MSLP: mRNA subcellular localization predictor based on machine learning techniques.

BACKGROUND: Subcellular localization of messenger RNA (mRNAs) plays a pivotal role in the regulation of gene expression, cell migration as well as in cellular adaptation. Experiment techniques for pinpointing the subcellular localization of mRNAs are laborious, time-consuming and expensive. Therefore, in silico approaches for this purpose are attaining great attention in the RNA community. METHODS: In this article, we propose MSLP, a machine learning-based method to predict the subcellular localization of mRNA. We propose a novel combination of four types of features representing k-mer, pseudo k-tuple nucleotide composition (PseKNC), physicochemical properties of nucleotides, and 3D representation of sequences based on Z-curve transformation to feed into machine learning algorithm to predict the subcellular localization of mRNAs. RESULTS: Considering the combination of the above-mentioned features, ennsemble-based models achieved state-of-the-art results in mRNA subcellular localization prediction tasks for multiple benchmark datasets. We evaluated the performance of our method  in ten subcellular locations, covering cytoplasm, nucleus, endoplasmic reticulum (ER), extracellular region (ExR), mitochondria, cytosol, pseudopodium, posterior, exosome, and the ribosome. Ablation study highlighted k-mer and PseKNC to be more dominant than other features for predicting cytoplasm, nucleus, and ER localizations. On the other hand, physicochemical properties and Z-curve based features contributed the most to ExR and mitochondria detection. SHAP-based analysis revealed the relative importance of features to provide better insights into the proposed approach. AVAILABILITY: We have implemented a Docker container and API for end users to run their sequences on our model. Datasets, the code of API and the Docker are shared for the community in GitHub at: https://github.com/smusleh/MSLP .

The Horizontal Rain-Cell Span and Wind Impact on Multisite Diversity Scheme in a Tropical Region during El-Niño and La-Niña.

Site diversity is the most effective way to recover a signal lost during heavy downpours, especially in tropical regions since other mitigation techniques such as adaptive power control and code modulation may be unreliable during such. Duplicated links at diverse sites are deployed, and the least-attenuated signal of either site will be routed to the prime site for further operation. Since the deployment is costly, a diversity-gain model is used to estimate the appropriateness of selected sites. Diversity gain is known to depend on site-separation distance and elevation angle and, optionally, baseline angle and signal frequency, based on the region of research. In addition to these factors, the horizontal rain-cell span and the wind's impact on the gain are ongoing investigations, especially in tropical regions. This article presented the rain analysis from the year 2014 to mid-July 2017 at eight sites in the Gombak and Sepang districts of Malaysia to investigate the dependency relevancies. The rain rates were then used to predict the attenuation using the ITU-R P.618-13 rain-attenuation model, and the inter- and cross-district gain characteristics were evaluated. The observation of diurnal rain during the northeast seasons yielded that the northeast wind stimulates intense rain at locations along its direction, thus, extending the horizontal rain-cell span to 15 km distant from a host. Meanwhile, sites located at 5 km distant, slightly perpendicular to the wind direction, and from 90° to 180° from due north of the host, experience less rain. The baseline angle variation establishes nonimpact to the gain and lengthening the site-separation distance presented equal chances to the shorter span towards diversity-gain increment. The research outcome is necessary to formulate a more reliable diversity-gain model to be used in the industry.

Cardiovascular Disease Diagnosis from DXA Scan and Retinal Images Using Deep Learning.

