Improving flat fluorescence microscopy in scattering tissue through deep learning strategies.

Intravital microscopy in small animals growingly contributes to the visualization of short- and long-term mammalian biological processes. Miniaturized fluorescence microscopy has revolutionized the observation of live animals' neural circuits. The technology's ability to further miniaturize to improve freely moving experimental settings is limited by its standard lens-based layout. Typical miniature microscope designs contain a stack of heavy and bulky optical components adjusted at relatively long distances. Computational lensless microscopy can overcome this limitation by replacing the lenses with a simple thin mask. Among other critical applications, Flat Fluorescence Microscope (FFM) holds promise to allow for real-time brain circuits imaging in freely moving animals, but recent research reports show that the quality needs to be improved, compared with imaging in clear tissue, for instance. Although promising results were reported with mask-based fluorescence microscopes in clear tissues, the impact of light scattering in biological tissue remains a major challenge. The outstanding performance of deep learning (DL) networks in computational flat cameras and imaging through scattering media studies motivates the development of deep learning models for FFMs. Our holistic ray-tracing and Monte Carlo FFM computational model assisted us in evaluating deep scattering medium imaging with DL techniques. We demonstrate that physics-based DL models combined with the classical reconstruction technique of the alternating direction method of multipliers (ADMM) perform a fast and robust image reconstruction, particularly in the scattering medium. The structural similarity indexes of the reconstructed images in scattering media recordings were increased by up to 20% compared with the prevalent iterative models. We also introduce and discuss the challenges of DL approaches for FFMs under physics-informed supervised and unsupervised learning.

Source-Detector Spectral Pairing-Related Inaccuracies in Pulse Oximetry: Evaluation of the Wavelength Shift.

Pulse oximetry enables oxygen saturation estimation ( S p O 2) non-invasively in real time with few components and modest processing power. With the advent of affordable development kits dedicated to the monitoring of biosignals, capabilities once reserved to hospitals and high-end research laboratories are becoming accessible for rapid prototyping. While one may think that medical-grade equipment differs greatly in quality, surprisingly, we found that the performance requirements are not widely different from available consumer-grade components, especially regarding the photodetection module in pulse oximetry. This study investigates how the use of candidate light sources and photodetectors for the development of a custom S p O 2 monitoring system can lead to inaccuracies when using the standard computational model for oxygen saturation without calibration. Following the optical characterization of selected light sources, we compare the extracted parameters to the key features in their respective datasheet. We then quantify the wavelength shift caused by spectral pairing of light sources in association with photodetectors. Finally, using the widely used approximation, we report the resulting absolute error in S p O 2 estimation and show that it can lead up to 8% of the critical 90-100% saturation window.

A Low-Cost, Wireless, 3-D-Printed Custom Armband for sEMG Hand Gesture Recognition.

Wearable technology can be employed to elevate the abilities of humans to perform demanding and complex tasks more efficiently. Armbands capable of surface electromyography (sEMG) are attractive and noninvasive devices from which human intent can be derived by leveraging machine learning. However, the sEMG acquisition systems currently available tend to be prohibitively costly for personal use or sacrifice wearability or signal quality to be more affordable. This work introduces the 3DC Armband designed by the Biomedical Microsystems Laboratory in Laval University; a wireless, 10-channel, 1000 sps, dry-electrode, low-cost (∼150 USD) myoelectric armband that also includes a 9-axis inertial measurement unit. The proposed system is compared with the Myo Armband by Thalmic Labs, one of the most popular sEMG acquisition systems. The comparison is made by employing a new offline dataset featuring 22 able-bodied participants performing eleven hand/wrist gestures while wearing the two armbands simultaneously. The 3DC Armband systematically and significantly ( p < 0.05 ) outperforms the Myo Armband, with three different classifiers employing three different input modalities when using ten seconds or more of training data per gesture. This new dataset, alongside the source code, Altium project and 3-D models are made readily available for download within a Github repository.

Miniaturized FDDA and CMOS Based Potentiostat for Bio-Applications.

