System-Identification-Based Activity Recognition Algorithms With Inertial Sensors.

This paper focuses on activity recognition using a single wearable inertial measurement sensor placed on the subject's chest. The ten activities that need to be identified include lying down, standing, sitting, bending and walking, among others. The activity recognition approach is based on using and identifying a transfer function associated with each activity. The appropriate input and output signals for each transfer function are first determined based on the norms of the sensor signals excited by that specific activity. Then the transfer function is identified using training data and a Wiener filter based on the auto-correlation and cross-correlation of the output and input signals. The activity occurring in real-time is recognized by computing and comparing the input-output errors associated with all the transfer functions. The performance of the developed system is evaluated using data from a group of Parkinson's disease subjects, including data obtained in a clinical setting and data obtained through remote home monitoring. On average, the developed system provides better than 90% accuracy in identifying each activity as it occurs. Activity recognition is particularly useful for PD patients in order to monitor their level of activity, characterize their postural instability and recognize high risk-activities in real-time that could lead to falls.

Real world validation of activity recognition algorithm and development of novel behavioral biomarkers of falls in aged control and movement disorder patients.

The use of wearable sensors in movement disorder patients such as Parkinson's disease (PD) and normal pressure hydrocephalus (NPH) is becoming more widespread, but most studies are limited to characterizing general aspects of mobility using smartphones. There is a need to accurately identify specific activities at home in order to properly evaluate gait and balance at home, where most falls occur. We developed an activity recognition algorithm to classify multiple daily living activities including high fall risk activities such as sit to stand transfers, turns and near-falls using data from 5 inertial sensors placed on the chest, upper-legs and lower-legs of the subjects. The algorithm is then verified with ground truth by collecting video footage of our patients wearing the sensors at home. Our activity recognition algorithm showed >95% sensitivity in detection of activities. Extracted features from our home monitoring system showed significantly better correlation (~69%) with prospectively measured fall frequency of our subjects compared to the standard clinical tests (~30%) or other quantitative gait metrics used in past studies when attempting to predict future falls over 1 year of prospective follow-up. Although detecting near-falls at home is difficult, our proposed model suggests that near-fall frequency is the most predictive criterion in fall detection through correlation analysis and fitting regression models.

Toward Completely Sampled Extracellular Neural Recording During fMRI.

This work aims to estimate severe fMRI scanning artifacts in extracellular neural recordings made at ultrahigh magnetic field strengths in order to remove the artifact interferences and uncover the complete neural electrophysiology signal. We build on previous work that used PCA to denoise EEG recorded during fMRI, adapting it to cover the much larger frequency range (1-6000 Hz) of the extracellular field potentials (EFPs) observed by extracellular neural recordings. We examine the singular value decomposition (SVD)-PCA singular value shrinkage (SVS) and compare two shrinkage rules and a sliding template subtraction approach. Additionally, we present a new technique for estimating the singular value upper bounds in spontaneous neural activity recorded in the isoflurane anesthetized rat that uses the temporal first difference of the neural signal. The approaches are tested on artificial datasets to examine their efficacy in detecting extracellular action potentials (EAPs: 300-6000 Hz) recorded during fMRI gradient interferences. Our results indicate that it is possible to uncover the EAPs recorded during gradient interferences. The methods are then tested on natural (non-artificial) datasets recorded from the cortex of isoflurane anesthetized rats, where both local field potential (LFP: 1-300 Hz) and EAP signals are analyzed. The SVS methods are shown to be advantageous compared to sliding template subtraction, especially in the high frequency range corresponding to EAPs. Our novel approach moves us towards simultaneous fMRI and completely sampled neural recording (1-6000Hz with no temporal gaps), providing the opportunity for further study of spontaneous brain function and neurovascular coupling at ultrahigh field in the isoflurane anesthetized rat.

Reference-Free Adaptive Filtering of Extracellular Neural Signals Recording in Ultra-High Field Magnetic Resonance Imaging Scanners: Removal of Periodic Interferences.

