In the past few months, several works were published in regards to the dynamics and early detection of COVID-19 via mathematical modeling and Artificial intelligence (AI). The aim of this work is to provide the research community with comprehensive overview of the methods used in these studies as well as a compendium of available open source datasets in regards to COVID-19. In all, 61 journal articles, reports, fact sheets, and websites dealing with COVID-19 were studied and reviewed. It was found that most mathematical modeling done were based on the Susceptible-Exposed-Infected-Removed (SEIR) and Susceptible-infected-recovered (SIR) models while most of the AI implementations were Convolutional Neural Network (CNN) on X-ray and CT images. In terms of available datasets, they include aggregated case reports, medical images, management strategies, healthcare workforce, demography, and mobility during the outbreak. Both Mathematical modeling and AI have both shown to be reliable tools in the fight against this pandemic. Several datasets concerning the COVID-19 have also been collected and shared open source. However, much work is needed to be done in the diversification of the datasets. Other AI and modeling applications in healthcare should be explored in regards to this COVID-19.

Current sources are widely used in bio-impedance spectroscopy (BIS) measurement systems to maximize current injection for increased signal to noise while keeping within medical safety specifications. High-performance current sources based on the Howland current pump with optimized impedance converters are able to minimize stray capacitance of the cables and setup. This approach is limited at high frequencies primarily due to the deteriorated output impedance of the constant current source when situated in a real measurement system. For this reason, voltage sources have been suggested, but they require a current sensing resistor, and the SNR reduces at low impedance loads due to the lower current required to maintain constant voltage. In this paper, we compare the performance of a current source-based BIS and a voltage source-based BIS, which use common components. The current source BIS is based on a Howland current pump and generalized impedance converters to maintain a high output impedance of more than 1 MΩ at 2 MHz. The voltage source BIS is based on voltage division between an internal current sensing resistor (Rs) and an external sample. To maintain high SNR, Rs is varied so that the source voltage is divided more or less equally. In order to calibrate the systems, we measured the transfer function of the BIS systems with several known resistor and capacitor loads. From this we may estimate the resistance and capacitance of biological tissues using the least-squares method to minimize error between the measured transimpedance excluding the system transfer function and that from an impedance model. When tested on realistic loads including discrete resistors and capacitors, and saline and agar phantoms, the voltage source-based BIS system had a wider bandwidth of 10 Hz to 2.2 MHz with less than 1% deviation from the expected spectra compared to more than 10% with the current source. The voltage source also showed an SNR of at least 60 dB up to 2.2 MHz in comparison to the current source-based BIS system where the SNR drops below 40 dB for frequencies greater than 1 MHz.

When we use a conductive fabric as a pressure sensor, it is necessary to quantitatively understand its electromechanical property related with the applied pressure. We investigated electromechanical properties of three different conductive fabrics using the electrical impedance spectroscopy (EIS). We found that their electrical impedance spectra depend not only on the electrical properties of the conductive yarns, but also on their weaving structures. When we apply a mechanical tension or compression, there occur structural deformations in the conductive fabrics altering their apparent electrical impedance spectra. For a stretchable conductive fabric, the impedance magnitude increased or decreased under tension or compression, respectively. For an almost non-stretchable conductive fabric, both tension and compression resulted in decreased impedance values since the applied tension failed to elongate the fabric. To measure both tension and compression separately, it is desirable to use a stretchable conductive fabric. For any conductive fabric chosen as a pressure-sensing material, its resistivity under no loading conditions must be carefully chosen since it determines a measurable range of the impedance values subject to different amounts of loadings. We suggest the EIS method to characterize the electromechanical property of a conductive fabric in designing a thin and flexible fabric pressure sensor.

Purpose – The aim of this paper is to introduce and to evaluate the performance of a multiple frequency complex impedance reconstruction for fabric-based EIT pressure sensor. Pressure mapping is an important and challenging area of modern sensing technology. It has many applications in areas such as artificial skins in Robotics and pressure monitoring on soft tissue in biomechanics. Fabric-based sensors are being developed in conjunction with electrical impedance tomography (EIT) for pressure mapping imaging. This is potentially a very cost-effective pressure mapping imaging solution in particular for imaging large areas. Fabric-based EIT pressure sensors aim to provide a pressure mapping image using current carrying and voltage sensing electrodes attached on the boundary of the fabric patch. Design/methodology/approach – Recently, promising results are being achieved in conductivity imaging for these sensors. However, the fabric structure presents capacitive behaviour that could also be exploited for pressure mapping imaging. Complex impedance reconstructions with multiple frequencies are implemented to observe both conductivity and permittivity changes due to the pressure applied to the fabric sensor. Findings – Experimental studies on detecting changes of complex impedance on fabric-based sensor are performed. First, electrical impedance spectroscopy on a fabric-based sensor is performed. Secondly, the complex impedance tomography is carried out on fabric and compared with traditional EIT tank phantoms. Quantitative image quality measures are used to evaluate the performance of a fabric-based sensor at various frequencies and against the tank phantom. Originality/value – The paper demonstrates for the first time the useful information on pressure mapping imaging from the permittivity component of fabric EIT. Multiple frequency EIT reconstruction reveals spectral behaviour of the fabric-based EIT, which opens up new opportunities in exploration of these sensors.