arXiv subjects

Rikky Muller

Publications and source records attributed to Rikky Muller.

At least 19 recordsLinked to original sources

arXiv2025

DustNet: A Wireless Network of Ultrasonic Neural Implants

Spatially distributed peripheral nerve recordings can be used to reconstruct motor intention and improve natural control of prosthetics However, many existing clinical solutions rely on percutaneous wires to access peripheral nerves; these sites are prone to infection and motion-induced electrode degradation, preventing chronic use. To address the need for fully wireless neural recording systems, this paper presents DustNet: a spatially-distributed network of ultrasonically-powered neural recording implants capable of supporting up to 8 simultaneously recording nodes over a single ultrasound link. To enable high throughput multi-implant communication, DustNet implements a time-division multiple-access (TDMA) protocol with up to 16-level amplitude modulation of the ultrasound backscatter that achieves up to 4x higher data rates than traditional on-off keying methods. Each neural implant consists of a 0.7x0.7x0.7 mm$^3$ piezoceramic transducer, a 100 nF off-chip capacitor, and an IC mounted on a flexible PCB. The implant IC was fabricated in a 28nm CMOS process and occupies an area of 0.43 mm$^2$. System functionality was verified at 90mm depth in oil, achieving a maximum measured data rate of 200 kb/s at 2 MHz ultrasound carrier frequency, with each implant transmitting uplink data at 50 kb/s and dissipating just 7 $\mu$W; the system is demonstrated to support up to 400 kb/s total data rate over the same link.

eess.SY↗

arXiv2025

Correcting Fabrication-Induced Curvature in Micromirror-Based Spatial Light Modulators with a Microlens Array

Computer generated holography requires high-speed spatial light modulators (SLMs) for dynamically patterning light in 3D. Piston-motion micromirror-based SLMs support high-speed ($\geq$ 10 kHz) phase modulation; however, fabricating micromirror arrays with sufficient fill factor necessary for high diffraction efficiency is challenging. In particular, the larger mirrors of high fill factor designs are susceptible to stress-induced curvature that significantly degrades optical performance. In this work, we introduce an optical compensation method using a pitch-matched microlens array (MLA) to focus light onto just the center of each mirror. Our approach thus avoids curvature-induced artifacts and improves optical fill factor to nearly 100$\%$, independent of the original mechanical fill factor. Through simulations and experiments on a fabricated micromirror array with bowed mirrors, we show that the Pearson correlation coefficient of the imparted phase profile is improved from 0.11 to 0.85 and the brightness of a holographically-generated single spot is enhanced by 8$\times$ with our microlens array in place. Our hybrid optical-electromechanical strategy thus provides a scalable path toward high-speed, high-fidelity wavefront control for applications such as adaptive optics, holographic displays, and optogenetics.

physics.optics↗

arXiv2025

Neural Recording Power Optimization Through Machine Learning Guided Resolution Reconfiguration

Neural recording implants are a crucial tool for both neuroscience research and enabling new clinical applications. The power consumption of high channel count implants is dominated by the circuits used to amplify and digitize neural signals. Since circuit designers have pushed the efficiency of these circuits close to the theoretical physical limits, reducing power further requires system level optimization. Recent advances use a strategy called channel selection, in which less important channels are turned off to save power. We demonstrate resolution reconfiguration, in which the resolution of less important channels is scaled down to save power. Our approach leverages variable importance of each channel inside machine-learning-based decoders and we trial this methodology across three applications: seizure detection, gesture recognition, and force regression. With linear decoders, resolution reconfiguration saves 8.7x, 12.8x, and 23.0x power compared to a traditional recording array for each task respectively. It further saves 1.6x, 3.4x, and 5.2x power compared to channel selection. The results demonstrate the power benefits of resolution reconfigurable front-ends and their wide applicability to neural decoding problems.

q-bio.NC↗

arXiv2025

MRDust: Wireless Implant Data Uplink & Localization via Magnetic Resonance Image Modulation