A novel fully differential difference CMOS potentiostat suitable for neurotransmitter sensing is presented. The described architecture relies on a fully differential difference amplifier (FDDA) circuit to detect a wide range of reduction-oxidation currents, while exhibiting low-power consumption and low-noise operation. This is made possible thanks to the fully differential feature of the FDDA, which allows to increase the source voltage swing without the need for additional dedicated circuitry. The FDDA also reduces the number of amplifiers and passive elements in the potentiostat design, which lowers the overall power consumption and noise. The proposed potentiostat was fabricated in 0.18 µm CMOS, with 1.8 V supply voltage. The device achieved 5 µA sensitivity and 0.99 linearity. The input-referred noise was 6.9 µV rms and the flicker noise was negligible. The total power consumption was under 55 µW. The complete system was assembled on a 20 mm × 20 mm platform that includes the potentiostat chip, the electrode terminals and an instrumentation amplifier for redox current buffering, once converted to a voltage by a series resistor. the chip dimensions were 1 mm × 0.5 mm and the other PCB components were off-chip resistors, capacitors and amplifiers for data acquisition. The system was successfully tested with ferricyanide, a stable electroactive compound, and validated with dopamine, a popular neurotransmitter.

The anterolateral thigh fold-over flap for total and subtotal glossectomy reconstruction.

BACKGROUND: We present a fold-over technique for total and subtotal glossectomy reconstruction utilizing the ALT flap with the goal of increasing soft tissue bulk. METHODS: Retrospective review of total and subtotal glossectomy patients undergoing ALT fold-over flap reconstruction from January 2007 to December 2012. RESULTS: Eight patients fit inclusion criteria. Average length of follow-up was 35 months. There were no flap failures. At last follow-up, no patients were tracheostomy dependent. Six out of 8 patients (75%) did not require a g-tube and were able to maintain their weight. Speech intelligibility was good in 6/8 patients (75%). CONCLUSIONS: The ALT fold-over flap technique provides sufficient bulk to allow decannulation and airway protection as well as PO intake in most patients. Given the minimal donor site morbidity, it is our technique of choice for total and subtotal glossectomy defects. © 2016 Wiley Periodicals, Inc. Microsurgery 36:297-302, 2016.

A Wireless Multi-Sensor Dielectric Impedance Spectroscopy Platform.

This paper describes the development of a low-cost, miniaturized, multiplexed, and connected platform for dielectric impedance spectroscopy (DIS), designed for in situ measurements and adapted to wireless network architectures. The platform has been tested and used as a DIS sensor node on ZigBee mesh and was able to interface up to three DIS sensors at the same time and relay the information through the network for data analysis and storage. The system is built from low-cost commercial microelectronics components, performs dielectric spectroscopy ranging from 5 kHz to 100 kHz, and benefits from an on-the-fly calibration system that makes sensor calibration easy. The paper describes the microelectronics design, the Nyquist impedance response, the measurement sensitivity and accuracy, and the testing of the platform for in situ dielectric impedance spectroscopy applications pertaining to fertilizer sensing, water quality sensing, and touch sensing.

Electrochemical Impedance Sensors for Monitoring Trace Amounts of NO3 in Selected Growing Media.

With the advent of smart cities and big data, precision agriculture allows the feeding of sensor data into online databases for continuous crop monitoring, production optimization, and data storage. This paper describes a low-cost, compact, and scalable nitrate sensor based on electrochemical impedance spectroscopy for monitoring trace amounts of NO3- in selected growing media. The nitrate sensor can be integrated to conventional microelectronics to perform online nitrate sensing continuously over a wide concentration range from 0.1 ppm to 100 ppm, with a response time of about 1 min, and feed data into a database for storage and analysis. The paper describes the structural design, the Nyquist impedance response, the measurement sensitivity and accuracy, and the field testing of the nitrate sensor performed within tree nursery settings under ISO/IEC 17025 certifications.

A Wireless Optogenetic Headstage with Multichannel Electrophysiological Recording Capability.

