it from the instantaneous value of signal (in theprocessing phase) either at the input or the output of the circuits[76]. AZ was claimed to perform better than CHS in reducingpower consumption because the modulation frequency in CHSmust be twice the input signal in order to meet the Nyquist cri-teria, resulting in a higher bandwidth and bias current [82]. Anamplifier using the AZ technique has to be disconnected fromthe input terminal in order to sample and hold its own offset andnoise in the sampling phase. Therefore, the amplified signal isonly available during the signal processing phase when using asingle auto-zeroed amplifier [76].

In addition, designers tend to use more specific ways toreduce noises, such as elaborate layout topology considerationsand design special circuits. For example, for the high-PSRRmicrophone preamplifier [83], several methods were employedto reduce the noise caused by power supply. The output ofthe buffer of the microphone was taken from the drain of theMOSFET instead of the source to avoid injecting noise fromthe power supply to the output due to the MOSFET parasiticcoupling effect. In addition, wide-band power supply rejectionwas achieved by using a novel power supply filter. Another ex-ample is an OTA for neural recording [61], which was designedwith a maximized transconductance under a given total currentto achieve a low input-referred noise. The NEF of the amplifierwas 2.47. It nearly achieved the theoretical limit of the NEF(2.02) of an OTA that used a differential pair as an input stage.

V. INTEGRATED CIRCUITS FOR PROCESSING DIFFERENT

PHYSIOLOGICAL SIGNALS

Different IC designs for processing different physiologicalsignals have been published in recent years.

A. Bioelectrical Signal Processing

Cardiovascular diseases are considered to be the leadingcause of death globally. ECG monitoring can be used to fore-cast possible heart diseases. ECG is a small amplitude, lowfrequency signal in the range of 0.5–4 mV and 0.01–250 Hz[10]. It is usually affected by the noises from the skin-electrodeinterface, muscular activity, etc. High CMRR and low cutofffrequency amplifiers are needed to detect the small differentialsignal. Several groups have developed ICs for ECG measure-ment. Lasanen et al. [84] implemented an ECG measurementchip in a 0.18- m CMOS process with 1 V–1.8 V supply,3 A averaged current and 82 dB CMRR for miniature deviceslike heart rate detector. There were only two electrodes in thiscircuit, and the analog ground was internally generated by thebias circuit. Wong et al. [85] implemented an ECG measure-ment chip with a driven-right-leg circuit in a 0.35- m CMOSprocess. The technique of using an electrode connected to theright leg as a reference has been widely used in discrete com-ponent circuits, but it was the first time that the driven-right-legcircuit was implemented on chip. In addition, Fay et al. [86]proposed an ECG processing amplifier in a 0.5 um processwith 2.8 W power consumption and 90 dB CMRR, whichused an active grounding electrode to attenuate the 60 Hz noise,weak inversion transistors to improve the noise efficiency, andcapacitor-based amplification to improve matching. Featuresof ECG, including P, Q, R, S, and T waves, can be attained inthis design.

EEG is another small amplitude (5–300 V), low frequency(dc-150 Hz for scalp EEG) bioelectrical signal that plays an im-portant role in diagnosing disorders like epilepsy, coma, stroke,and investigating cognitive state. Ng et al. [87] presented a16-channel analog front-end chip for EEG/ECG monitoring.CHSDDIA previously mentioned was used in this work toachieve a high CMRR (115 dB) and a low input-referred noise(0.86 Vrms, 0.3–150 Hz). Another example is the 8-channelEEG acquisition ASIC proposed by Refet et al. [88]. In thisdesign, each channel utilized a new ac coupled chopper stabi-lized IA (ACCIA) with coarse-fine servoloop to improve thenoise performance and reduce the power consumption. Thedesign achieved 120 dB CMRR, 0.59 Vrms input-referrednoise (0.5–100 Hz) and consumed 200 W. Compared withthe former work [87] (485 A from 1.5 V supply), the powerconsumption was largely reduced in this design (66 A from3 V supply).

Implantable neural recording systems with microelectrodearrays for observing the activity of the neurons in the brainare useful for understanding how the brain works. Many ex-cellent designs have been published in this area, including awireless 100-electrode neural recording system proposed byHarrison et al. [89] and a wireless 64-channel neural recordingsystem published by Sodagar et al. [90]. The two systems aretelemetry-powered systems that consumed 13.5 and 14.4 mW,respectively. In addition, there are some designs focusing onthe front-end amplifiers like the neural recording amplifierdesigned for recording neural spikes and local field potentials[61]. As mentioned previously, the amplifier nearly achieved


Fig. 11. (a) Pacemaker [92]. (b) Bionic cochlea system [95], which usually includes microphone, transmitter, receiver/stimulator, electrodes and speech processor.

the theoretical limit of the NEF of an OTA that uses a differen-tial pair as an input stage.

Another important application is a pacemaker that is com-monly used for treating bradycardia. The device monitors theheart’s rate and rhythm by sensing cardiac signals and provideselectrical stimulation when the heart does not beat or beats tooslowly [91]. Fig. 11(a) shows a typical pacemaker. Gerosa etal. [51] proposed a fully integrated preamplifier and filter for animplantable cardiac pacemaker to detect the spontaneous heartactivity. This circuit was fabricated in a 0.35- m CMOS processwith a 1.8 V supply voltage and 1.8 A current. Almost all thetransistors were biased in the subthreshold region to meet theexponential I-V relationship required by log-domain approachto result in ultra low power consumption. Wong et al. [92] pro-posed a very low power IC for implantable pacemaker, whichwas fabricated in a 0.5- m CMOS process. Most of the transis-tors in the analog part worked in the deep subthreshold region.The device dissipated an average power of 8 W at approxi-mately 2.8 V supply voltage and had an estimated longevity of5–10 years with a primary battery.