Magnetic resonance imaging (MRI) exhibits rich and clinically useful endogenous contrast mechanisms, which can differentiate soft tissues and are sensitive to flow, diffusion, magnetic susceptibility, blood oxygenation level, and more. However, MRI sensitivity is ultimately constrained by Nuclear Magnetic Resonance (NMR) physics, and its spatiotemporal resolution is limited by SNR and spatial encoding. On the other hand, miniaturized implantable sensors offer highly localized physiological information, yet communication and localization can be challenging when multiple implants are present. This paper introduces the MRDust, an active ``contrast agent" that integrates active sensor implants with MRI, enabling the direct encoding of highly localized physiological data into MR images to augment the anatomical images. MRDust employs a micrometer-scale on-chip coil to actively modulate the local magnetic field, enabling MR signal amplitude and phase modulation for digital data transmission. Since MRI inherently captures the anatomical tissue structure, this method has the potential to enable simultaneous data communication, localization, and image registration with multiple implants. This paper presents the underlying physical principles, design tradeoffs, and design methodology for this approach. To validate the concept, a 900 $\times$ 990 $\mu$m$^2$ chip was designed using TSMC 28 nm technology, with an on-chip coil measuring 630 $\mu$m in diameter. The chip was tested with custom hardware in an MR750W GE3T MRI scanner. Successful voxel amplitude modulation is demonstrated with Spin-Echo Echo-Planar-Imaging (SE-EPI) sequence, achieving a contrast-to-noise ratio (CNR) of 25.58 with a power consumption of 130 $\mu$W.

eess.SY↗

arXiv2024

SPIRIT: Low Power Seizure Prediction using Unsupervised Online-Learning and Zoom Analog Frontends

Early prediction of seizures and timely interventions are vital for improving patients' quality of life. While seizure prediction has been shown in software-based implementations, to enable timely warnings of upcoming seizures, prediction must be done on an edge device to reduce latency. Ideally, such devices must also be low-power and track long-term drifts to minimize maintenance from the user. This work presents SPIRIT: Stochastic-gradient-descent-based Predictor with Integrated Retraining and In situ accuracy Tuning. SPIRIT is a complete system-on-a-chip (SoC) integrating an unsupervised online-learning seizure prediction classifier with eight 14.4 uW, 0.057 mm2, 90.5 dB dynamic range, Zoom Analog Frontends. SPIRIT achieves, on average, 97.5%/96.2% sensitivity/specificity respectively, predicting seizures an average of 8.4 minutes before they occur. Through its online learning algorithm, prediction accuracy improves by up to 15%, and prediction times extend by up to 7x, without any external intervention. Its classifier consumes 17.2 uW and occupies 0.14 mm2, the lowest reported for a prediction classifier by >134x in power and >5x in area. SPIRIT is also at least 5.6x more energy efficient than the state-of-the-art.

eess.SP↗

arXiv2024

Towards EMG-to-Speech with a Necklace Form Factor

Electrodes for decoding speech from electromyography (EMG) are typically placed on the face, requiring adhesives that are inconvenient and skin-irritating if used regularly. We explore a different device form factor, where dry electrodes are placed around the neck instead. 11-word, multi-speaker voiced EMG classifiers trained on data recorded with this device achieve 92.7% accuracy. Ablation studies reveal the importance of having more than two electrodes on the neck, and phonological analyses reveal similar classification confusions between neck-only and neck-and-face form factors. Finally, speech-EMG correlation experiments demonstrate a linear relationship between many EMG spectrogram frequency bins and self-supervised speech representation dimensions.

eess.AS↗

arXiv2024

Wireless Ear EEG to Monitor Drowsiness

Neural wearables can enable life-saving drowsiness and health monitoring for pilots and drivers. While existing in-cabin sensors may provide alerts, wearables can enable monitoring across more environments. Current neural wearables are promising but most require wet-electrodes and bulky electronics. This work showcases in-ear, dry-electrode earpieces used to monitor drowsiness with compact hardware. The employed system integrates additive-manufacturing for dry, user-generic earpieces, existing wireless electronics, and offline classification algorithms. Thirty-five hours of electrophysiological data were recorded across nine subjects performing drowsiness-inducing tasks. Three classifier models were trained with user-specific, leave-one-trial-out, and leave-one-user-out splits. The support-vector-machine classifier achieved an accuracy of 93.2% while evaluating users it has seen before and 93.3% when evaluating a never-before-seen user. These results demonstrate wireless, dry, user-generic earpieces used to classify drowsiness with comparable accuracies to existing state-of-the-art, wet electrode in-ear and scalp systems. Further, this work illustrates the feasibility of population-trained classification in future electrophysiological applications.

q-bio.QM↗

arXiv2023

Fast non-iterative algorithm for 3D point-cloud holography

Recently developed iterative and deep learning-based approaches to computer-generated holography (CGH) have been shown to achieve high-quality photorealistic 3D images with spatial light modulators. However, such approaches remain overly cumbersome for patterning sparse collections of target points across a photoresponsive volume in applications including biological microscopy and material processing. Specifically, in addition to requiring heavy computation that cannot accommodate real-time operation in mobile or hardware-light settings, existing sampling-dependent 3D CGH methods preclude the ability to place target points with arbitrary precision, limiting accessible depths to a handful of planes. Accordingly, we present a non-iterative point cloud holography algorithm that employs fast deterministic calculations in order to efficiently allocate patches of SLM pixels to different target points in the 3D volume and spread the patterning of all points across multiple time frames. Compared to a matched-performance implementation of the iterative Gerchberg-Saxton algorithm, our algorithm's relative computation speed advantage was found to increase with SLM pixel count, exceeding 100,000x at 512x512 array format.