We present a small and lightweight fully wireless optogenetic headstage capable of optical neural stimulation and electrophysiological recording. The headstage is suitable for conducting experiments with small transgenic rodents, and features two implantable fiber-coupled light-emitting diode (LED) and two electrophysiological recording channels. This system is powered by a small lithium-ion battery and is entirely built using low-cost commercial off-the-shelf components for better flexibility, reduced development time and lower cost. Light stimulation uses customizable stimulation patterns of varying frequency and duty cycle. The optical power that is sourced from the LED is delivered to target light-sensitive neurons using implantable optical fibers, which provide a measured optical power density of 70 mW/mm² at the tip. The headstage is using a novel foldable rigid-flex printed circuit board design, which results into a lightweight and compact device. Recording experiments performed in the cerebral cortex of transgenic ChR2 mice under anesthetized conditions show that the proposed headstage can trigger neuronal activity using optical stimulation, while recording microvolt amplitude electrophysiological signals.

Novel wireless-communicating textiles made from multi-material and minimally-invasive fibers.

The ability to integrate multiple materials into miniaturized fiber structures enables the realization of novel biomedical textile devices with higher-level functionalities and minimally-invasive attributes. In this work, we present novel textile fabrics integrating unobtrusive multi-material fibers that communicate through 2.4 GHz wireless networks with excellent signal quality. The conductor elements of the textiles are embedded within the fibers themselves, providing electrical and chemical shielding against the environment, while preserving the mechanical and cosmetic properties of the garments. These multi-material fibers combine insulating and conducting materials into a well-defined geometry, and represent a cost-effective and minimally-invasive approach to sensor fabrics and bio-sensing textiles connected in real time to mobile communications infrastructures, suitable for a variety of health and life science applications.

High Precision Ping-pong Auto-zeroed Lock-in Fluorescence Photometry Sensor.

This paper presents a high-precision CMOS fluorescence photometry sensor using a novel lock-in amplification scheme based on switched-biasing and ping-pong auto-zeroing techniques. The CMOS sensor includes two photodiodes and a lock-in amplifier (LIA) operating at 1 kHz. The LIA comprises a differential low-noise amplifier using a novel switched-biasing ping-pong auto-zeroed scheme, an automatic phase aligner, a programmable gain amplifier, a band-pass filter, a mixer, and an output low-pass filter. The design is fabricated in 0.18-µm CMOS process, and the measurement shows that the LIA can retrieve noisy input signals with a dynamic reserve of 42 dB, while consuming only 0.7 mW from a 1.8 V supply voltage. The measured results show that the LIA can detect a wide range of incident light power from 8 nW to 24 µW. The proposed design is encapsulated in a 3D-printed housing allowing for real-time in vitro biomarker detection. This ambulatory platform uses an LED and a fiber optic to convey the excitation light to the sample and retrieve the fluorescence signal. Experiments with a beads solution diluted in PBS demonstrate that the sensor has a sensitivity of 1:100 k. Experimental results obtained in vitro with NIH3T3 mouse cells tagged with membrane dye show the ability of the prototype to detect different densities of cell culture. The portable prototype, which includes optical filters and a small 30 mm × 36 mm × 30 mm printed circuit board enclosed inside the 3D-printed housing, consumes 36.7 mW and weighs 120 g.

Portable impedance-sensing device for microorganism characterization in the field.

A variety of biosensors have been proposed to quickly detect and measure the properties of individual microorganisms among heterogeneous populations, but challenges related to cost, portability, stability, sensitivity, and power consumption limit their applicability. This study proposes a portable microfluidic device based on impedance flow-cytometry and electrical impedance spectroscopy that can detect and quantify the size of microparticles larger than 45 µm, such as algae and microplastics. The system is low cost ($300), portable (5 cm [Formula: see text] 5 cm), low-power (1.2 W), and easily fabricated utilizing a 3D-printer and industrial printed circuit board technology. The main novelty we demonstrate is the use of square wave excitation signal for impedance measurements with quadrature phase-sensitive detectors. A linked algorithm removes the errors associated to higher order harmonics. After validating the performance of the device for complex impedance models, we used it to detect and differentiate between polyethylene microbeads of sizes between 63 and 83 µm, and buccal cells between 45 and 70 µm. A precision of 3% is reported for the measured impedance and a minimum size of 45 µm is reported for the particle characterization.

Deep Learning Empowered Fresnel-based Lensless Fluorescence Microscopy.

