Enhanced regularization for on-chip training using analog and temporary memory weights.

In-memory computing techniques are used to accelerate artificial neural network (ANN) training and inference tasks. Memory technology and architectural innovations allow efficient matrix-vector multiplications, gradient calculations, and updates to network weights. However, on-chip learning for edge devices is quite challenging due to the frequent updates. Here, we propose using an analog and temporary on-chip memory (ATOM) cell with controllable retention timescales for implementing the weights of an on-chip training task. Measurement results for Read-Write timescales are presented for an ATOM cell fabricated in GlobalFoundries' 45 nm RFSOI technology. The effect of limited retention and its variability is evaluated for training a fully connected neural network with a variable number of layers for the MNIST hand-written digit recognition task. Our studies show that weight decay due to temporary memory can have benefits equivalent to regularization, achieving a ∼33% reduction in the validation error (from 3.6% to 2.4%). We also show that the controllability of the decay timescale can be advantageous in achieving a further ∼26% reduction in the validation error. This strongly suggests the utility of temporary memory during learning before on-chip non-volatile memories can take over for the storage and inference tasks using the neural network weights. We thus propose an algorithm-circuit codesign in the form of temporary analog memory for high-performing on-chip learning of ANNs.

Part II: Impedance-based DNA biosensor for detection of isolated strains of phytopathogen Ralstonia solanacearum.

In Part I, we demonstrated the complete development of a label-free, ultra-low sample volume requiring DNA-based biosensor to detect Ralstonia solanacearum, an aerobic non-spore-forming, Gram-negative, plant pathogenic bacterium, using non-faradaic electrochemical impedance spectroscopy (nf-EIS). We also presented the sensor's sensitivity, specificity, and electrochemical stability. In this article, we highlight the specificity study of the developed DNA-based impedimetric biosensor to detect various strains of R. solanacearum. We have collected seven isolates of R. solanacearum isolated from locally infected host plants (eggplant, potato, tomato, chilli, and ginger) from different parts of Goa, India. The pathogenicity of these isolates was tested on the eggplant, and the pathogen was confirmed by microbiological plating and polymerase chain reaction (PCR). We further report the insight into the DNA hybridization on the surface of Interdigitated Electrodes (IDEs) and the expansion of the Randles model for more accurate analysis. The interpretation of the sensor specificity is clearly demonstrated by the capacitance change observed at the electrode-electrolyte interface.

Plant pathogenicity and associated/related detection systems. A review.

There is an increasing demand for the development of various tools for diagnosis and control of plant infections. The early diagnosis of plant disease serves as a vital element to improve crop productivity and meet demands of the ever-growing world population. The traditional methods of plant disease detection are time consuming, laborious and require 3-5 days to estimate the disease incidence. In this review, we focus on the advances in the detection techniques, mainly the miniaturized systems that has developed in the last decade. The analytical techniques for plant pathogen detection have been classified as direct and indirect detection methods. The direct methods involving laboratory techniques such as polymerase chain reaction, enzyme-linked immune-sorbent assays, and immunofluorescence and their recent advances have been discussed. Similarly, indirect methods which rely on sensing the plant stress indicators to detect plant diseases have been categorized and reviewed. In the last decade, various detection platforms with high sensitivity and selectivity have been developed and commercialized into handheld devices and products for on-field plant disease detection. This review focusses on the transition from the gold standard techniques to the advanced on-field biosensors to detect plant diseases with higher accuracy, cost-effective and making timely diagnosis possible. A growing trend for pathogen detection based on biosensors has been highlighted and further categorized into electrochemical, optical, and mass-based sensors. These innovative advancements in plant pathogen detection systems help to make the agricultural sector more safe, reliable, and sustainable for the ever-growing population.

A review of recent advances in plant-pathogen detection systems.

Worldwide, a substantial economic loss in agricultural products is caused by plant pathogens. The increased losses in agriculture have drawn attention towards the development of miniaturized pathogen detection systems for phytopathology. This review paper's main selling point supports recent research (from 2015 to 2022) and technological advancements in the field of plant pathogen detection. The article discusses in depth important developments in the loop-mediated isothermal amplification (LAMP) assay, microfluidics, Molecular Imprinted Polymer (MIP) based biosensors, digital droplet PCR (ddPCR), disposable all-printed electronics, and nanoparticle-based sensors for instantaneous pathogen detection in agricultural applications. Utilizing nanoparticles to identify agricultural pathogens is a crucial topic that is explored. A brief on various commercially available detection systems worldwide have been listed. Finally, we discuss the perspective in the development of portable miniaturized systems and novel assay technologies based on advanced nanomaterials. Gold standard techniques: Although Polymerase Chain Reaction (PCR) and culture counting have been widely used for plant pathogen detection, they are not appropriate for measurements made in the field due to their higher installation costs, lack of portability, need for well-equipped laboratories, and requirement of skilled personnel. Therefore, these recent trends are overtaking the traditional methods in Agri-diagnostics because of their superior performances and suitability for the task.

