ICSE 2008 Proc. 2008, Johor Bahru, Malaysia
Design of MEMS Biomedical Pressure Sensor for Gait Analysis Yufridin Wahab1, Member, IEEE, Aladin Zayegh1, Member, IEEE, Rezaul K. Begg2, Member, IEEE, and Ronny Veljanovski1, Member, IEEE, 1 Faculty of Health, Engineering and Science 2 Biomechanics Unit, CARES/HMRP Victoria University, PO Box 14428, Melbourne, VIC 8001, AUSTRALIA Email:
[email protected]
Abstract – Measurement of the foot and shoe interface pressure underpins a number of important applications. Abnormal pressure may indicate instability in gait, risks of diabetic ulceration and many other biomedical and sports applications. As the current foot pressure sensors in the market exhibit many limitations, a new sensor design based on the more promising MEMS technology was therefore explored. As such, this paper reports the analysis and optimization of a MEMS pressure sensor for foot pressure measurement. The pressure sensor had a high linearity output with pressure span of more than 2-MPa. This characteristic indicates excellent potential for a wide spectrum of biomechanical activities. I. INTRODUCTION THE interface pressure between foot plantar surface and shoe soles is one of several key parameters frequently measured in biomechanical field. This parameter has wide applications, for example, screening for high risk diabetic foot ulceration, design of orthotics for diabetes mellitus and peripheral neuropathy, footwear design [1], improvement of balance [2], sports injury prevention in athletes [3] plus many more. For example, foot ulceration is estimated to cause over $1 billion per year worth of medical expenses in the United States alone [4]. It is therefore critical to ensure the availability of an accurate and efficient technique of measuring this type of pressure. Among the key requirements of the pressure sensor specification for the application is its pressure range, output linearity and low hysteresis [5]. It is suggested that for walking analysis, pressure sensor should be able to measure up to 1000kPa, or 10kgcm-2 (about 100Ncm-2) [6]. For sports application, the range should be even more. There are a number
166
1-4244-2561-7/08/$20.00 ©2008 IEEE
of commercially available foot-pressure sensors at current but due to specification limitations, these sensors are not fully fulfilling the requirement of various biomechanical applications. The limitations include the specified pressure span, physical sensor dimensions [7] and hysteresis [5]. These weaknesses may cause erroneous readings or even worse, the malfunctioning of the device as some application may introduce pressure values way too far beyond the specified range. To provide a better solution to the needs, we, therefore, explore the development a microsensor uniquely designed for foot pressure measurement using a proven micro-electromechanical system (MEMS) technology. The sensing membrane and sensing circuit designs are modelled and analyzed. A significant performance enhancement is achieved recently following an extensive optimization works. In this paper, the focus is directed towards total performance optimizations of the sensor output signal. It is divided into five sections. Section II briefly explains the device design, followed by a section each for optimization works, results and discussion and the concluding remarks. II. DEVICE DESIGN MEMS may advance the specification of foot pressure measurement devices in many ways. The highly miniaturized structures exhibit no mechanical hysteresis [8]. Moreover, as proven in [5], silicon MEMS pressure sensor package design for biomechanical application is technologically feasible. The study involves modelling of the sensor design, finite-element analysis, simulations and fabrication. The device is based on the Infineon SensoNor [9] technology with membrane area of 100µmX100µm and thickness of 23.1µm. It can measure up to more than 2MPa of pressure with highly linear output.
ICSE 2008 Proc. 2008, Johor Bahru, Malaysia
III. DESIGN OPTIMIZATION A series of optimization are performed on a reported design of a silicon micromachined pressure sensor [10][11]. Continuous improvements are required to further increase the output voltage level of the sensor. The design optimization only involves the sensing elements, namely the piezoresistors. Fig. 1 shows the design parameters available as part of the design optimization space involved in the whole study. Piezoresistors’ locations length width
Piezoresistors in the membrane layer
Membrane width, length and thickness
Fig. 1 The parameters optimized in the whole study.
The design, computer modelling and optimization studies are performed using the CoventorWare [12] design tool. The Wheatstone bridge aligned four sensing piezoresistors are involved in this optimization study. The optimization begins with the effect of sensing resistor size. In this step, the width and length of the sensing resistors are swept across all possible values. The values are limited by the process design rules [9] and the physical dimension of the sensor. The next analysis was the influence of device location on the maximum output voltage magnitude. The literature pointed that the regions of high stress were the best options for sensing resistor placement as one of the main objective is always to obtain the maximum voltage at the output. For this reason, the locations of resistors are varied according to the parameter y offset and x offset (or Offset_y and Offset_x, respectively). These offsets represent the distance between the selected location from the membrane edge in the x or y direction. Again, the limiting factors that have to be considered are the sensor dimension and process design rules. Based on both the element dimension and location analysis results, the final optimization is performed. At this stage, only the significant factors that are affecting the output magnitude were considered. The parameters are swept
across all possible values allowed by the device dimension and process design rules. The optimization work reported here is highly needed due to the recent technical changes occurred to the process representation [13]. IV. RESULTS AND DISCUSSION Initial design is reported in [10]. However, no optimization work is properly done. The first optimization study is explained in [11]. A more extensive study is performed recently with an aim to further strengthen the output values. The result of the analysis of the placement optimization is shown in Fig. 2. In this analysis, the Offset_y is varied from 0µm to 33µm, while the Offset_x is swept from -10µm to 10µm. The maximum voltage is achieved when Offset_y is 3µm and Offset_x of around 2µm, with value of about 0.57V. Further optimization is performed to find an even better location following the indications from the result in Fig. 2. The result of the extended analysis is given in Fig. 3. From the final placement optimization results, it is found that the better locations for the piezoresistors are that with Offset_y and Offset_x of 4.5µm and 2µm respectively. The value of the output voltage at these locations is very close to 0.6V. This is about 0.3V improvement over the depicted values in Fig. 2. As Fig. 2 indicates the potential of getting higher output voltage if the Offset_y is further analyzed, the performed study proves its validity. In terms of piezoresistor dimensions, Fig. 4 and Fig. 5 depict the results. As can be seen in Fig. 4, similar to the reported result in [11], the best value for piezoresistor width is 4µm. This is obviously the minimum size allowed by the fabrication process [9]. On the other hand, the output voltage is better optimized when the piezoresistor length is chosen to be 15.5µm, as shown in Fig. 5. The maximum value for the voltage as recorded at this stage is about 1.12V, which is about double the achievement shown in Fig. 3. This new result shows that the output voltage can be optimized very efficiently using a step by step analysis and optimization. Using the optimized locations and dimension, the relationship of the output voltage and the applied pressure is obtained. Fig. 6 shows the optimized relationship of the output voltage and pressure when the applied pressure is varied from 0Pa to 2MPa on the membrane. The result of Fig. 5 is hereby verified. The sensor output is increased 100 times by use of operational amplifier circuit. 167
Output voltage (V)
Output voltage (V)
ICSE 2008 Proc. 2008, Johor Bahru, Malaysia
Piezoresistor length (µm)
Offset_x (µm)
Fig. 5 Output voltage against piezoresistor length.