cs.GR↗

arXiv2022

A Sparse Sampling Sensor Front-end IC for Low Power Continuous SpO$_2$ \& HR Monitoring

Photoplethysmography (PPG) is an attractive method to acquire vital signs such as heart rate and blood oxygenation and is frequently used in clinical and at-home settings. Continuous operation of health monitoring devices demands a low power sensor that does not restrict the device battery life. Silicon photodiodes (PD) and LEDs are commonly used as the interface devices in PPG sensors; however, using of flexible organic devices can enhance the sensor conformality and reduce the cost of fabrication. In most PPG sensors, most of system power consumption is concentrated in powering LEDs, traditionally consuming mWs. Using organic devices further increases this power demand since these devices exhibit larger parasitic capacitances and typically need higher drive voltages. This work presents a sensor IC for continuous SpO$_2$ and HR monitoring that features an on-chip reconstruction-free sparse sampling algorithm to reduce the overall system power consumption by $\sim$70\% while maintaining the accuracy of the output information. The designed frontend is compatible with a wide range of devices from silicon PDs to organic PDs with parasitic capacitances up to 10 nF. Implemented in a 40 nm HV CMOS process, the chip occupies 2.43 mm$^2$ and consumes 49.7 $μ$W and 15.2 $μ$W of power in continuous and sparse sampling modes respectively. The performance of the sensor IC has been verified \textit{in vivo} with both types of devices and the results are compared against a clinical grade reference. Less than 1 bpm and 1\% mean absolute errors were achieved in both continuous and sparse modes of operation.

eess.SY↗

arXiv2022

Phased array beamforming methods for powering biomedical ultrasonic implants

Millimeter-scale implants using ultrasound for power and communication have been proposed for a range of deep-tissue applications, including neural recording and stimulation. However, published implementations have shown high sensitivity to misalignment with the external ultrasound transducer. Ultrasonic beamforming using a phased array to these implants can improve tolerance to misalignment, reduce implant volume, and allow multiple implants to be operated simultaneously in different locations. This paper details the design of a custom planar phased array ultrasound system, which is capable of steering and focusing ultrasound power within a 3D volume. Analysis and simulation is performed to determine the choice of array element pitch, with special attention given to maximizing the power available at the implant while meeting FDA limits for diagnostic ultrasound. Time reversal is proposed as a computationally simple approach to beamforming that is robust despite scattering and inhomogeneity of the acoustic medium. This technique is demonstrated both in active drive and pulse-echo modes, and it is experimentally compared with other beamforming techniques by measuring energy transfer efficiency. Simultaneous power delivery to multiple implants is also demonstrated.

eess.SY↗

arXiv2022

SOUL: An Energy-Efficient Unsupervised Online Learning Seizure Detection Classifier

Implantable devices that record neural activity and detect seizures have been adopted to issue warnings or trigger neurostimulation to suppress epileptic seizures. Typical seizure detection systems rely on high-accuracy offline-trained machine learning classifiers that require manual retraining when seizure patterns change over long periods of time. For an implantable seizure detection system, a low power, at-the-edge, online learning algorithm can be employed to dynamically adapt to the neural signal drifts, thereby maintaining high accuracy without external intervention. This work proposes SOUL: Stochastic-gradient-descent-based Online Unsupervised Logistic regression classifier. After an initial offline training phase, continuous online unsupervised classifier updates are applied in situ, which improves sensitivity in patients with drifting seizure features. SOUL was tested on two human electroencephalography (EEG) datasets: the CHB-MIT scalp EEG dataset, and a long (>100 hours) NeuroVista intracranial EEG dataset. It was able to achieve an average sensitivity of 97.5% and 97.9% for the two datasets respectively, at >95% specificity. Sensitivity improved by at most 8.2% on long-term data when compared to a typical seizure detection classifier. SOUL was fabricated in TSMC's 28 nm process occupying 0.1 mm2 and achieves 1.5 nJ/classification energy efficiency, which is at least 24x more efficient than state-of-the-art.