B. Acoustic Signal Processing

The disturbance of the normal breathing process may causesevere metabolic, organic, and central nervous disorders or evendeath [93]. Small wearable devices with low power consump-tion are needed to monitor the respiration process and warnof the cessation of breathing. Corbishley et al. [29] proposeda nanopower OTA-C band-pass filter to be used in a wearablebreathing detector for capturing the acoustic signal caused bybreathing. The filter, which was designed with transconductanceamplifiers biased in the subthreshold and fabricated in a 0.35-m CMOS process with a 1 V supply voltage, consumed only70 nW. The acoustic signal received by the microphone can befirst processed by the band-pass filter before sending to the recti-fier, low-pass filter and comparator. By comparing with the pro-cessed signal with a predetermined threshold, the system willdecide whether respiration is detected. An alarm signal can besent if respiration ceases or the device is dislodged [29].

The incurable damage of some important organs, like ear,eye, and some nerves is suffering for people. Generating neuralaction potentials by electrical stimulations to control the dys-functional organs is proved a promising way to relieve people’spain. The first example is bionic ear. The inner ear includesthe cochlea and the vestibule, which is responsible for trans-mitting sound and inertia to the vestibulocochlear nerve, re-spectively. The bionic cochlea usually consists of an implantedmodule, including a receiver-stimulator and electrodes, and anexternal speech processor [94], as shown in Fig. 11(b). Thesound is received and processed by the speech processor. Theprocessed signal will then be transmitted to the internal moduleto stimulate the auditory nerve. In order to be more comfort-able and convenient, the next generation bionic cochlea will befully implantable. Sarpeshkar et al. [96] reported an ultra-lowpower programmable analog bionic ear processor, which aimedto be used in fully implantable bionic cochleas. Many tech-niques were used in this design to reduce power and improvenoise performance, as mentioned in Section IV. The implantcan be operated on a 100 mAh battery with a 1000 charge-and-discharge-cycle lifetime for 30 years. Another example is the126 W cochlear chip for a totally implantable system designedby Georgiou et al. [97]. The speech processor and the stimu-lator were implemented on one mixed-signal chip with size andpower consumption sufficiently small for a fully implantableapplication. As another part of the inner ear, the vestibule is re-sponsible for transmitting inertia to the vestibulocochlear nerve.A vestibular prosthesis, commonly includes an inertia sensor,interfacing, processing, and stimulating modules, brings anotherimportant function to bionic ears [98]–[100].

Heart sounds, noises generated by the closing of the heartvalves, and the resultant flow of blood through them, areuseful for diagnosing diseases such as stenosis that restricts theopening of a heart valve. A battery-free tag, which includeda low power IC, an antenna and microphones, to wirelesslymonitor heart sounds has been developed [101]. The IC con-sumed only 1 W. The tag was battery-less and powered byharvesting radiated RF energy. It was demonstrated that thetag can reliably measure heart rate through heart sounds at adistance up to 7 m from an FCC-compliant RF power source.


It is worth mentioning that heart activity was monitored byacoustic and not electrical signals in this work such that noelectrical contact is needed.

C. Physical Quantity Processing

Photoplethysmograhy is a noninvasive method to measure thevolume changes in vessels and has been used in SpO evalua-tion [102], cuffless blood pressure estimation [103], and heartrate measurements. Photoplethysmogram (PPG) is a very lowfrequency signal (0.5–17 Hz) and its ratio between the ac anddc components is as small as about 0.001–0.015 [104], [105].It is easily disturbed by motion artifact and environment noises.Wong et al. [75] presented a near-infrared heart rate measure-ment chip by processing PPG. A low-pass current steering filtermentioned in Section III was employed in this circuit to get acutoff frequency as low as 0.25 Hz without off chip components.This is the first chip for processing PPG.

Bladder diseases may lead to various complications or evendeath. Many bladder diseases can be observed or predicted bylong-term invasive monitoring of the bladder urine pressure forsyndromes of urinary anomalisms. Wang et al. [106] proposedan invasive long term bladder urine pressure measuring systemthat included a controlling ASIC, a pressure sensor and a RFmodule. The ASIC was implemented in the TSMC 0.35- mCMOS process, and the whole system consumed 1.25 mV. Theoutput voltage of the sensor that was proportional to the abso-lute pressure was processed by the ASIC, and then the data waswirelessly delivered to an external data analyzer for diagnosis.

D. Chemical Quantity Processing

Chemical sensors based on ion-sensitive field effect transis-tors (ISFETs) have been widely used in ion concentration mea-surement. An ISFET is a MOS transistor with the gate connec-tion separated from the device in the form of a reference elec-trode inserted into an aqueous solution which is in contact withthe gate oxide [107]. The drain current of the ISFET can be ex-pressed as a function of the hydrogen ion concentration both instrong inversion region and weak inversion region [108]. As anexample, Pantelis et al. [109] proposed a silicon pancreatic betacell, which was used for real-time glucose sensing and insulinrelease for diabetes. The silicon beta cell has been fabricatedin the UMC 0.25- m CMOS process with a measured powerconsumption of 4.5 W. In this paper, the ISFET biased in thesubthreshold region was employed to model the metabolic func-tions. The one-to-one relationship between hydrogen ions andglucose ions can be constructed according the reaction of glu-cose with the enzyme. By using the relation between hydrogenion concentration and drain current of an ISFET, the connectionbetween the glucose ion concentration and the drain current ofthe ISFET can be founded.