This paper focuses on the removal of periodic artifacts from neural signals recorded in rats in ultra-high field (UHF) MRI scanners, using a reference free adaptive feedforward method. Recording extracellular neural signals in the UHF environment is motivated by the desire to combine neural recording and UHF functional magnetic resonance imaging (fMRI) to better understand brain function. However, the neural signals are found to have extremely high noise artifacts of a periodic nature due to electromagnetic interference and due to small oscillatory motions. In particular, noise at 60 Hz and several harmonics of 60 Hz, sinusoidal noise from a pump, and low frequency breathing motion artifacts are observed. Due to significant overlap between the noise frequencies and the neural frequency region of interest, band pass filters cannot be effectively utilized in this application. Hence, this paper develops adaptive least squares feedforward cancellation filters to remove the periodic artifacts. The interference fundamental frequency is identified precisely using an implementation of k-means in an iterative approach. The paper includes significant animal data from rats recorded in an IACUC-approved procedure in 9.4T and 16.4T MRI machines. For breathing artifacts filtered from 4 rats, the mean signal cancellation values at the harmonic interference frequencies are 5.18, 12.97, and 20.87 dB/Hz for a sliding template subtraction, a single-stage impulse reference method, and the cascaded adaptive filtering approach respectively. For pump artifacts filtered from 2 chronically implanted rats, mean signal cancellation values are 2.85, 9.52 and 12.06 dB/Hz respectively. The experimental results show that periodic noise is very effectively removed by the developed cascaded adaptive least squares feedforward algorithm.

Novel Composite Gold-Aluminum Electrode with Application to Neural Recording and Stimulation in Ultrahigh Field Magnetic Resonance Imaging Scanners.

Traditional electrodes used for neural recording and stimulation generate large regions of signal void (no functional MRI signal) when used in ultrahigh field (UHF) MRI scanners. This is a significant disadvantage when simultaneous neural recording/stimulation and fMRI signal acquisition is desired, for example in understanding the functional mechanisms of deep brain stimulation (DBS). In this work, a novel gold-aluminum microwire neural electrode is presented which overcomes this disadvantage. The gold-aluminum design greatly reduces the magnetic susceptibility difference between the electrode and brain tissue leading to significantly reduced regions of signal void. Gold-aluminum microwire samples are imaged at ultrahigh field 16.4 Tesla and compared with gold-only and aluminum-only microwire samples. First, B0 field mapping was used to quantify field distortions at 16.4T and compared with analytical computations in an agarose phantom. The gold-aluminum microwire samples generated substantially less field distortion and signal loss in comparison with gold-only and aluminum-only samples at 16.4T using gradient echo imaging and echo planar imaging sequences. Next, the proposed gold-aluminum electrode was used to successfully record local field potential signals from a rat cortex. The newly proposed gold-aluminum microwire electrode exhibits reduced field distortions and signal loss at 16.4T, a finding which translates to MRI scanners of lower magnetic field strengths as well. The design can be easily reproduced for widespread study of DBS using MRI in animal models. Additionally, the use of non-reactive gold and aluminum materials presents an avenue for translation to human implant applications in the future.

Vibrotactile perception in Dupuytren disease.

PURPOSE: Dupuytren disease (DD) has been associated with enlarged Pacinian corpuscles (PCs) and with PCs having a greater number of lamellae. Based on these associations, we hypothesized that subjects with DD would have altered sensitivity to high-frequency vibrations and that the changes would be more prominent at 250 Hz, where healthy subjects demonstrate the highest sensitivity. METHODS: A novel device was created to deliver vibrations of specific frequencies and amplitudes to the fingers and palm. Using a Psi-marginal adaptive algorithm, vibrotactile perception thresholds (VPTs) were determined in 36 subjects with DD and 74 subjects without DD. Experiments were performed at 250 Hz and 500 Hz at the fingertip and palm. The VPTs were statistically analyzed with respect to disease status, age, gender, location tested, and frequency tested. RESULTS: We found that VPT increases with age, which agrees with findings by others. Women showed greater sensitivity (i.e. lower VPT) than men. Men exhibited lower sensitivity in DD versus healthy subjects, but the results were not statistically significant. In subjects with DD presenting unilaterally, the unaffected hand was more sensitive than the affected hand, in particular for a 250 Hz stimulus applied to the finger. CONCLUSIONS: The data on vibration sensitivity obtained from a large group of subjects with and without DD present interesting trends that may serve as a useful reference to future DD researchers. Understanding additional symptoms of DD may facilitate development of novel diagnostic or prognostic protocols.