High temporal resolution transcriptomic profiling delineates distinct patterns of interferon response following Covid-19 mRNA vaccination and SARS-CoV2 infection

Knowledge of the mechanisms underpinning the development of protective immunity conferred by mRNA vaccines is fragmentary. Here we investigated responses to COVID-19 mRNA vaccination via ultra-low-volume sampling and high-temporal-resolution transcriptome profiling (23 subjects across 22 timepoints, and with 117 COVID-19 patients used as comparators). There were marked differences in the timing and amplitude of the responses to the priming and booster doses. Notably, we identified two distinct interferon signatures. The first signature (A28/S1) was robustly induced both post-prime and post-boost and in both cases correlated with the subsequent development of antibody responses. In contrast, the second interferon signature (A28/S2) was robustly induced only post-boost, where it coincided with a transient inflammation peak. In COVID19 patients, a distinct phenotype dominated by A28/S2 was associated with longer duration of intensive care. In summary, high-temporal-resolution transcriptomic permitted the identification of post- vaccination phenotypes that are determinants of the course of COVID-19 disease.

IFI27 transcription is an early predictor for COVID-19 outcomes; a multi-cohort observational study

BackgroundRobust biomarkers that predict disease outcomes amongst COVID-19 patients are necessary for both patient triage and resource prioritisation. Numerous candidate biomarkers have been proposed for COVID-19. However, at present, there is no consensus on the best diagnostic approach to predict outcomes in infected patients. Moreover, it is not clear whether such tools would apply to other potentially pandemic pathogens and therefore of use as stockpile for future pandemic preparedness. MethodsWe conducted a multi-cohort observational study to investigate the biology and the prognostic role of interferon alpha-inducible protein 27 (IFI27) in COVID-19 patients. FindingsWe show that IFI27 is expressed in the respiratory tract of COVID-19 patients and elevated IFI27 expression is associated with the presence of a high viral load. We further demonstrate that systemic host response, as measured by blood IFI27 expression, is associated with COVID-19 severity. For clinical outcome prediction (e.g. respiratory failure), IFI27 expression displays a high positive (0.83) and negative (0.95) predictive value, outperforming all other known predictors of COVID-19 severity. Furthermore, IFI27 is upregulated in the blood of infected patients in response to other respiratory viruses. For example, in the pandemic H1N1/09 swine influenza virus infection, IFI27-like genes were highly upregulated in the blood samples of severely infected patients. InterpretationThese data suggest that prognostic biomarkers targeting the family of IFI27 genes could potentially supplement conventional diagnostic tools in future virus pandemics, independent of whether such pandemics are caused by a coronavirus, an influenza virus or another as yet-to-be discovered respiratory virus. Research in contextO_ST_ABSEvidence before this studyC_ST_ABSWe searched the scientific literature using PubMed to identify studies that used the IFI27 biomarker to predict outcomes in COVID-19 patients. We used the search terms "IFI27", "COVID-19, "gene expression" and "outcome prediction". We did not identify any study that investigated the role of IFI27 biomarker in outcome prediction. Although ten studies were identified using the general terms of "gene expression" and "COVID-19", IFI27 was only mentioned in passing as one of the identified genes. All these studies addressed the broader question of the host response to COVID-19; none focused solely on using IFI27 to improve the risk stratification of infected patients in a pandemic. Added value of this studyHere, we present the findings of a multi-cohort study of the IFI27 biomarker in COVID-19 patients. Our findings show that the host response, as reflected by blood IFI27 gene expression, accurately predicts COVID-19 disease progression (positive and negative predictive values; 0.83 and 0.95, respectively), outperforming age, comorbidity, C-reactive protein and all other known risk factors. The strong association of IFI27 with disease severity occurs not only in SARS-CoV-2 infection, but also in other respiratory viruses with pandemic potential, such as the influenza virus. These findings suggest that host response biomarkers, such as IFI27, could help identify high-risk COVID-19 patients - those who are more likely to develop infection complications - and therefore may help improve patient triage in a pandemic. Implications of all the available evidenceThis is the first systemic study of the clinical role of IFI27 in the current COVID-19 pandemic and its possible future application in other respiratory virus pandemics. The findings not only could help improve the current management of COVID-19 patients but may also improve future pandemic preparedness.

A 400 mV 160 nW/Ch Compact Energy Efficient ∆Σ Modulator for Multichannel Biopotential Signal Acquisition System.