VI. CONCLUSION

In this paper, low power, low frequency, and low noise analogIC design techniques for processing physiological signals arereviewed and some excellent related designs are listed. It is en-visaged that there will be a strong demand for medical ICs inan increasing number of novel medical applications, in additionto applications such as monitoring physiological parameters or

developing bionic organs. Moreover, the analog, digital, and RFmodules as well as sensors or electrodes will be integrated orpackaged into a system to perform multiple and more complexfunctions. New techniques for designing ICs of medical applica-tions are therefore needed in order to improve the performance,lower the power consumption, and reduce the physical size ofthe wearable and implantable medical instrumentations.

ACKNOWLEDGMENT

The authors are grateful to the reviewers for their constructivecomments and valuable inputs, which were useful in improvingthe quality of this paper. The authors are grateful to Stan-dard Telecommunication Ltd., Jetfly Technology Ltd., GoldenMeditech Company Ltd., Bird International Ltd., Bright StepsCorporation and PCCW for their support to the ITF projects.

REFERENCES

[1] P. Bonato, “Wearable sensors/systems and their impact on biomedicalengineering,” IEEE Eng. Med. Biol. Mag., vol. 22, no. 3, pp. 18–20,May/Jun. 2003.

[2] D. W. Ryoo, Y. S. Kim, and J. W. Lee, “Wearable systems for ser-vice based on physiological signals,” in Proc. 27th Conf. IEEE EMBS,Shanghai, China, 2005, pp. 2437–2440.

[3] X. F. Teng and Y. T. Zhang, “M-health: Trends in wearable medicaldevices,” Chinese J. Med. Instrum., vol. 30, no. 5, pp. 330–340, Sep.2006.

[4] A. Lymberis, “Smart wearable systems for personalised health man-agement: Current R&D and future challenges,” in Proc. 25th Ann. Int.Conf. IEEE EMBS, Cancun, Mexico, 2003, pp. 3716–3719.

[5] G. Tröster, “The agenda of wearable healthcare,” in IMIA Year-book of Medical Informatics 2005: Ubiquitous Health Care Sys-tems. Stuttgart, Germany: Schattauer, 2004, pp. 125–138.

[6] J. E. Mezzich, “Psychiatry for the person: Articulating medicine’s sci-ence and humanism,” World Psychiatry, vol. 6, no. 2, pp. 65–67, Jun.2007.

[7] X. F. Teng, Y. T. Zhang, C. C. Y. Poon, and P. Bonato, “Wearablemedical systems for p-health,” IEEE Rev. Biomed. Eng., vol. 1, pp.62–74, 2008.

[8] C. C. Y. Poon and Y. T. Zhang, “Perspectives on high technologies forlow-cost healthcare,” IEEE Eng. Med. Biol., vol. 27, no. 5, pp. 42–47,Oct. 2008.

[9] Y. T. Zhang, C. C. Y. Poon, and E. Macpherson, “Editorial note onhealth informatics,” IEEE Trans. Inform. Technol. Biomed., vol. 13, no.3, pp. 281–283, May 2009.

[10] J. G. Webster, Medical Instrumentation Application and De-sign. New York: Wiley, 1998, pp. 10–11.

[11] T. M. Seese, H. Harasaki, G. M. Saidel, and C. R. Davies, “Charac-terization of tissue morphology, angiogenesis, and temperature in theadaptive response of muscle tissue to chronic heating,” Lab. Invest.,vol. 78, no. 12, pp. 1553–1562, 1998.

[12] E. H. Liu, G. M. Saidel, and H. Harasaki, “Model analysis of tissueresponses to transient and chronic heating,” Ann. Biomed. Eng., vol.31, no. 8, pp. 1007–1014, Sep. 2003.

[13] Radio electronics: Battery technology overview [Online]. Available:http://www.radio-electronics.com/info/data/battery-technology/tuto-rial-basics-overview.php

[14] M. H. Miles, “Recent advances in lithium battery technology,” in Proc.Gallium Arsenide Integrated Circuit (GaAs IC) Symp., 23rd Annu.Technical Dig., Baltimore, MD, 2001, pp. 219–222.

[15] Alternative energy news: Battery technology [Online]. Available:http://www.alternative-energy-news.info/technology/battery-power/

[16] P. E. Allen and D. R. Holberg, CMOS Analog Circuit Design. NewYork: Oxford Univ. Press, 2002, p. 99.

[17] B. Razavi, Design of analog CMOS Integrated Circuits. Singapore:McGraw-Hill, 2002, pp. 27–28.

[18] R. R. Troutman and S. N. Chakravarti, “Subthreshold characteristicsof insulated-gate field-effect transistors,” IEEE Trans. Circuit Theory,vol. CT-20, pp. 659–665, Nov. 1973.

[19] R. R. Troutman, “Subthreshold design considerations for insulated-gate field-effect transistors,” IEEE J. Solid-State Circuits, vol. SC-9,pp. 55–60, Apr. 1974.


[20] E. Vittoz and J. Fellrath, “CMOS analog integrated circuits based onweak inversion operation,” IEEE J. Solid-State Circuits, vol. 12, no. 3,pp. 224–231, Jun. 1977.