Miniaturized fluorescence microscopy has revolutionized the way neuroscientists study the brain in-vivo. Recent developments in computational lensless imaging promise a next generation of miniaturized microscopes in lensless fluorescence microscopy. We developed a microscope prototype using an optimized Fresnel amplitude mask. While many lensless imaging modalities have reported excellent performance using Deep Learning (DL) approaches, DL application in fluorescence imaging has been left untouched. We generated a computational dataset based on experimental system calibration to evaluate DL capabilities on biological cell morphologies. We show that our DL-assisted microscope can provide high-quality imaging with a structural similarity index of 89%. The least absolute error was decreased by 63% using the DL-assisted method compared with the classical models. The state-of-the-art performance of this prototype enhances the expected potential of amplitude masks in lensless microscopy applications, which are critical for robust in-vivo flat microscopy with engineered image sensors.Clinical Relevance- This study aids in advancing miniaturized fluorescence microscopy, which greatly impacts long-term brain circuit and disease studies in freely moving animal models.

Intensive care photoplethysmogram datasets and machine-learning for blood pressure estimation: Generalization not guarantied.

The large MIMIC waveform dataset, sourced from intensive care units, has been used extensively for the development of Photoplethysmography (PPG) based blood pressure (BP) estimation algorithms. Yet, because the data comes from patients in severe conditions-often under the effect of drugs-it is regularly noted that the relationship between BP and PPG signal characteristics may be anomalous, a claim that we investigate here. A sample of 12,000 records from the MIMIC waveform dataset was stacked up against the 219 records of the PPG-BP dataset, an alternative public dataset obtained under controlled experimental conditions. The distribution of systolic and diastolic BP data and 31 PPG pulse morphological features was first compared between datasets. Then, the correlation between features and BP, as well as between the features themselves, was analysed. Finally, regression models were trained for each dataset and validated against the other. Statistical analysis showed significant p < 0.001 differences between the datasets in diastolic BP and in 20 out of 31 features when adjusting for heart rate differences. The eight features showing the highest rank correlation ρ   >   0.40 to systolic BP in PPG-BP all displayed muted correlation levels ρ   <   0.10 in MIMIC. Regression tests showed twice higher baseline predictive power with PPG-BP than with MIMIC. Cross-dataset regression displayed a practically complete loss of predictive power for all models. The differences between the MIMIC and PPG-BP dataset exposed in this study suggest that BP estimation models based on the MIMIC dataset have reduced predictive power on the general population.

A Low-Power High Input Range PPG Readout Amplifier with a Current Buffer Input.

This paper presents ultra-low power photoplethysmography (PPG) readout circuits. The proposed system architecture uses a current buffer between the photodiode (PD) and the transimpedance amplifier (TIA) to isolate the large parasitic capacitance of the PD leading to improves the power consumption of the TIA. A class AB topology is exploited at the output of the amplifier, which allows for increased drive capability without the use of auxiliary circuits. The maximum input current range of the TIA is 160 µA, so the large DC current of the input signal does not saturate the circuit. In the LED driver circuit, by varying the duty cycle of a pulse wave modulation (PWM) signal, the ON and OFF times of the circuits. The amplifier and LED driver are manufactured in the 130 nm TSMC CMOS process. The power consumption of the circuits with a duty cycle of 1% is 3.28 µW (at VDD = 1.2V).Clinical Relevance- Vital signs are becoming a very important research topic due to the recent prevalence of COVID-19 and other respiratory diseases. This research aims to develop and interface circuits to monitor vital signs including blood pressure, heart rate, and respiratory rate to study respiratory disease, drug safety, and efficacy.

A fully-flexible and thermally adjustable implantable neural probe with a U-turn polyester microchannel.