Analog to digitalconverter circuit design for biomedical systems with multiple recording channels presents challenges in high density and very low power consumption. Passive integrator and loop-filter based delta-sigma modulators (DSMs) have been recently reported for ultra-low-power and highly energy-efficient data converters for multi-channel biopotential acquisition. However, these modulators rely on a very high oversampling ratio (OSR) to achieve the target resolution. Higher OSR leads to higher power consumption in the modulator and the digital low-pass and decimation filter succeeding the DSM. We present a low OSR passive integrator-based DSM in this work by relying on a duty-cycled resistor (DCR). DCR enables the realization of large time constants in the already passive loop-filter, with minimal area and overhead power consumption. This leads to design of DSMs that are highly area, power and energy-efficient, suitable for multi-channel biopotential recording systems. We demonstrate a second order, duty-cycled passive integrator based CTDSM in a 65 nm CMOS technology for a 10 kHz biopotential bandwidth. Measurement results show that the fabricated design achieves an SNDR/DR of 56.36/63.1 dB while consuming only 160 nW power with an OSR of 32 and occupies an area of 0.035 mm 2 with a state-of-the-art energy efficiency of 14.9 fJ/conv. In-vitro and in-vivo measurements are provided to further demonstrate the operation of the proposed DSM.

Reasons for delayed spinal cord decompression in individuals with traumatic spinal cord injuries in Iran: A qualitative study from the perspective of neurosurgeons / 中华创伤杂志(英文版)

PURPOSE@#The median time from the event leading to the spinal cord injury (SCI) to the time of decompressive surgery is estimated to be 6.9 days in Iran, which is much longer than the proposed ideal time (less than 24 h) in published guidelines. The current qualitative study aimed to determine the reasons for the observed decompression surgery delay in Iran from the perspective of neurosurgeons.@*METHODS@#This qualitative study is designed to perform content analysis on the gathered data from face-to-face semi-structured interviews with 12 Iranian neurosurgeons.@*RESULTS@#The findings of the current study suggest that patient-related factors constitute more than half of the codes extracted from the interviews. Overall, the type of injury, presence of polytrauma, and surgeons' wrong attitude are the main factors causing delayed spinal cord decompression in Iranian patients from the perspective of neurosurgeons. Other notable factors include delay in transferring patients to the trauma center, delay in availability of necessary equipment, and scarce medical personnel.@*CONCLUSION@#In the perspective of neurosurgeons, the type of injury, presence of polytrauma, and surgeons' wrong attitude are the leading reasons for delayed decompressive surgery of individuals with SCI in Iran.

A 6-mRNA host response whole-blood classifier trained using patients with non-COVID-19 viral infections accurately predicts severity of COVID-19

BackgroundDetermining the severity of COVID-19 remains an unmet medical need. Our objective was to develop a blood-based host-gene-expression classifier for the severity of viral infections and validate it in independent data, including COVID-19. MethodsWe developed the classifier for the severity of viral infections and validated it in multiple viral infection settings including COVID-19. We used training data (N=705) from 21 retrospective transcriptomic clinical studies of influenza and other viral illnesses looking at a preselected panel of host immune response messenger RNAs. ResultsWe selected 6 host RNAs and trained logistic regression classifier with a cross-validation area under curve of 0.90 for predicting 30-day mortality in viral illnesses. Next, in 1,417 samples across 21 independent retrospective cohorts the locked 6-RNA classifier had an area under curve of 0.91 for discriminating patients with severe vs. non-severe infection. Next, in independent cohorts of prospectively (N=97) and retrospectively (N=100) enrolled patients with confirmed COVID-19, the classifier had an area under curve of 0.89 and 0.87, respectively, for identifying patients with severe respiratory failure or 30-day mortality. Finally, we developed a loop-mediated isothermal gene expression assay for the 6-messenger-RNA panel to facilitate implementation as a rapid assay. ConclusionsWith further study, the classifier could assist in the risk assessment of COVID-19 and other acute viral infections patients to determine severity and level of care, thereby improving patient management and reducing healthcare burden.

Predicting Early Stage Drug Induced Parkinsonism using Unsupervised and Supervised Machine Learning.

Drug Induced Parkinsonism (DIP) is the most common, debilitating movement disorder induced by antipsychotics. There is no tool available in clinical practice to effectively diagnose the symptoms at the onset of the disease. In this study, the variations in gait accelerometer data due to the intermittency of tremor at the initial stages is examined. These variations are used to train a logistic regression model to predict subjects with early-stage DIP. The logistic classifier predicts if a subject is a DIP or control with approximately 89% sensitivity and 96% specificity. This paper discusses the algorithm used to extract the features in gait data for training the classifier to predict DIP at the earliest.Clinical Relevance- Diagnosing the disease and the causative drug is vital as the physical health of a patient who is mentally unstable can deteriorate with prolonged usage of the drug. The proposed model helps clinicians to diagnose the disease at the onset of tremors with an accuracy of 93.58%.