[21] C. Jakobson, I. Bloom, and Y. Nemirovsky, “1/f noise in CMOS tran-sistors for analog applications from subthreshold to saturation,” Solid-State Electron, vol. 42, no. 10, pp. 1807–1817, 1998.

[22] D. J. Comer and D. T. Comer, “Operation of analog MOS circuits inthe weak or moderate inversion region,” IEEE Trans. Education, vol.47, no. 4, pp. 430–435, Nov. 2004.

[23] D. J. Comer and D. T. Comer, “Using the weak inversion region tooptimize input stage design of CMOS op amps,” IEEE Trans. CircuitsSyst. II: Expr. Briefs, vol. 51, no. 1, pp. 8–14, Jan. 2004.

[24] T. M. Hollis, D. J. Comer, and D. T. Comer, “Optimization of MOSamplifier performance through channel length and inversion level se-lection,” IEEE Trans. Circuits Syst. II: Expr. Briefs, vol. 52, no. 9, pp.545–549, Sep. 2005.

[25] D. Bol, R. Ambroise, D. Flandre, and J. D. Legat, “Interests and limi-tations of technology scaling for subthreshold logic,” IEEE Trans. VeryLarge Scale Integration (VLSI) Syst., vol. 17, no. 10, pp. 1508–1519,Oct. 2009.

[26] C. C. Enz, F. Krummenacher, and E. A. Vittoz, “An analytical MOStransistor model valid in all regions of operation and dedicated to low-voltage and low-current applications,” Analog Integrat. Circuits SignalProcess., vol. 8, no. 1, pp. 83–114, Jul. 1995.

[27] A. I. A. Cunha, M. C. Schneider, and C. Galup-Montoro, “An MOStransistor model for analog circuit design,” IEEE J. Solid-State Cir-cuits, vol. 33, pp. 1510–1519, Oct. 1998.

[28] S. S. Rajput and S. S. Jamuar, “Low voltage analog circuit design tech-niques,” IEEE Mag. Circuits Syst., vol. 2, no. 1, pp. 24–42, 2002.

[29] P. Corbishley and E. Rodriguez-Villegas, “A nanopower bandpass filterfor detection of an acoustic signal in a wearable breathing detector,”IEEE Trans. Biomed. Circuits Syst., vol. 1, no. 3, pp. 163–171, Sep.2007.

[30] J. Kim, M. S. Chae, and W. Liu, “A 220 nW neural amplifier for multi-channel neural recording systems,” in IEEE Int. Symp. Circuits Syst.,Taipei, Taiwan, 2009, pp. 1257–1260.

[31] E. Lopez-Morillo, R. G. Carvajal, F. Munoz, H. E. Gmili, A. Lopez-Martin, J. Ramirez-Angulo, and E. Rodriguez-Villegas, “A 1.2 V 140nW 10 bit sigma-delta modulator for electroencephalogram applica-tions,” IEEE Trans. Biomed. Circuits Syst., vol. 2, no. 3, pp. 223–230,Sep. 2008.

[32] R. H. Dennard, F. H. Gaensslen, H. N. Yu, V. L. Rideout, E. Bassous,and A. R. LeBlanc, “Design of ion-implanted MOSFET’s with verysmall physical dimensions,” IEEE J. Solid-State Circuits, vol. SC-9,pp. 256–268, 1974.

[33] T. H. Ning, “A perspective on the theory of MOSFET scaling andits impact,” IEEE J. Solid-State Circuits Newslett., vol. 12, no. 1, pp.27–30, Jan. 2007.

[34] A. Guzinski, M. Bialko, and J. C. Matheau, “Body driven differen-tial amplifier for application in continuous time active-C filer,” in Eur.Conf. Circuit Theory Design, Paris, France, 1987, pp. 315–320.

[35] Y. Haga, H. Zare-Hoseini, L. Berkovi, and I. Kale, “Design of a0.8 volt fully differential CMOS OTA using the bulk-driven tech-nique,” in IEEE Int. Symp. Circuits Syst., May 23–26, 2005, pp.220–223.

[36] Y. Haga, R. C. S. Morling, and I. Kale, “A new bulk-driven input stagedesign for sub 1 volt CMOS opamps,” in Int. Sym. IEEE Circuits Syst.,Kos, Greece, 2006, pp. 1547–1550.

[37] J. M. Carrillo, P.-A. Raquel, J. M. Valverde, J. F. Duque-Carrillo,and G. Torelli, “Compact low-voltage rail-to-rail bulk-driven CMOSopamp for scaled technologies,” in Eur. Conf. Circuit Theory Design,Antalya, 2009, pp. 263–266.

[38] E. Sanchez-Sinencio, Bulk Driven Transistors [Online]. Available:http://amsc.tamu.edu/

[39] K. Lasanen, E. Raisanen-Ruotsalainen, and J. Kostamovaara, “A 1V 5 uW CMOS opamp with bulk-driven input transistors,” in Proc.43rd IEEE Midwest Symp. Circuits System, Lansing, Michigan, pp.1038–1041.

[40] S. W. Pan, C. C. Chuang, C. H. Yang, and Y. S. Lai, “A novel OTAwith dual bulk-driven input stage,” in Proc. Int. Sym. IEEE CircuitsSyst., Taipei, Taiwan, 2009, pp. 2721–2724.

[41] D. Kahng and S. M. Sze, “A floating-gate and its application to memorydevices,” Bell Syst. Tech. J., vol. 46, no. 4, pp. 1288–1295, 1967.