An Instrumented Urethral Catheter with a Distributed Array of Iontronic Force Sensors.

This paper develops a novel instrumented urethral catheter with an array of force sensors for measuring the distributed pressure in a human urethra. The catheter and integrated portions of the force sensors are fabricated by the use of 3D printing using a combination of both soft and hard polymer substrates. Other portions of the force sensors consisting of electrodes and electrolytes are fabricated separately and assembled on top of the 3D-printed catheter to create a soft flexible device. The force sensors use a novel supercapacitive (iontronic) sensing mechanism in which the contact area between a pair of electrodes and a paper-based electrolyte changes in response to force. This provides a highly sensitive measure of force that is immune to parasitic noise from liquids. The developed catheter is tested using a force calibration test rig, a cuff-based pressure application device, an extracted bladder and urethra from a sheep and by dipping inside a beaker of water. The force sensors are found to have a sensitivity of 30-50 nF/N, which is 1000 times larger than that of traditional capacitive force sensors. They exhibit negligible capacitance change when dipped completely in water. The pressure cuff tests and the extracted sheep tissue tests also verify the ability of the sensor array to work reliably in providing distributed force measurements. The developed catheter could help diagnose ailments related to urinary incontinence and inadequate urethral closure pressure.

Adaptive virtual referencing for the extraction of extracellularly recorded action potentials in noisy environments.

OBJECTIVE: Removal of common mode noise and artifacts from extracellularly measured action potentials, herein referred to as spikes, recorded with multi-electrode arrays (MEAs) which included severe noise and artifacts generated by an ultrahigh field (UHF) 16.4 Tesla magnetic resonance imaging (MRI) scanner. APPROACH: An adaptive virtual referencing (AVR) algorithm is used to remove artifacts and thus enable extraction of neural spike signals from extracellular recordings in anesthetized rat brains. A 16-channel MEA with 150-micron inter-site spacing is used, and a virtual reference is created by spatially averaging the 16 signal channels which results in a reference signal without extracellular spiking activity while preserving common mode noise and artifacts. This virtual reference signal is then used as the input to an adaptive FIR filter which optimally scales and time-shifts the reference to each specific electrode site to remove the artifacts and noise. MAIN RESULTS: By removing artifacts and reducing noise, the neural spikes at each electrode site can be well extracted, even from data originally recorded with a high noise floor due to electromagnetic interference and artifacts generated by a 16.4T MRI scanner. The AVR method enables many more spikes to be detected than would otherwise be possible. Further, the filtered spike waveforms can be well separated from each other using PCA feature extraction and semi-supervised k-means clustering. While data in a 16.4T MRI scanner contains significantly more noise and artifacts, the developed AVR method enables similar data quality to be extracted as recorded on benchtop experiments outside the MRI scanner. SIGNIFICANCE: AVR of extracellular spike signals recorded with MEAs has not been previously reported and fills a technical need by enabling low-noise extracellular spike extraction in noisy and challenging environments such as UHF MRI that will enable further study of neuro-vascular coupling at UHF.

Observer-Based Deconvolution of Deterministic Input in Coprime Multichannel Systems With Its Application to Noninvasive Central Blood Pressure Monitoring.