eess.SP↗

arXiv2022

Rapid Fabrication of Low Impedance, 3D Dry Electrodes for Physiological Sensing

Medical electrophysiological sensors that can study the body and diagnose diseases depend on consistently low impedance electrode-skin interfaces. Clinical-standard wet electrodes use hydrogels and skin abrasion to improve the interface and thus the recorded signal quality. These electrodes are challenging to self-administer and impede in-home care. Wearable dry electrodes are more practical; however, they show higher impedances than wet electrodes and are costly to customize. This work presents a fabrication method for rapidly producing low impedance, anatomically fit dry electrodes that do not require hydrogels. By using electroless copper and gold plating with 3D printing, biocompatible electrodes can be optimized for individuals at a fraction of the cost of existing vacuum deposition-based techniques. Example 3D dry electrodes made with this process are evaluated alongside clinical-standard devices in typical scenarios to compare electrical performance and comfort. The resulting dry electrodes exhibited an average electrode-skin impedance of 66.7 kΩ at 50Hz and DC offset of -20 mV without any hydrogel, which, when area normalized, is within the range achieved by wet electrodes without skin abrasion.

physics.med-ph↗

arXiv2021

A MEMS-based optical scanning system for precise, high-speed neural interfacing

Optical scanning is a prevalent technique for optical neural interfaces where light delivery with high spatial and temporal precision is desired. However, due to the sequential nature of point-scanning techniques, the settling time of optical modulators is a major bottleneck for throughput and limits random-access targeting capabilities. While fast lateral scanners exist, commercially available varifocal elements are constrained to >3ms settling times, limiting the speed of the overall system to hundreds of Hz. Faster focusing methods exist but cannot combine fast operation and dwelling capability with electrical and optical efficiency. Here, we present a varifocal mirror comprised of an array of piston-motion MEMS micromirrors and a custom driver ASIC, offering fast operation with dwelling capability while maintaining high diffraction efficiency. The ASIC features a reconfigurable nonlinear DAC to simultaneously compensate for the built-in nonlinearity of electrostatic actuators and the global process variations in MEMS mirrors. Experimental results demonstrate a wide continuous sweeping range that spans 22 distinctly resolvable depth planes with refresh rates greater than 12 kHz.

physics.app-ph↗

arXiv2021

Memory-Efficient, Limb Position-Aware Hand Gesture Recognition using Hyperdimensional Computing

Electromyogram (EMG) pattern recognition can be used to classify hand gestures and movements for human-machine interface and prosthetics applications, but it often faces reliability issues resulting from limb position change. One method to address this is dual-stage classification, in which the limb position is first determined using additional sensors to select between multiple position-specific gesture classifiers. While improving performance, this also increases model complexity and memory footprint, making a dual-stage classifier difficult to implement in a wearable device with limited resources. In this paper, we present sensor fusion of accelerometer and EMG signals using a hyperdimensional computing model to emulate dual-stage classification in a memory-efficient way. We demonstrate two methods of encoding accelerometer features to act as keys for retrieval of position-specific parameters from multiple models stored in superposition. Through validation on a dataset of 13 gestures in 8 limb positions, we obtain a classification accuracy of up to 93.34%, an improvement of 17.79% over using a model trained solely on EMG. We achieve this while only marginally increasing memory footprint over a single limb position model, requiring $8\times$ less memory than a traditional dual-stage classification architecture.

cs.LG↗

arXiv2020

A micromirror array with annular partitioning for high-speed random-access axial focusing

Dynamic axial focusing functionality has recently experienced widespread incorporation in microscopy, augmented/virtual reality (AR/VR), adaptive optics, and material processing. However, the limitations of existing varifocal tools continue to beset the performance capabilities and operating overhead of the optical systems that mobilize such functionality. The varifocal tools that are the least burdensome to drive (ex: liquid crystal, elastomeric or optofluidic lenses) suffer from low (~ 100 Hz) refresh rates. Conversely, the fastest devices sacrifice either critical capabilities such as their dwelling capacity (ex: acoustic gradient lenses or monolithic micromechanical mirrors) or low operating overhead (e.g., deformable mirrors). Here, we present a general-purpose random-access axial focusing device that bridges these previously conflicting features of high speed, dwelling capacity and lightweight drive by employing low-rigidity micromirrors that exploit the robustness of defocusing phase profiles. Geometrically, the device consists of an 8.2 mm diameter array of piston-motion and 48 um-pitch micromirror pixels that provide 2pi phase shifting for wavelengths shorter than 1 100 nm with 10-90 % settling in 64.8 us (i.e., 15.44 kHz refresh rate). The pixels are electrically partitioned into 32 rings for a driving scheme that enables phase-wrapped operation with circular symmetry and requires less than 30 V per channel. Optical experiments demonstrated the array's wide focusing range with a measured ability to target 29 distinct, resolvable depth planes. Overall, the features of the proposed array offer the potential for compact, straightforward methods of tackling bottlenecked applications including high-throughput single-cell targeting in neurobiology and the delivery of dense 3D visual information in AR/VR.