[42] E. R. Villegas, “A low-power wide range transimpedance amplifierfor biochemical sensing,” in Proc. 29th Annu. Int. Conf. IEEE Eng.Medicine Biol. Soc., Lyon, France, Aug. 2007, pp. 2673–2676.

[43] A. E. Mourabit, G. N. Lu, and P. Pittet, “Rail-to-rail differential linearrange OTA with pico-A/V transconductance for sub-Hertz OTA-Cfilter,” in Proc. 16th Int. Conf. Microelectronics, Tunis, 2004, pp.647–650.

[44] E. Rodriguez-Villegas, A. Yufera, and A. Rueda, “A 1.25 V microp-ower Gm-C filter based on FGMOS transistors operating in weak in-version,” IEEE J. Solid-State Circuits, vol. 39, no. 1, pp. 100–111, Jan.2004.

[45] E. Rodriguez-Villegas, A. Yufera, and A. Rueda, “A 1 V micropowerlog-domain integrator based on FGMOS transistors operating in weakinversion,” IEEE J. Solid-State Circuits, vol. 39, no. 1, pp. 256–259,Jan. 2004.

[46] J. Ramirez-Angulo, R. G. Carvajal, J. Tombs, and A. Torralba,“Low voltage CMOS opamp with rail-to-rail input and output signalswing for continuous-time signal processing using multiple-inputfloating-gate transistors,” IEEE Trans. Circuits Syst. II, vol. 48, no. 1,pp. 111–116, Jan. 2001.

[47] P. Hasler, “Foundations of learning in analog VLSI,” Ph.D. disserta-tion, California Inst. Technology, Pasadena, CA, 1997.

[48] E. Rodriguez-Villegas and H. Barnes, “Solution to trapped chargein FGMOS transistors,” IEE Electron. Lett., vol. 39, no. 19, pp.1416–1417, Sep. 2003.

[49] J. Ramirez-Angulo, C. A. Urquidi, R. Gonzalez-Carvajal, A. Torralba,and A. Lopez-Martin, “A new family of very low-voltage analogcircuits based on quasi-floating-gate transistors,” IEEE Trans. CircuitsSystems II: Analog Digital Signal Processing, vol. 50, no. 5, pp.214–220, May 2003.

[50] F. Serra-Graells, A. Rueda, and J. L. Huertas, Low-Voltage CMOSLog Companding Analog Design. Hingham, MA: Kluwer, 2003, pp.5–10.

[51] A. Gerosa, A. Maniero, and A. Neviani, “A fully integrateddual-channel log-domain programmable preamplifier and filterfor an implantable cardiac pacemaker,” IEEE Trans. Circuits Syst. I:Reg. Papers, vol. 51, no. 10, pp. 1916–1925, Oct. 2004.

[52] C. Bartolozzi, S. Mitra, and G. Indiveri, “An ultra low power cur-rent-mode filter for neuromorphic systems and biomedical signal pro-cessing,” in Proc. IEEE Biomedical Circuits Syst. Conf., London, U.K.,2006, pp. 130–133.

[53] E. C. M. Lim, X. D. Zou, Y. J Zheng, and J. Tan, “Design of low-power low-voltage biomedical amplifier for electrocardiogram signalrecording,” in Proc. IEEE Biomedical Circuits Syst. Conf., Montréal,Canada, 2007, pp. 191–194.

[54] A. P. Chandran, K. Najafi, and K. D. Wise, “A new dc baseline stabi-lization scheme for neural recording microprobes,” in Proc. 21th Conf.IEEE EMBS, Atlanta, GA, 1999, p. 386.

[55] P. Mohseni and K. Najafi, “A fully integrated neural recording ampli-fier with DC input stabilization,” IEEE Trans. Biomed. Eng., vol. 51,no. 5, pp. 832–837, May 2004.

[56] R. H. Olsson, III, M. N. Gulari, and K. D. Wise, “Silicon neuralrecording arrays with on-chip electronics for in vivo data acquisition,”in Proc. 2nd Annu. Int. Conf. Microtechnologies Medicine Biology,Madison, WI, 2002, pp. 237–240.

[57] R. H. Olsson, III, D. L. Buhl, A. M. Sirota, G. Buzsaki, and K.D. Wise, “Band-tunable and multiplexed integrated circuits for si-multaneous recording and stimulation with microelectrode arrays,”IEEE Trans. Biomed. Eng., vol. 52, no. 7, pp. 1303–1311, Jul.2005.

[58] X. D. Zou, X. Y. Xu, L. B. Yao, and Y. Lian, “A 1 V 450 nW fullyintegrated programmable biomedical sensor interface chip,” IEEE J.Solid-State Circuit, vol. 44, no. 4, pp. 1067–1077, Apr. 2009.

[59] T. Delbrück and C. A. Mead, “Analog VLSI adaptive, logarithmic widedynamic range photoreceptor,” in Proc. Int. Sym. IEEE Circuits Syst.,London, U.K., 1994, pp. 339–342.

[60] R. R. Harrison and C. Charles, “A low-power low-noise CMOS am-plifier for neural recording applications,” IEEE J. Solid-State Circuits,vol. 38, no. 6, pp. 958–965, Jun. 2003.

[61] W. Wattanapanitch, M. Fee, and R. Sarpeshkar, “An energy-efficientmicropower neural recording amplifier,” IEEE Trans. Biomed. CircuitsSyst., vol. 1, no. 2, pp. 136–147, Jun. 2007.

[62] B. Gosselin, M. Sawan, and C. A. Chapman, “A low-power integratedbioamplifier with active low-frequency suppression,” IEEE Trans.Biomed. Circuits Syst., vol. 1, no. 3, pp. 184–192, Sep. 2007.