Estimating central aortic blood pressure (BP) is important for cardiovascular (CV) health and risk prediction purposes. CV system is a multichannel dynamical system that yields multiple BPs at various body sites in response to central aortic BP. This paper concerns the development and analysis of an observer-based approach to deconvolution of unknown input in a class of coprime multichannel systems applicable to noninvasive estimation of central aortic BP. A multichannel system yields multiple outputs in response to a common input. Hence, the relationship between any pair of two outputs constitutes a hypothetical input-output system with unknown input embedded as a state. The central idea underlying our approach is to derive the unknown input by designing an observer for the hypothetical input-output system. In this paper, we developed an unknown input observer (UIO) for input deconvolution in coprime multichannel systems. We provided a universal design algorithm as well as meaningful physical insights and inherent performance limitations associated with the algorithm. The validity and potential of our approach were illustrated using a case study of estimating central aortic BP waveform from two noninvasively acquired peripheral arterial pulse waveforms. The UIO could reduce the root-mean-squared error (RMSE) associated with the central aortic BP by up to 27.5% and 28.8% against conventional inverse filtering (IF) and peripheral arterial pulse scaling techniques.

Paper-Based Supercapacitive Mechanical Sensors.

Paper has been pursued as an interesting substrate material for sensors in applications such as microfluidics, bio-sensing of analytes and printed microelectronics. It offers advantages of being inexpensive, lightweight, environmentally friendly and easy to use. However, currently available paper-based mechanical sensors suffer from inadequate range and accuracy. Here, using the principle of supercapacitive sensing, we fabricate force sensors from paper with ultra-high sensitivity and unprecedented configurability. The high sensitivity comes from the sensitive dependence of a supercapacitor's response on the contact area between a deformable electrolyte and a pair of electrodes. As a key component, we develop highly deformable electrolytes by coating ionic gel on paper substrates which can be cut and shaped into complex three-dimensional geometries. Paper dissolves in the ionic gel after determining the shape of the electrolytes, leaving behind transparent electrolytes with micro-structured fissures responsible for their high deformability. Exploiting this simple paper-based fabrication process, we construct diverse sensors of different configurations that can measure not just force but also its normal and shear components. The new sensors have range and sensitivity several orders of magnitude higher than traditional MEMS capacitive sensors, in spite of their being easily fabricated from paper with no cleanroom facilities.

Wearable Water Content Sensor Based on Ultrasound and Magnetic Sensing.

Fluid accumulation in the lower extremities is an early indicator of disease deterioration in cardiac failure, chronic venous insufficiency and lymphedema. At-home wearable monitoring and early detection of fluid accumulation can potentially lead to prompt medical intervention and avoidance of hospitalization. Current methods of fluid accumulation monitoring either suffer from lack of specificity and sensitivity or are invasive and cost-prohibitive to use on a daily basis. Ultrasound velocity in animal and human tissue has been found to change with water content. However, previous prototype fluid monitoring sensors based on ultrasound are cumbersome and not wearable. Hence, in this research a compact water content sensor based on a wearable instrumented elastic band is proposed. A novel integration of magnetic sensing and ultrasonic sensing is utilized, where the magnetic sensor provides distance measurement and the ultrasonic sensor produces time-of-flight measurement. Magnetic field modeling with a Kalman filter and least squares linear fitting algorithms are employed to ensure robust sensor performance on a wearable device. The combination of the two measurements yields ultrasound velocity measurement in tissue. The water content sensor prototype was tested on a tissue phantom, on animal tissue and on a human leg. The error in velocity measurement is shown to be small enough for early detection of tissue edema.

Computation of Magnetic Field Distortions and Impact on T2*-weighted MRI with Application to Magnetic Susceptibility Parameter Estimation.

A two-step numerical computation of T2* signal weighting maps in gradient echo magnetic resonance imaging in the presence of an object with varied susceptibility property is presented. In the first step, the magnetic scalar potential is computed for an arbitrary 2D magnetic susceptibility distribution using an algebraic solver. The corresponding magnetic field disturbance is computed from the magnetic scalar potential. In the second step, nonlinear operations are used to compute T2* from the magnetic field disturbance and then to generate a map of T2* signal weighting. The linearity of the first step of the solution process is used to implement a superposition of basis solutions approach that increases computational efficiency. Superposition of basis solutions, computed from a system composed of a single node of differing magnetic susceptibility from the surround, herein referred to as the base system, is found to provide an accurate estimation of the scalar potential for arbitrary susceptibility distributions. Afterwards, nonlinear computation of the T2* signal weighting maps can be performed. The properties of the algebraic magnetic scalar potential solver are discussed in this work. Finally, the linearity of the magnetic scalar potential solver is used to estimate the magnetic susceptibility of various objects from in vitro MR-imaging data acquired at 9.4T.