Cardiovascular diseases (CVD) are the leading cause of death worldwide. People affected by CVDs may go undiagnosed until the occurrence of a serious heart failure event such as stroke, heart attack, and myocardial infraction. In Qatar, there is a lack of studies focusing on CVD diagnosis based on non-invasive methods such as retinal image or dual-energy X-ray absorptiometry (DXA). In this study, we aimed at diagnosing CVD using a novel approach integrating information from retinal images and DXA data. We considered an adult Qatari cohort of 500 participants from Qatar Biobank (QBB) with an equal number of participants from the CVD and the control groups. We designed a case-control study with a novel multi-modal (combining data from multiple modalities-DXA and retinal images)-to propose a deep learning (DL)-based technique to distinguish the CVD group from the control group. Uni-modal models based on retinal images and DXA data achieved 75.6% and 77.4% accuracy, respectively. The multi-modal model showed an improved accuracy of 78.3% in classifying CVD group and the control group. We used gradient class activation map (GradCAM) to highlight the areas of interest in the retinal images that influenced the decisions of the proposed DL model most. It was observed that the model focused mostly on the centre of the retinal images where signs of CVD such as hemorrhages were present. This indicates that our model can identify and make use of certain prognosis markers for hypertension and ischemic heart disease. From DXA data, we found higher values for bone mineral density, fat content, muscle mass and bone area across majority of the body parts in CVD group compared to the control group indicating better bone health in the Qatari CVD cohort. This seminal method based on DXA scans and retinal images demonstrate major potentials for the early detection of CVD in a fast and relatively non-invasive manner.

Quad-Band Polarization-Insensitive Square Split-Ring Resonator (SSRR) with an Inner Jerusalem Cross Metamaterial Absorber for Ku- and K-Band Sensing Applications.

The development of metamaterial absorbers has become attractive for various fields of application, such as sensing, detectors, wireless communication, antenna design, emitters, spatial light modulators, etc. Multiband absorbers with polarization insensitivity have drawn significant attention in microwave absorption and sensing research. In this paper, we propose a quad-band polarization-insensitive metamaterial absorber (MMA) for Ku- and K-band applications. The proposed patch comprises two square split-ring resonators (SSRR), four microstrip lines, and an inner Jerusalem cross to generate four corresponding resonances at 12.62 GHz,14.12 GHz, 17.53 GHz, and 19.91 GHz with 97%, 99.51%, 99%, and 99.5% absorption, respectively. The complex values of permittivity, permeability, refractive index, and impedance of MMA were extracted and discussed. The absorption mechanism of the designed MMA was explored by impedance matching, equivalent circuit model, as well as magnetic field and electric field analysis. The overall patch has a rotational-symmetrical structure, which plays a crucial role in acquiring the polarization-insensitive property. The design also shows stable absorption for both transverse electric (TE) and transverse magnetic (TM) modes. Its near-unity absorption and excellent sensing performance make it a potential candidate for sensing applications.

Low Profile Monopole Meander Line Antenna for WLAN Applications.

An antenna assumes a significant role in expanding the levels of communication to meet the demands of contemporary technologically based industry and private data services. In this paper, a printed compact meander line patch antenna array for wireless local-area network (WLAN) applications in the frequency span of 2.3685-2.4643 GHz is presented. The impedance matching of the antenna is generated by applying a partial rectangular-shaped ground plane backside of the meander line antenna. The proposed antenna evolved on the Rogers RT5880 substrate with a dielectric permittivity of 2.2, and the height of the substrate was 1.575 mm to accomplish the lowest possible return loss. The proposed antenna was developed to achieve particular outcomes, for example, voltage standing wave ratio (VSWR) 1.32, reflection coefficient 20 dB with a bandwidth of 94.2 MHz, a gain of 2.8 dBi, and an efficacy measurement of 97%. This antenna is appropriate for WLAN applications that utilize a 2.4 GHz resonance frequency. The overall dimensions of the antenna are 15 mm × 90.86 mm.

Slotted Monopole Patch Antenna for Microwave-Based Head Imaging Applications.

A modified monopole patch antenna for microwave-based hemorrhagic or ischemic stroke recognition is presented in this article. The designed antenna is fabricated on a cost-effective FR-4 lossy material with a 0.02 loss tangent and 4.4 dielectric constant. Its overall dimensions are 0.32 λ × 0.28 λ × 0.007 λ, where λ is the lower bandwidth 1.3 GHz frequency wavelength. An inset feeding approach is utilized to feed the antenna to reduce the input impedance (z = voltage/current). A total bandwidth (below -10 dB) of 2.4 GHz (1.3-3.7 GHz) is achieved with an effective peak gain of over 6 dBi and an efficiency of over 90%. A time-domain analysis confirms that the antenna produces minimal signal distortion. Simulated and experimental findings share a lot of similarities. Brain tissue is penetrated by the antenna to a satisfactory degree, while still exhibiting a safe specific absorption rate (SAR). The maximum SAR value measured for the head model is constrained to be equal to or below 0.1409 W/kg over the entire usable frequency band. Evaluation of theoretical and experimental evidence indicates the intended antenna is appropriate for Microwave Imaging (MWI) applications.