[63] S. Solis-Bustos, J. Silva-Martinez, F. Maloberti, and E. Sanchez-Sinencio, “A 60 dB dynamic-range CMOS sixth-order 2.4 Hzlow-pass filter for medical applications,” IEEE Trans. Circuits Syst. II,vol. 47, no. 12, pp. 1391–1398, Dec. 2000.


[64] C. D. Salthouse and R. Sarpeshkar, “A practical micropower pro-grammable bandpass filter for use in bionic ears,” IEEE J. Solid-StateCircuits, vol. 38, no. 1, pp. 63–70, Jan. 2003.

[65] X. B. Qian, Y. P. Xu, and X. P. Li, “A CMOS continuous-time low-pass notch filter for EEG systems,” Analog Integrated Circuits SignalProcessing, vol. 44, no. 3, pp. 231–238, Sep. 2005.

[66] O. Omeni, E. Rodriguez-Villegas, and C. Toumazou, “A micropowerCMOS continuous-time filter with on-chip automatic tuning,” IEEETrans. Circuits Syst. I: Reg. Papers, vol. 52, no. 4, pp. 695–705, Apr.2005.

[67] C. L. Hsu, M. H. Ho, Y. K. Wu, and T. H. Chen, “Design of low-frequency low-pass filters for biomedical applications,” in Proc. AsiaPacific Conf. IEEE Circuits Systems, Singapore, 2006, pp. 690–695.

[68] S. Y. Lee, C. J. Cheng, C. P. Wang, and W. C. Kao, “Low-power signalprocessing devices for portable ECG detection,” in Proc. 25th Ann. Int.Conf. IEEE EMBS, British Columbia, Canada, 2008, pp. 1683–1686.

[69] J. Silva-Martínez and J. Salcedo-Suñer, “IC voltage-to-current trans-ducers with very-small transconductance,” Analog Inte. Circ. SignalProcessing, vol. 13, no. 3, pp. 285–293, 1997.

[70] P. Garde, “Transconductance cancellation for operational amplifiers,”IEEE J. Solid-State Circuits, vol. 12, no. 3, pp. 310–311, Jun. 1977.

[71] A. Arnaud, R. Fiorell, and C. Galup-Montoro, “Nanowatt, sub-nsOTAs, with sub-10-mV input offset, using series-parallel currentmirrors,” IEEE J. Solid-State Circuits, vol. 41, no. 9, pp. 2009–2018,Set. 2006.

[72] A. Arnaud and C. Galup-Montoro, “Pico-A/V range CMOS transcon-ductors using series-parallel current division,” Electron. Lett., vol. 39,no. 18, pp. 1295–1296, Sep. 2003.

[73] J. Silva-Martinez and A. Vazquez-Gonzalez, “Impedance scalers for ICactive filters,” in Proc. Int. Sym. IEEE Circuits Systems, Monterey, CA,1998, pp. 151–154.

[74] J. Silva-Martinez and S. Solis-Bustos, “Design considerations for highperformance very low frequency filters,” in Proc. Int. Sym. IEEE Cir-cuits Systems, Orlando, FL, 1999, pp. 648–651.

[75] A. K. Y. Wong, K. P. Pun, Y. T. Zhang, and K. Hung, “A near-infraredheart rate measurement IC with very low cutoff frequency using currentsteering technique,” IEEE Trans. Circuits Syst. I, vol. 52, no. 12, pp.2642–2647, Dec. 2005.

[76] C. C. Enz and G. C. Temes, “Circuit techniques for reducing the ef-fects of op-amp imperfections: Autozeroing, correlated double sam-pling, and chopper stabilization,” Proc. IEEE, no. 8, pp. 1584–1614,Aug. 1996.

[77] P. K. Chan, K. A. Ng, and X. L. Zhang, “CMOS chopper-stabilizeddifferential difference amplifier for biomedical integrated circuits,” inProc. 47th Midwest Symp. Circuits Syst., Hiroshima, Japan, 2004, pp.25–28.

[78] B. Gosselin, V. Simard, and M. Sawan, “An ultra low-power chopperstabilized front-end for multichannel cortical signals recording,” inProc. Conf. Canadian Electrical Computer Engineering, 2004, pp.2259–2262.

[79] A. Uranga, X. Navarro, and N. Barniol, “Integrated CMOS amplifierfor ENG signal recording,” IEEE Trans. Biomed. Eng., vol. 51, no. 12,pp. 2188–2194, Dec. 2004.

[80] T. Denison, K. Consoer, W. Santa, A. T. Avestruz, J. Cooley, and A.Kelly, “A 2 uW 100 nV/rtHz chopper-stabilized instrumentation am-plifier for chronic measurement of neural field potentials,” IEEE J.Solid-State Circuits, vol. 42, no. 12, pp. 2934–2945, Dec. 2007.

[81] T. Denison, K. Consoer, A. Kelly, A. Hachenburg, and W. Santa, “A2.2 uW 94 ��

��, chopper-stabilized instrumentation amplifier for

EEG detection in chronic implants,” in Proc. Int. Conf. IEEE Solid-State Circuits, San Francisco, CA, 2007, pp. 162–594.

[82] C. H. Chan, J. Wills, J. LaCoss, J. J. Granacki, and J. Choma, “A micro-power low-noise auto-zeroing CMOS amplifier for cortical neural pros-theses,” in Proc. Conf. IEEE Biomedical Circuits Systems, London,U.K., 2006, pp. 214–217.