physics.optics↗

arXiv2020

Optimizing Volumetric Efficiency and Modeling Backscatter Communication in Biosensing Ultrasonic Implants

Ultrasonic backscatter communication has gained popularity in recent years with the advent of deep-tissue sub-mm scale biosensing implants in which piezoceramic (piezo) resonators are used as acoustic antennas. Miniaturization is a key design goal for such implants to reduce tissue displacement and enable minimally invasive implantation techniques. Here, we provide a systematic design approach for the implant piezo geometry and operation frequency to minimize the overall volume of the implant. Moreover, a critical design aspect of an ultrasonic backscatter communication link is the response of the piezo acoustic reflection coefficient $Γ$ with respect to the variable shunt impedance, $Z_E$, of the implant uplink modulator. Due to the complexity of the piezo governing equations and multi-domain, electro-acoustical nature of the piezo, $Γ(Z_E)$ has often been characterized numerically and the implant uplink modulator has been designed empirically resulting in sub-optimal performance in terms of data rate and linearity. Here, we present a SPICE friendly end-to-end equivalent circuit model of the channel as a piezo-IC co-simulation tool that incorporates inherent path losses present in a typical ultrasonic backscatter channel. The circuit model is then used to simulate the channel transient response in a common CAD tool. To provide further insight into the channel response, we present experimentally validated closed form expressions for $Γ(Z_E)$ under various boundary conditions. These expressions couple $Γ$ to the commonly used Thevenin equivalent circuit model of the piezo, facilitating systematic design and synthesis of ultrasonic backscatter uplink modulators

physics.app-ph↗

arXiv2020

Wireless User-Generic Ear EEG

In the past few years it has been demonstrated that electroencephalography (EEG) can be recorded from inside the ear (in-ear EEG). To open the door to low-profile earpieces as wearable brain-computer interfaces (BCIs), this work presents a practical in-ear EEG device based on multiple dry electrodes, a user-generic design, and a lightweight wireless interface for streaming data and device programming. The earpiece is designed for improved ear canal contact across a wide population of users and is fabricated in a low-cost and scalable manufacturing process based on standard techniques such as vacuum forming,plasma-treatment, and spray coating. A 2.5x2.5 cm^2 wireless recording module is designed to record and stream data wirelessly to a host computer. Performance was evaluated on three human subjects over three months and compared with clinical-grade wet scalp EEG recordings. Recordings of spontaneous and evoked physiological signals, eye-blinks, alpha rhythm, and the auditory steady-state response (ASSR), are presented. This is the first wireless in-ear EEG to our knowledge to incorporate a dry multielectrode, user-generic design. The user-generic ear EEG recorded a mean alpha modulation of 2.17, outperforming the state-of-the-art in dry electrode in-ear EEG systems.

q-bio.QM↗

arXiv2020

StimDust: A mm-scale implantable wireless precision neural stimulator with ultrasonic power and communication

Neural stimulation is a powerful technique for modulating physiological functions and for writing information into the nervous system as part of brain-machine interfaces. Current clinically approved neural stimulators require batteries and are many cubic centimetres in size -- typically much larger than their intended targets. We present a complete wireless neural stimulation system consisting of a 1.7 mm3 wireless, batteryless, leadless implantable stimulator (the "mote"), an ultrasonic wireless link for power and bi-directional communication, and a hand-held external transceiver. The mote consists of a piezoceramic transducer, an energy storage capacitor, and a stimulator integrated circuit (IC). The IC harvests ultrasonic power with high efficiency, decodes stimulation parameter downlink data, and generates current-controlled stimulation pulses. Stimulation parameters are time-encoded on the fly through the wireless link rather than being programmed and stored on the mote, reducing power consumption and on-chip memory requirements and enabling complex stimulation protocols with high-temporal resolution and low-latency feedback for use in closed-loop stimulation. Uplink data indicates whether the mote is currently stimulating; it is encoded by the mote via backscatter modulation and is demodulated at the external transceiver. We show that the mote operates at an acoustic intensity that is 7.8% of the FDA limit for diagnostic ultrasound and characterize the acoustic wireless link's robustness to expected real-world misalignment. We demonstrate the in vivo performance of the system with motes acutely implanted with a cuff on the sciatic nerve of anesthetized rats and show highly repeatable stimulation across a wide range of physiological responses.

q-bio.NC↗