Shafaat A. Bazaz

Design of an electrostatic MEMS microgripper system integrated with force sensor

This work presents the design of a microgripper integrated with force sensor using electrostatic actuation. The single capacitance sensor used in the design can also operate as an actuator when force sensing is not required. The design is optimized in the standard MEMS technology SOIMUMPs. An input voltage of 85 V produces a force of 384 ¿N and a displacement of 15 ¿m. In dual actuation mode, the gripper jaws are displaced 15 ¿m at an input voltage of 60 V. A concept of two stage jaw is given which increases the range of object size gripped to 30 ¿m without increasing the size or input voltage of the actuator.

Automated segmentation of subretinal layers for the detection of macular edema

Macular edema (ME) is considered as one of the major indications of proliferative diabetic retinopathy and it is commonly caused due to diabetes. ME causes retinal swelling due to the accumulation of protein deposits within subretinal layers. Optical coherence tomography (OCT) imaging provides an early detection of ME by showing the cross-sectional view of macular pathology. Many researchers have worked on automated identification of macular edema from fundus images, but this paper proposes a fully automated method for extracting and analyzing subretinal layers from OCT images using coherent tensors. These subretinal layers are then used to predict ME from candidate images using a support vector machine (SVM) classifier. A total of 71 OCT images of 64 patients are collected locally in which 15 persons have ME and 49 persons are healthy. Our proposed system has an overall accuracy of 97.78% in correctly classifying ME patients and healthy persons. We have also tested our proposed implementation on spectral domain OCT (SD-OCT) images of the Duke dataset consisting of 109 images from 10 patients and it correctly classified all healthy and ME images in the dataset.

Thermal Actuation Based 3-DoF Non-Resonant Microgyroscope Using MetalMUMPs.

High force, large displacement and low voltage consumption are a primary concern for microgyroscopes. The chevron-shaped thermal actuators are unique in terms of high force generation combined with the large displacements at a low operating voltage in comparison with traditional electrostatic actuators. A Nickel based 3-DoF micromachined gyroscope comprising 2-DoF drive mode and 1-DoF sense mode oscillator utilizing the chevron-shaped thermal actuators is presented here. Analytical derivations and finite element simulations are carried out to predict the performance of the proposed device using the thermo-physical properties of electroplated nickel. The device sensitivity is improved by utilizing the dynamical amplification of the oscillation in 2-DoF drive mode using an active-passive mass configuration. A comprehensive theoretical description, dynamics and mechanical design considerations of the proposed gyroscopes model are discussed in detail. Parametric optimization of gyroscope, its prototype modeling and fabrication using MetalMUMPs has also been investigated. Dynamic transient simulation results predicted that the sense mass of the proposed device achieved a drive displacement of 4.1μm when a sinusoidal voltage of 0.5V is applied at 1.77 kHz exhibiting a mechanical sensitivity of 1.7μm /°/s in vacuum. The wide bandwidth frequency response of the 2-DoF drive mode oscillator consists of two resonant peaks and a flat region of 2.11 kHz between the peaks defining the operational frequency region. The sense mode resonant frequency can lie anywhere within this region and therefore the amplitude of the response is insensitive to structural parameter variations, enhancing device robustness against such variations. The proposed device has a size of 2.2 × 2.6 mm2, almost one third in comparison with existing M-DoF vibratory gyroscope with an estimated power consumption of 0.26 Watts. These predicted results illustrate that the chevron-shaped thermal actuator has a large voltage-stroke ratio shifting the paradigm in MEMS gyroscope design from the traditional interdigitated comb drive electrostatic actuator. These actuators have low damping compared to electrostatic comb drive actuators which may result in high quality factor microgyroscopes operating at atmospheric pressure.

Performance of a Compliant Structure for a Thermal Resonant Microgyroscope

Realizing a structure that allows integration of actuation and sensing components operating at identical frequency is critical from functional standpoint of resonant microgyroscope. This paper presents the design and experimental test results of a compliant nickel structure for a microgyroscope. The device consists of three suspended proof masses supported by in-plane flexures anchored to the substrate thereby enabling to realize distinct driving and sensing components. Chevron-shaped beams driven electrothermally were employed for actuating drive component while parallel electrostatic combs were employed for sensing the Coriolis force induced rotation. Experimental investigation of the dynamic performance of the actuator and sensor structures revealed that their operating frequencies are very close (within 1%) thereby enabling to realize resonant microgyroscope with increased sensitivity. Comparison of the experimental results with the predictions made by analytical and finite-element models agreed well within 15%. Design guidelines for realizing microgyroscope with improved performance were briefly discussed highlighting the challenges associated with the frequency tuning.

Papilledema Detection in Fundus Images Using Hybrid Feature Set

Papilledema is the swelling of the optic disc. It is a symptom of several diseases including some life-threatening diseases as well. Therefore it needs to be detected as soon as possible. If it is not detected on time and reaches its final stage, it can also lead to permanent vision loss. In this paper, we have given a simple and efficient method to automatically detect papilledema through fundus images. The dataset used is taken from STARE. After preprocessing of the fundus images, 13 features are extracted which are helpful in detecting this disease and combined to form a feature set. After feature extraction, classification is done using SVM classifier on this feature set. The method gives an accuracy of 96.67%.