Injectable Sensors Based on Passive Rectification of Volume-Conducted Currents.

Sensing implants that can be deployed by catheterization or by injection are preferable over implants requiring invasive surgery. However, present powering methods for active implants and present interrogation methods for passive implants require bulky parts within the implants that hinder the development of such minimally invasive devices. In this article, we propose a novel approach that potentially enables the development of passive sensing systems overcoming the limitations of previous implantable sensing systems in terms of miniaturization. In this approach implants are shaped as thread-like devices suitable for implantation by injection. Their basic structure consists of a thin elongated body with two electrodes at opposite ends and a simple and small circuit made up of a diode, a capacitor and a resistor. The interrogation method to obtain measurements from the implants consists in applying innocuous bursts of high frequency (≥1 MHz) alternating current that reach the implants by volume conduction and in capturing and processing the voltage signals that the implants produce after the bursts. As proof-of-concept, and for illustrating how to put in practice this novel approach, here we describe the development and characterization of a system for measuring the conductivity of tissues surrounding the implant. We also describe the implementation and the in vitro validation of a 0.95 mm-thick, flexible injectable implant made of off-the-shelf components. For conductivities ranging from about 0.2 to 0.8 S/m, when compared to a commercial conductivity meter, the accuracy of the implemented system was about ±10%.

Bio-WiTel: A Low-Power Integrated Wireless Telemetry System for Healthcare Applications in 401-406 MHz Band of MedRadio Spectrum.

This paper presents a low-power integrated wireless telemetry system (Bio-WiTel) for healthcare applications in 401-406 MHz frequency band of medical device radiocommunication (MedRadio) spectrum. In this paper, necessary design considerations for telemetry system for short-range (upto 3 m) communication of biosignals are presented. These considerations help greatly in making important design decisions, which eventually lead to a simple, low power, robust, and reliable wireless system implementation. Transmitter (TX) and receiver (RX) of Bio-WiTel system have been fabricated in 180 nm mixed mode CMOS technology. While radiating -18 dBm output power to a 50 antenna, the packaged TX IC consumes 250 µW power in 100% on state from 1 V supply, whereas the RX IC consumes 990 µW power from 1.8 V supply with a sensitivity of -75 dBm. Measurement results show that TX fulfils the spectral mask requirement at a maximum data rate of 72 kb/s. The measured bit error rate (BER) of RX is less than for a data rate of 200 kb/s. The proposed Bio-WiTel system is tested successfully in home and hospital environments for the communication of electrocardiogram and photoplethysmogram signals at a data rate of 57.6 kb/s with a measured BER of <10 for a maximum distance of 3 m.

Nanostructured Anodic Multilayer Dielectric Stacked Metal-Insulator-Metal Capacitors.

This paper presents the fabrication of Al2O3/TiO2/Al2O3 metal-insulator-metal (MIM) capacitor using anodization technique. High capacitance density of > 3.5 fF/µm2, low quadratic voltage coefficient of capacitance of < 115 ppm/V2 and a low leakage current density of 4.457 x 10(-11) A/cm2 at 3 V are achieved which are suitable for analog and mixed signal applications. We found that the anodization voltage played a major role in electrical and structural properties of the thin film. This work suggests that the anodization method can offer crystalline multilayer dielectric stack required for high performance MIM capacitor.

Nanostructured bilayer anodic TiO2/Al2O3 metal-insulator-metal capacitor.

This paper presents the fabrication of high performance bilayer TiO2/Al2O3 Metal-Insulator-Metal capacitor using anodization technique. A high capacitance density of 7 fF/microm2, low quadratic voltage coefficient of capacitance of 150 ppm/V2 and a low leakage current density of 9.1 nA/cm2 at 3 V are achieved which are suitable for Analog and Mixed signal applications. The influence of anodization voltage on structural and electrical properties of dielectric stack is studied in detail. At higher anodization voltages, we have observed the transformation of amorphous to crystalline state of TiO2/Al2O3 and improvement of electrical properties.

Comparing stress markers across various cohorts in a mobile setting.

There exist multiple markers for measuring psychological stress, with varying specificities and sensitivities. However, in a real-life setting, there is limited data on how robust these methods may be especially in a relatively mobile context where the signal fidelity maybe limited. Thus any large scale data to inform how these methods perform, using commonly available sensors, based on both context and cohort characterization, can greatly add to our knowledge of their respective utility in real-life settings. This paper presents a study of 253 subjects which provides crucial data for analysing various stress markers in a mobile setting. We also provide early analysis results.