Instrumented Urethral Catheter and Its Ex Vivo Validation in a Sheep Urethra.

This paper designs and fabricates an instrumented catheter for instantaneous measurement of distributed urethral pressure profiles. Since the catheter enables a new type of urological measurement, a process for accurate ex-vivo validation of the catheter is developed. A flexible sensor strip is first fabricated with nine pressure sensors and integrated electronic pads for an associated sensor IC chip. The flexible sensor strip and associated IC chip are assembled on a 7 Fr Foley catheter. A sheep bladder and urethra are extracted and used in an ex-vivo set up for verification of the developed instrumented catheter. The bladder-urethra are suspended in a test rig and pressure cuffs placed to apply known static and dynamic pressures around the urethra. A significant challenge in the performance of the sensor system is the presence of parasitics that introduce large bias and drift errors in the capacitive sensor signals. An algorithm based on use of reference parasitic transducers is used to compensate for the parasitics. Extensive experimental results verify that the developed compensation method works effectively. Results on pressure variation profiles circumferentially around the urethra and longitudinally along the urethra are presented. The developed instrumented catheter will be useful in improved urodynamics to more accurately diagnose the source of urinary incontinence in patients.

Carbon Nano-Structured Neural Probes Show Promise for Magnetic Resonance Imaging Applications.

OBJECTIVE: Previous animal studies have demonstrated that carbon nanotube (CNT) electrodes provide several advantages of preferential cell growth and better signal-to-noise ratio when interfacing with brain neural tissue. This work explores another advantage of CNT electrodes, namely their MRI compatibility. MRI-compatible neural electrodes that do not produce image artifacts will allow simultaneous co-located functional MRI and neural signal recordings, which will help improve our understanding of the brain. APPROACH: Prototype CNT electrodes on polyimide substrates are fabricated and tested in vitro and in vivo in rat brain at 9.4T. To understand the results of the in vitro and in vivo studies, a simulation model based on numerical computation of the magnetic field around a two-dimensional object in a tissue substrate is developed. MAIN RESULTS: The prototype electrodes are found to introduce negligible image artifacts in structural and functional imaging sequences in vitro and in vivo. Simulation results confirm that CNT prototype electrodes produce less magnetic field distortion than traditional metallic electrodes due to a combination of both superior material properties and geometry. By using CNT films, image artifacts can be nearly eliminated at magnetic fields of strength up to 9.4T. At the same time, the high surface area of a CNT film provides high charge transfer and enables neural local field potential (LFP) recordings with an equal or better signal-to-noise ratio (SNR) than traditional electrodes. SIGNIFICANCE: CNT film electrodes can be used for simultaneous MRI and electrophysiology in animal models to investigate fundamental neuroscience questions and clinically relevant topics such as epilepsy.

Novel Supercapacitor-Based Force Sensor Insensitive to Parasitic Noise.

Traditional capacitive sensors suffer from significant parasitic noise when used in liquid environments or inside the human body. The parasitic noise overwhelms the force response of the sensor and makes it impossible to calculate the absolute force experienced by the sensor. This article focuses on the development of a supercapacitor based force sensor that is immune to parasitic noise. The supercapacitor consists of co-planar electrodes and a solid state ionic gel electrolyte on a deformable membrane. Force exertion causes deformation of the electrolyte membrane, increases its area of contact with the electrodes, resulting in a change of capacitance. The sensor is sealed, waterproof, and shows absolutely no changes in capacitance when immersed in water or enclosed in extracted sheep tissue. At the same time, its force sensitivity of 0.13 µ F/N exceeds the 0.3 pF/N sensitivity of a traditional capacitive sensor by 6 orders of magnitude. The developed sensor could have many biomedical applications in which parasitic capacitance is a serious challenge.