This work presents a fully flexible implantable neural probe fabricated with Polydimethylsiloxane (PDMS) and including a thermally-tunable stiffness microchannel filled with Polyester. The probe includes an optimized microfluidics mixer for drug delivery. Polyester, which is solid at room temperature and has a low melting point close to body temperature, is used to decrease the stiffness of the probe after insertion, after getting in contact with tissues. We designed a U-turn microchannel inside the PDMS neural probe and filled it up with melted polyester. The microchannel has a cross-section of 30 µm × 5 µm and a length of 14.7 mm. The following probe dimensions were chosen after extensive simulation: thickness = 20 µm, width = 300 µm, and length = 7 mm. These values yield a buckling force above 1 mN, which is sufficient for proper insertion into the brain tissues. Simulation results show that the microfluidics mixer with a cross-section of 90 µm × 5 µm and a length of 7 mm has optimum performance for the desired flow rate and quantity of drug to deliver. The pressure drop inside the microfluidic channel is less than 0.43 kPa, which is appropriate for PDMS-PDMS bonding, whereas the Reynolds number is near 1.91k in the laminar regime. No leakage or bubble occurred during the experimental validation, which suggests an appropriate pressure and a laminar flow in the channel.

Recent Advances in CMOS Electrochemical Biosensor Design for Microbial Monitoring: Review and Design Methodology.

Rapid, high-sensitivity, and real-time characterization of microorganisms plays a significant role in several areas, including clinical diagnosis, human healthcare, early detection of outbreaks, and the protection of living beings. Integrating microbiology and electrical engineering promises the development of low-cost, miniaturized, autonomous, and high-sensitivity sensors to quantify and characterize bacterial strains at various concentrations. Electrochemical-based biosensors are receiving particular attention in microbiological applications among the different biosensing devices. Several approaches have been adopted to design and fabricate cutting-edge, miniaturized, and portable electrochemical biosensors to track and monitor bacterial cultures in real time. These techniques differ in their sensing interface circuits and microelectrode fabrication. The goals of this review are (1) to summarize the current state of CMOS sensing circuit designs in label-free electrochemical biosensors for bacteria monitoring and (2) to discuss the material and size of the electrodes used in electrochemical biosensors in microbiological applications. In this paper, we reviewed the latest and most advanced CMOS integrated interface circuits that have recently been used in electrochemical biosensors to identify and characterize bacteria species, such as impedance spectroscopy, capacitive, amperometry, and voltammetry, etc. In addition to the interface circuit design, other crucial factors, such as the material and scale of the electrodes, must be considered to increase the sensitivity of electrochemical biosensors. Surveying the literature in this field improves our knowledge about the impact of electrode designs and materials on sensing precision and will help future designers adapt, design, and fabricate appropriate electrode configurations based on their application. Thus, we summarized the conventional microelectrode designs and materials mainly employed in microbial sensors, including interdigitated electrodes (IDEs), microelectrode arrays (MEAs), paper, and carbon-based electrodes, etc.

EMaGer: A Wearable Full-Circumference HD-EMG Sensor and Data Augmentation Method for Robust Hand Gesture Recognition.

This work presents EMaGer, a new 360° 64-channel high-density electromyography (HD-EMG) bracelet combined with an original data augmentation method for improved robustness in gesture recognition. By leveraging homogeneous electrode density and powerful deep learning techniques, the sensor is capable of rotation invariance around the arm axis, thus increasing gesture recognition robustness to electrode movement and inter-session evaluation. The system is made of a 4x16 electrode array covering the full circumference of the limb, and uses a sampling frequency of 1 kHz and a 16-bit resolution. The sensor's uniform and adjustable geometry paired with an array barrel shifting data augmentation (ABSDA) technique allows a convolutional neural network to maintain a 76.98% inter-session classification accuracy for a 6 gestures dataset, from a baseline intra-session accuracy of 93.75%. High inter-session classification accuracy decreases the training burden for users of EMG control systems such as myoelectric prostheses by minimizing calibration requirements. The same methods applied with different state-of-the-art sensors are demonstrated to be less effective. Thus, this work evidences the importance of co-designing the EMG sensor system with the gesture inference algorithms to leverage synergistic properties and solve state-of-the-art challenges.Clinical relevance- This paper establishes a method that alleviates clinical manipulations in setting up and calibrating myoelectric prosthetic devices.

Novel Wearable HD-EMG Sensor With Shift-Robust Gesture Recognition Using Deep Learning.