[83] M. W. Baker and R. Sarpeshkar, “A low-power high-PSRR current-mode microphone preamplifier,” IEEE J. Solid-State Circuits, vol. 38,no. 10, pp. 1671–1678, Oct. 2003.

[84] K. Lasanen and J. Kostamovaara, “A 1 V analog CMOS front-end fordetecting QRS complexes in a cardiac signal,” IEEE Trans. CircuitsSyst. I: Reg. Papers, vol. 52, no. 12, pp. 2584–2594, Dec. 2005.

[85] A. K. Y. Wong, P. Pun, Y. T. Zhang, and C. S. Choy, “An ECG mea-surement IC using driven-right-leg circuit,” in IEEE Int. Symp. CircuitsSyst., Island of Kos, Greece, 2006, pp. 345–348.

[86] L. Fay, V. Misra, and R. Sarpeshkar, “A micropower electrocardio-gram amplifier,” IEEE Trans. Biomed. Circuits Syst., vol. 3, no. 5, pp.312–320, Oct. 2009.

[87] K. A. Ng and P. K. Chan, “A CMOS analog front-end IC for portableEEG/ECG monitoring applications,” IEEE Trans. Circuits Syst. I: Reg.Papers, vol. 52, no. 11, pp. 2335–2347, Nov. 2005.

[88] R. F. Yazicioglu, P. Merken, R. Puers, and C. V. Hoof, “A 200 uWeight-channel EEG acquisition ASIC for ambulatory EEG systems,”IEEE J. Solid-State Circuits, vol. 43, no. 12, pp. 3025–3038, Dec. 2008.

[89] R. R. Harrison, P. T. Watkins, R. J. Kier, R. O. Lovejoy, D. J. Black,B. Greger, and F. Solzbacher, “A low-power integrated circuit for awireless 100-electrode neural recording system,” IEEE J. Solid-StateCircuits, vol. 42, no. 1, pp. 123–133, Jan. 2007.

[90] A. M. Sodagar, G. E. Perlin, Y. Yao, K. Najafi, and K. D. Wise, “Animplantable 64-channel wireless microsystem for single-unit neuralrecording,” IEEE J. Solid-State Circuits, vol. 44, no. 9, pp. 2591–2604,Sep. 2009.

[91] M. A. Wood and K. A. Ellenbogen, “Cardiac pacemakers from the pa-tient’s perspective,” Circulation, vol. 105, pp. 2136–2138, May 2002.

[92] L. S. Y. Wong, S. Hossain, A. Ta, J. Edvinsson, D. H. Rivas, andH. Naas, “A very low-power CMOS mixed-signal IC for implantablepacemaker applications,” IEEE J. Solid-State Circuits, vol. 39, no. 12,pp. 2446–2456, Dec. 2004.

[93] P. Varady, T. Micsik, S. Benedek, and Z. Benyo, “A novel methodfor the detection of apnea and hypopnea events in respiration signals,”IEEE Trans. Biomed. Eng., vol. 49, no. 9, pp. 936–942, Sep. 2002.

[94] W. Germanovix and C. Toumazou, “Design of a micropower current-mode log-domain analog cochlear implant,” IEEE Trans. Circuits Syst.II: Analog Digital Signal Processing, vol. 47, no. 10, pp. 1023–1046,Oct. 2000.

[95] Kidshealth [Online]. Available: http://kidshealth.org/parent/gen-eral/eyes/cochlear.html

[96] R. Sarpeshkar, C. Salthouse, J. J. Sit, M. W. Baker, S. M. Zhak, T. K. T.Lu, L. Turicchia, and S. Balster, “An ultra-low-power programmableanalog bionic ear processor,” IEEE Trans. Biomed. Eng., vol. 52, no. 4,pp. 711–727, Apr. 2005.

[97] J. Georgiou and C. Toumazou, “A 126-uW cochlear chip for a totallyimplantable system,” IEEE J. Solid-State Circuits, vol. 40, no. 2, pp.430–443, Feb. 2005.

[98] T. G. Constandinou, J. Georgiou, and C. Toumazou, “A fully-integratedsemicircular canal processor for an implantable vestibular prosthesis,”in Proc. 15th Int. Conf. IEEE Electronics, Circuits Syst., St. Julien’s,2008, pp. 81–84.

[99] T. G. Constandinou, J. Georgiou, and C. Toumazou, “A micropowerfront-end interface for differential-capacitive sensor systems,” in Proc.IEEE ISCAS, Seattle, WA, 2008, pp. 2474–2477.

[100] T. G. Constandinou, J. Georgiou, and C. Toumazou, “A partial-cur-rent-steering biphasic stimulation driver for neural prostheses,” in Proc.IEEE ISCAS, Seattle, WA, 2008, pp. 2506–2509.

[101] S. Mandal, L. Turicchia, and R. Sarpeshkar, “A battery-free tag forwireless monitoring of heart sounds,” in Proc. BSN 2009, Berkeley,CA, 2009, pp. 201–206.

[102] Y. S. Yan and Y. T. Zhang, “A model-based artifact reduction methodfor the non-invasive estimation of blood oxygen saturation,” in Proc.Int. Workshop IEEE Computer Architectures for Machine Perception,2003, pp. 90–92.

[103] C. C. Y. Poon, Y. T. Zhang, G. Wong, and W. S. Poon, “Thebeat-to-beat relationship between pulse transit time and systolic bloodpressure,” in Proc. Int. Conf. Technology Applications in Biomedicine,Shenzhen, China, 2008, pp. 342–343.