Output voltage (V)
Output voltage (V)
Fig. 2 Output voltage against Offset_x.
Applied Pressure (MPa) Offset_x (m)
Fig. 3 Output voltage against Offset_x.
Fig. 6 The Output voltage vs Pressure
V. CONCLUSION
Output voltage (V)
The optimization of a pressure sensor for biomedical application is described. The performance improvement is increased significantly to enable better representation of foot plantar pressure. The high pressure range enables the application of this sensor encompassing wide spectrum of biomechanical activities and clinical diagnosis. The improvement is achieved by thorough analysis of the sensing piezoresistors’ placement and sizing. Piezoresistor length (µm)
Fig. 4 Output voltage against piezoresistor length.
The dashed line rectangle is the key finding of this analysis. It is enlarged and reproduced as Fig. 5. It shows the maximum values of output voltage when the piezoresistors’ locations and dimensions are highly optimized.
168
ACKNOWLEDGMENT The authors wish to thank Dr Daniel Lapadatu of Infineon Technologies SensoNor, Dr Gerold Schröpfer and Aurelie Cruau of Coventor Inc. for their technical support. This work was supported in part by grants from Victoria University, Melbourne and University Malaysia Perlis (Unimap), Perlis, Malaysia.
ICSE 2008 Proc. 2008, Johor Bahru, Malaysia
REFERENCES [1]
[2]
[3]
[4]
[5]
[6] [7]
[8]
[9]
M. J. Mueller, "Application of plantar pressure assessment in footwear and insert design," Journal of Orthopaedics and. Sports Physical Therapy, vol. 29, pp. 747-756, 1999. E. Santarmou, M. Dozza, M. Lannocca, L. Chiari, and A. Cappello, "Insole pressure sensor-based audio-biofeedback for balance improvement," Gait & Posture, vol. 24, pp. S30-S31, 2006. A. Gefen, "Biomechanical analysis of fatigue-related foot injury mechanisms in athletes and recruits during intensive marching," Medical and Biological Engineering and Computing, vol. 40, pp. 302-310, 2002. J. R. Mackey and B. L. Davis, "Simultaneous shear and pressure sensor array for accessing pressure and shear at foot/ground interface,"Journal of Biomechanics, vol. 39, pp. 2893-2897, 2006. N. K. S. Lee, R. S. Goonetilleke, Y. S. Cheung, and G. M. Y. So, "A flexible encapsulated MEMS pressure sensor system for biomechanical applications," Microsystem Technologies, vol. 7, pp. 55-62, 2001 S. Urry, "Plantar pressure-measurement sensors," Measurement Science and Technology, vol. 10, pp. R16-R32, 1999. M. W. Carol, M. G. Sarah and J. D. Sharon,"The shock attenuation characteristics of four different insoles when worn in a military boot during running and marching,"Gait &Posture,vol. 9(1), pp.31-37, 1999. J. Bryzek, S. Roundy, B. Bircumshaw, C. A. C. C. Chung, K. A. C. K. Castellino, J. R. A. S. J. R. Stetter, and M. A. V. M. Vestel, "Marvelous MEMS," Circuits and Devices Magazine, IEEE, vol. 22, pp. 8-28, 2006. Infineon Technologies SensoNor, MultiMEMS Design Handbook Version 4.1, Horten, Norway: Infineon Technologies SensoNor, 2007.
[10] Y. Wahab, A. Zayegh, R. Begg and R. Veljanovski, “Micro-sensor for foot pressure measurement,” IEEE TENCON, accepted for presentation. [11] Y. Wahab, A. Zayegh, R. Begg and R. Veljanovski, “Sensitivity optimization of a foot plantar pressure micro-sensor,” IEEE International Conference on Microelectronics, accepted for presentation. [12]
[13]
G. Schröpfer and A. Cruau, microBUILDER design Handbook Module VII, Process Access Kits in CoventorWare 2008.000 for MultiMEMS and Tronics SOI HARM, Version 1.0, February 2008,http://www.microBUILDER.org G. Schröpfer and A. Cruau, (personal communication), Oct. 10, 2008.
169