Radio Frequency Energy Harvesting Technologies: A Comprehensive Review on Designing, Methodologies, and Potential Applications.

Radio frequency energy harvesting (RF-EH) is a potential technology via the generation of electromagnetic waves. This advanced technology offers the supply of wireless power that is applicable for battery-free devices, which makes it a prospective alternative energy source for future applications. In addition to the dynamic energy recharging of wireless devices and a wide range of environmentally friendly energy source options, the emergence of the RF-EH technology is advantageous in facilitating various applications that require quality of service. This review highlights the abundant source of RF-EH from the surroundings sources, including nearby mobile phones, Wi-Fi, wireless local area network, broadcast television signal or DTS, and FM/AM radio signals. In contrast, the energy is captured by a receiving antenna and rectified into a working direct current voltage. This review also summarizes the power of RF-EH technology, which would provide a guideline for developing RF-EH units. The energy harvesting circuits depend on cutting-edge electrical technology to achieve significant efficiency, given that they are built to perform with considerably small current and voltage. Hence, the review includes a thorough analysis and discussion of various RF designs and their pros and cons. Finally, the latest applications of RF-EH are presented.

A Shallow U-Net Architecture for Reliably Predicting Blood Pressure (BP) from Photoplethysmogram (PPG) and Electrocardiogram (ECG) Signals.

Cardiovascular diseases are the most common causes of death around the world. To detect and treat heart-related diseases, continuous blood pressure (BP) monitoring along with many other parameters are required. Several invasive and non-invasive methods have been developed for this purpose. Most existing methods used in hospitals for continuous monitoring of BP are invasive. On the contrary, cuff-based BP monitoring methods, which can predict systolic blood pressure (SBP) and diastolic blood pressure (DBP), cannot be used for continuous monitoring. Several studies attempted to predict BP from non-invasively collectible signals such as photoplethysmograms (PPG) and electrocardiograms (ECG), which can be used for continuous monitoring. In this study, we explored the applicability of autoencoders in predicting BP from PPG and ECG signals. The investigation was carried out on 12,000 instances of 942 patients of the MIMIC-II dataset, and it was found that a very shallow, one-dimensional autoencoder can extract the relevant features to predict the SBP and DBP with state-of-the-art performance on a very large dataset. An independent test set from a portion of the MIMIC-II dataset provided a mean absolute error (MAE) of 2.333 and 0.713 for SBP and DBP, respectively. On an external dataset of 40 subjects, the model trained on the MIMIC-II dataset provided an MAE of 2.728 and 1.166 for SBP and DBP, respectively. For both the cases, the results met British Hypertension Society (BHS) Grade A and surpassed the studies from the current literature.

Review on Medical Implantable Antenna Technology and Imminent Research Challenges.

Implantable antennas are mandatory to transfer data from implants to the external world wirelessly. Smart implants can be used to monitor and diagnose the medical conditions of the patient. The dispersion of the dielectric constant of the tissues and variability of organ structures of the human body absorb most of the antenna radiation. Consequently, implanting an antenna inside the human body is a very challenging task. The design of the antenna is required to fulfill several conditions, such as miniaturization of the antenna dimension, biocompatibility, the satisfaction of the Specific Absorption Rate (SAR), and efficient radiation characteristics. The asymmetric hostile human body environment makes implant antenna technology even more challenging. This paper aims to summarize the recent implantable antenna technologies for medical applications and highlight the major research challenges. Also, it highlights the required technology and the frequency band, and the factors that can affect the radio frequency propagation through human body tissue. It includes a demonstration of a parametric literature investigation of the implantable antennas developed. Furthermore, fabrication and implantation methods of the antenna inside the human body are summarized elaborately. This extensive summary of the medical implantable antenna technology will help in understanding the prospects and challenges of this technology.

Low-Profile Slotted Metamaterial Antenna Based on Bi Slot Microstrip Patch for 5G Application.