[104] A. K. Y. Wong, K. P. Pun, Y. T. Zhang, and K. N. Leung, “A low-powerCMOS front-end for photoplethysmographic signal acquisition withrobust dc photocurrent rejection,” IEEE Trans. Biomed. Circuits Syst.,vol. 2, no. 4, pp. 280–288, Dec. 2008.

[105] P. D. Mannheimer, J. R. Cascini, M. E. Fein, and S. L. Nierlich,“Wavelength selection for low-saturation pulse oximetry,” IEEETrans. Biomed. Eng., vol. 44, no. 3, pp. 148–158, Mar. 1997.

[106] C. C. Wang, C. C. Huang, J. S. Liou, Y. J. Ciou, I. Y. Huang, C. P.Li, Y. C. Lee, and W. J. Wu, “A mini-invasive long-term bladder urinepressure measurement ASIC and system,” IEEE Trans. Biomed. Cir-cuits Syst., vol. 2, no. 1, pp. 44–49, Mar. 2008.

[107] P. Bergveld, “Thirty years of isfetology—What happened in the past30 years and what may happen in the next 30 years,” Sens. ActuatorsB-Chem., vol. 88, no. 1, Jan. 2003.

[108] L. Shepherd and C. Toumazou, “Weak inversion ISFETs for ultra-lowpower biochemical sensing and real-time analysis,” Sensors ActuatorsB: Chemical, vol. 107, no. 1, pp. 468–473, May 2005.

[109] P. Georgiou and C. Toumazou, “A silicon pancreatic beta cell for di-abetes,” IEEE Trans. Biomed. Circuits Syst., vol. 1, no. 1, pp. 39–49,Mar. 2007.


Yan Li received the B.E. and M.E. degrees fromLiaoning University, Shenyang, China, in 2004and 2007, respectively. She is currently workingtoward the Ph.D. degree at the Key Laboratory forBiomedical Informatics and Health Engineering,Chinese Academy of Sciences, Shenzhen, China,and SIAT—Institute of Biomedical and Health Engi-neering of Chinese Academy of Sciences, Shenzhen,China.

She is also a Research Assistant of the Divisionof Biomedical Engineering and the Joint Research

Center for Biomedical Engineering at the Chinese University of Hong Kong,Hong Kong. Her current research interests include low power analog integratedcircuit design of medical applications.

Carmen C. Y. Poon (M’08) received the B.A.Sc. de-gree in engineering science (biomedical option) andthe M.A.Sc. degree in biomedical engineering fromthe University of Toronto, ON, Canada, and the Ph.D.degree from The Chinese University of Hong Kong,Hong Kong.

She is currently a Research Assistant Professor atThe Chinese University of Hong Kong. Her researchinterests include biosignal processing, biosystemmodeling, and development of wearable medicaldevices and body sensor network for telemedicine,

m-Health, and p-Health.Dr. Poon is as an Associate Editor of the IEEE TRANSACTIONS ON

INFORMATION TECHNOLOGY IN BIOMEDICINE.

Yuan-Ting Zhang (M’90–SM’93–F’06) receivedthe M.S. degree from Shandong University, Jinan,China, and the Ph.D. degree from the University ofNew Brunswick, Fredericton, NB, Canada, in 1990.

He is currently Head of the Division of Biomed-ical Engineering and Director of the Joint ResearchCenter for Biomedical Engineering at the ChineseUniversity of Hong Kong, Hong Kong. He alsoserves currently as the Director of Key Laboratoryfor Biomedical Informatics and Health Engineeringof Chinese Academy of Sciences and Director of the

SIAT—Institute of Biomedical and Health Engineering of Chinese Academy ofScience. He was a Research Associate and Adjunct Assistant Professor at theUniversity of Calgary, Calgary, AB, Canada, from 1989 to 1994. He chairedthe Biomedical Division of Hong Kong Institution of Engineers in 1996/1997and 2000/2001. His current research interests include neural engineering,health informatics, THz imaging, and wearable medical devices and bodysensor networks particularly for mobile health. He has published more than300 scientific articles in the area of biomedical engineering.

Dr. Zhang was the Technical Program Chair of the 20th Annual InternationalConference in 1998 and the General Conference Chair of the 27th Annual Inter-national Conference in 2005. He served the TPC Chair of IEEE-EMBS SummerSchool and Symposium on Medical Devices and Biosensors (ISSS-MDBS)in 2006 and 2007. He was elected as an IEEE-EMBS AdCom member in1999 and served as Vice-President (Conferences) in 2000. He was an honoraryadvisor of Hong Kong Medical and Healthcare Device Manufacture Associa-tion. He served as Associate Editor for IEEE TRANSACTIONS ON BIOMEDICAL

ENGINEERING and IEEE TRANSACTIONS ON MOBILE COMPUTING. He was alsothe Guest Editor of IEEE Communications Magazine and IEEE TRANSACTIONS

ON INFORMATION TECHNOLOGY IN BIOMEDICINE. He currently serves as theEditor-in-Chief of IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN

BIOMEDICINE and Associate Editor of the Journal of NeuroEngineering andRehabilitation. He is also on a number of editorial boards, the Book Series ofBiomedical Engineering published by the IEEE Press, and the IEEE-EMBSTechnical Committee of Wearable Systems and Sensors. He is a Fellow ofthe International Academy of Medicinal and Biological Engineering and theAmerican Institute for Medical and Biological Engineering.

Documents

Document22