Comparison of noise removal algorithms on Optical Coherence Tomography (OCT) image

Automatic Identification of disease through image processing in biomedical field is the norm of modern era. Ophthalmologists have used several invasive and noninvasive techniques for early detection of disease. OCT is one such noninvasive modality that performs high resolution tomographic imaging in biological systems. OCT images are produced containing speckle noise. Noise is a major factor that decreases image quality, hence degrading performance of noise image processing algorithms. It is therefore at highest priority to apply effective method for denoising image, before further processing. In this paper we applied Wavelet denoising, bilateral and wiener filter on an OCT images and discuss benefits and drawback of each algorithms. The effectiveness of each algorithm is compared on basis of Signal to noise ratio SNR, peak signal to noise ratio PSNR and Mean square error MSE.

Behavioral modeling of a monolithically integrated three-axis capacitive accelerometer based on the MetalMUMPs process:

In this paper we present behavioral modeling of a monolithically integrated three-axis capacitive accelerometer using the standard MetalMUMPs process. The behavioral modeling is done in order to verify the design with respect to structural and electrostatic performance of the designed three-axis capacitive accelerometer. The proposed accelerometer is 3.2 mm × 3.5 mm in size, designed for sensing the acceleration of 25g in three axes. The mechanical noise floor for in-plane (x and y) and out-of-plane (z) axes is 0.291, 0.316, and 2.84 µg/√ (Hz), respectively. The total sense capacitance along the x, y, and z axes is 68.5 fF, 100 fF, and 6.19 pF, respectively. Sensitivity of 2.568 fF/g, 4 fF/g, and 0.252 pF/g is obtained for in-plane (x and y) and out-of-plane (z) axes, respectively. The resonance frequency for the designed accelerometer is 800 and 2500 Hz for in-plane and out-of-plane axes, respectively. The results obtained by behavioral modeling are compared with the analytical results, which are approximately the same.

Impedance characterization of mems based all diamond neural probe

Microelectrodes are becoming an important tool for sensing of neuro-chemical and neuro-electrical signals for studying neural activities. Although a lot of work has been done for the development of microelectrodes made up of different materials, such as metal micro-wires, ceramic oxide, polymers, carbon fiber microelectrodes and silicon but these are still far from being satisfactory. An ideal material should have excellent biocompatibility, high signal-to-noise ratio and excellent electrochemical properties. The work of this paper focused on the development and impedance characterization of MEMS based All-Diamond neural probe, a new generation of probe with excellent properties such as wider potential window, low back ground current, very high and low electrical resistivities when doped and undoped respectively, mechanical flexibility and biocompatibility. A novel diamond probe with multiple microelectrodes has been presented with the characterization and test results of these developed probes for biomedical application. In this paper, we have characterized diamond probe for neural prosthetic applications. A test cell has been specially designed for In-Vitro testing of microelectrodes on the probe. Impedance, Capacitance and Resistivity characteristics were studied as a function of size of microelectrode, which were found to be 16.273 × 103 ohm, 1.92 to 3.78 ¿F and 28.21 ohm that are in accordance with the theoratical electrochemical equations.

Comparative Study on Finite Element Analysis & System Model Extraction for Non-Resonant 3-DoF Microgyroscope

This paper reports a comparative study of full transient start up analysis of a Non-Resonant Micromachined Gyroscope using Finite Element Analysis (FEA) and System Model Extraction (SME) techniques. FEA is a popular numerical technique based on Finite Element Method (FEM) for carrying out different engineering analyses. But one of the major disadvantages of this FEA is its computational time. While running multiple optimization analyses with a FE Solver, it may take days, weeks and possibly months depending on extent of optimization. In this study we initially analyzed the MEMS based microgyro on a device level using FEM. We determined the natural frequencies of the gyro along with the both static and dynamic responses of the gyro. After these device level simulations in FEA we switched to the system level simulation. Using SME we generated system model of the gyro with system level components and run a full transient startup response analysis of the device by incorporating that extracted model in INTELLISUITE circuit simulator SYNPLE. When we compared the computational time required by both techniques, we found that SME with SYNPLE is orders of magnitude faster than FEA.

Real-time doctor-patient assignment in a telemedicine system

The aim of this paper is to address and overcome the limitation in number of doctors by assigning patients efficiently in Telemedicine System. In this paper, we present mathematical formulation and an algorithm to address the problem of effective and resourceful mapping between patients and doctors for online Telemedicine System so that overall patient wait time can be minimized. Our proposed algorithm is based on providing efficient allocation of patients to doctors considering two primary factors i.e. patient handling time and workload (queue size) of doctor. We compare our algorithm with baseline algorithm (Round Robin) and calculate percentile average difference of the mean average time of both with constraints of limited, moderate and maximum availability of resources under the observation in variation of patient handling time. Our simulation based evaluation shows increase in efficiency from 0.0% in worst case to 110.84% in best case when compared with Round Robin. Therefore our proposed algorithm can help to improve doctors efficiency and minimize patient wait time. The improvement in the assigning patients to doctors in efficient way yields to effective resource utilization of scarce health resources.