In this work, we present a hardware-software solution to improve the robustness of hand gesture recognition to confounding factors in myoelectric control. The solution includes a novel, full-circumference, flexible, 64-channel high-density electromyography (HD-EMG) sensor called EMaGer. The stretchable, wearable sensor adapts to different forearm sizes while maintaining uniform electrode density around the limb. Leveraging this uniformity, we propose novel array barrel-shifting data augmentation (ABSDA) approach used with a convolutional neural network (CNN), and an anti-aliased CNN (AA-CNN), that provides shift invariance around the limb for improved classification robustness to electrode movement, forearm orientation, and inter-session variability. Signals are sampled from a 4×16 HD-EMG array of electrodes at a frequency of 1 kHz and 16-bit resolution. Using data from 12 non-amputated participants, the approach is tested in response to sensor rotation, forearm rotation, and inter-session scenarios. The proposed ABSDA-CNN method improves inter-session accuracy by 25.67% on average across users for 6 gesture classes compared to conventional CNN classification. A comparison with other devices shows that this benefit is enabled by the unique design of the EMaGer array. The AA-CNN yields improvements of up to 63.05% accuracy over non-augmented methods when tested with electrode displacements ranging from -45 ° to +45 ° around the limb. Overall, this article demonstrates the benefits of co-designing sensor systems, processing methods, and inference algorithms to leverage synergistic and interdependent properties to solve state-of-the-art problems.

Recent advances in neural recording microsystems.

The accelerating pace of research in neuroscience has created a considerable demand for neural interfacing microsystems capable of monitoring the activity of large groups of neurons. These emerging tools have revealed a tremendous potential for the advancement of knowledge in brain research and for the development of useful clinical applications. They can extract the relevant control signals directly from the brain enabling individuals with severe disabilities to communicate their intentions to other devices, like computers or various prostheses. Such microsystems are self-contained devices composed of a neural probe attached with an integrated circuit for extracting neural signals from multiple channels, and transferring the data outside the body. The greatest challenge facing development of such emerging devices into viable clinical systems involves addressing their small form factor and low-power consumption constraints, while providing superior resolution. In this paper, we survey the recent progress in the design and the implementation of multi-channel neural recording Microsystems, with particular emphasis on the design of recording and telemetry electronics. An overview of the numerous neural signal modalities is given and the existing microsystem topologies are covered. We present energy-efficient sensory circuits to retrieve weak signals from neural probes and we compare them. We cover data management and smart power scheduling approaches, and we review advances in low-power telemetry. Finally, we conclude by summarizing the remaining challenges and by highlighting the emerging trends in the field.

Low-Cutoff Frequency Reduction in Neural Amplifiers: Analysis and Implementation in CMOS 65 nm.

Scaling down technology demotes the parameters of AC-coupled neural amplifiers, such as increasing the low-cutoff frequency due to the short-channel effects. To improve the low-cutoff frequency, one solution is to increase the feedback capacitors' value. This solution is not desirable, as the input capacitors have to be increased to maintain the same gain, which increases the area and decreases the input impedance of the neural amplifier. We analytically analyze the small-signal behavior of the neural amplifier and prove that the main reason for the increase of the low-cutoff frequency in advanced CMOS technologies is the reduction of the input resistance of the operational transconductance amplifier (OTA). We also show that the reduction of the input resistance of the OTA is due to the increase in the gate oxide leakage in the input transistors. In this paper, we explore this fact and propose two solutions to reduce the low-cutoff frequency without increasing the value of the feedback capacitor. The first solution is performed by only simulation and is called cross-coupled positive feedback that uses pseudoresistors to provide a negative resistance to increase the input resistance of the OTA. As an advantage, only standard CMOS transistors are used in this method. Simulation results show that a low-cutoff frequency of 1.5 Hz is achieved while the midband gain is 30.4 dB at 1 V. In addition, the power consumption is 0.6 µW. In the second method, we utilize thick-oxide MOS transistors in the input differential pair of the OTA. We designed and fabricated the second method in the 65 nm TSMC CMOS process. Measured results are obtained by in vitro recordings on slices of mouse brainstem. The measurement results show that the bandwidth is between 2 Hz and 5.6 kHz. The neural amplifier has 34.3 dB voltage gain in midband and consumes 3.63 µW at 1 V power supply. The measurement results show an input-referred noise of 6.1 µV rms and occupy 0.04 mm 2 silicon area.