Transparent Flexible Active Faraday Cage Enables In Vivo Capacitance Measurement in Assembled Microsensor.

Capacitive micro-sensors such as accelerometers, gyroscopes and pressure sensors are increasingly used in the modern electronic world. However, the in vivo use of capacitive sensing for measurement of pressure or other variables inside a human body suffers from significant errors due to stray capacitance. This paper proposes a solution consisting of a transparent thin flexible Faraday cage that surrounds the sensor. By supplying the active sensing voltage simultaneously to the deformable electrode of the capacitive sensor and to the Faraday cage, the stray capacitance during in vivo measurements can be largely eliminated. Due to the transparency of the Faraday cage, the top and bottom portions of a capacitive sensor can be accurately aligned and assembled together. Experimental results presented in the paper show that stray capacitance is reduced by a factor of 10 by the Faraday cage, when the sensor is subjected to a full immersion in water.

Note: Development of leg size sensors for fluid accumulation monitoring.

A number of diseases can lead to fluid accumulation and swelling in the lower leg. Early detection of leg swelling can be used to effectively predict potential health risks and allows for early intervention from medical providers. Hence this note develops a novel leg size sensor based on the use of magnetic field measurement. An electromagnet is combined with two magnetic field transducers to provide a drift-free leg size estimation technique immune to environmental disturbances. The sensor can measure changes as small as 1 mm in diameter reliably during in vitro tests. Its performance is compared with that of other size measurement techniques.

Sensors on instrumented socks for detection of lower leg edema--An in vitro study.

This paper presents the design, sensing principles and in vitro evaluation of a novel instrumented sock intended for prediction and prevention of acute decompensated heart failure. The sock contains a drift-free ankle size sensor and a leg tissue elasticity sensor. Both sensors are inexpensive and developed using innovative new sensing ideas. Preliminary tests with the sensor prototypes show promising results: The ankle size sensor is capable of measuring 1 mm changes in ankle diameter and the tissue elasticity sensor can detect 0.15 MPa differences in elasticity. A low-profile instrumented sock prototype with these two sensors has been successfully fabricated and will be evaluated in the future in an IRB-approved human study.

Distributed pressure sensors for a urethral catheter.

A flexible strip that incorporates multiple pressure sensors and is capable of being fixed to a urethral catheter is developed. The urethral catheter thus instrumented will be useful for measurement of pressure in a human urethra during urodynamic testing in a clinic. This would help diagnose the causes of urinary incontinence in patients. Capacitive pressure sensors are fabricated on a flexible polyimide-copper substrate using surface micromachining processes and alignment/assembly of the top and bottom portions of the sensor strip. The developed sensor strip is experimentally evaluated in an in vitro test rig using a pressure chamber. The sensor strip is shown to have adequate sensitivity and repeatability. While the calibration factors for the sensors on the strip vary from one sensor to another, even the least sensitive sensor has a resolution better than 0.1 psi.

Flexible Distributed Pressure Sensing Strip for a Urethral Catheter.

A multi-sensor flexible strip is developed for a urethral catheter to measure distributed pressure in a human urethra. The developed sensor strip has important clinical applications in urodynamic testing for analyzing the causes of urinary incontinence in patients. There are two major challenges in the development of the sensor. First, a highly sensitive sensor strip that is flexible enough for urethral insertion into a human body is required and second, the sensor has to work reliably in a liquid in-vivo environment in the human body. Capacitive force sensors are designed and micro-fabricated using polyimide/PDMS substrates and copper electrodes. To remove the parasitic influence of urethral tissues which create fringe capacitance that can lead to significant errors, a reference fringe capacitance measurement sensor is incorporated on the strip. The sensing strip is embedded on a catheter and experimental in-vitro evaluation is presented using a bench-top pressure chamber. The sensors on the strip are able to provide the required sensitivity and range. Preliminary experimental results also show promise that by using measurements from the reference parasitic sensor on the strip, the influence of parasitics from human tissue on the pressure measurements can be removed.