A low-profile high-directivity, and double-negative (DNG) metamaterial-loaded antenna with a slotted patch is proposed for the 5G application. The radiated slotted arm as a V shape has been extended to provide a low-profile feature with a two-isometric view square patch structure, which accelerates the electromagnetic (EM) resonance. Besides, the tapered patch with two vertically split parabolic horns and the unit cell metamaterial expedite achieve more directive radiation. Two adjacent splits with meta units enhance the surface current to modify the actual electric current, which is induced by a substrate-isolated EM field. As a result, the slotted antenna shows a 7.14 dBi realized gain with 80% radiation efficiency, which is quite significant. The operation bandwidth is 4.27-4.40 GHz, and characteristic impedance approximately remains the same (50 Ω) to give a VSWR (voltage Standing wave ratio) of less than 2, which is ideal for the expected application field. The overall size of the antenna is 60 × 40 × 1.52 mm. Hence, it has potential for future 5G applications, like Internet of Things (IoT), healthcare systems, smart homes, etc.

Ultralow Profile, Low Passive Intermodulation, and Super-Wideband Ceiling Mount Antennas for Cellular and Public Safety Distributed Antenna Systems.

This paper presents an ultralow profile, low passive intermodulation (PIM), and super-wideband in-building ceiling mount antenna that covers both the cellular and public safety ultra high frequency (UHF) band for distributed antenna system (DAS) applications. The proposed antenna design utilizes a modified 2-D planar discone design concept that is miniaturized to fit into a small disc-shaped radome. The 2-D planar discone has an elliptical-shaped disc monopole and a bell-shaped ground plane, a stub at the shorting path, with asymmetrical structure and an additional proximity coupling patch to maximize the available electrical path to support the 350 MHz band range. The proposed design maximizes the radome area with a reduction of about 62% compared to similar concept type antennas. Besides, the proposed design exhibits an improved radiation pattern with null reduction compared to a typical dipole/monopole when lies at the horizontal plane. A prototype was manufactured to demonstrate the antenna performance. The VSWR and radiation pattern results agreed with the simulated results. The proposed antenna achieves a band ratio of 28.57:1 while covering a frequency range of 350-10,000 MHz. The measured passive intermodulation levels are better than -150 dBc (2 × 20 Watts) for 350, 700 and 1920 MHz bands.

An Octagonal Ring-shaped Parasitic Resonator Based Compact Ultrawideband Antenna for Microwave Imaging Applications.

An Ultrawideband (UWB) octagonal ring-shaped parasitic resonator-based patch antenna for microwave imaging applications is presented in this study, which is constructed with a diamond-shaped radiating patch, three octagonal, rectangular slotted ring-shaped parasitic resonator elements, and partial slotting ground plane. The main goals of uses of parasitic ring-shaped elements are improving antenna performance. In the prototype, various kinds of slots on the ground plane were investigated, and especially rectangular slots and irregular zigzag slots are applied to enhance bandwidth, gain, efficiency, and radiation directivity. The optimized size of the antenna is 29 × 24 × 1.5 mm3 by using the FR-4 substrate. The overall results illustrate that the antenna has a bandwidth of 8.7 GHz (2.80 ̶ 11.50 GHz) for the reflection coefficient S11 < -10 dB with directional radiation pattern. The maximum gain of the proposed prototype is more than 5.7 dBi, and the average efficiency over the radiating bandwidth is 75%. Different design modifications are performed to attain the most favorable outcome of the proposed antenna. However, the prototype of the proposed antenna is designed and simulated in the 3D simulator CST Microwave Studio 2018 and then effectively fabricated and measured. The investigation throughout the study of the numerical as well as experimental data explicit that the proposed antenna is appropriate for the Ultrawideband-based microwave-imaging fields.

Synthesis, Characterization and Development of Energy Harvesting Techniques Incorporated with Antennas: A Review Study.

The investigation into new sources of energy with the highest efficiency which are derived from existing energy sources is a significant research area and is attracting a great deal of interest. Radio frequency (RF) energy harvesting is a promising alternative for obtaining energy for wireless devices directly from RF energy sources in the environment. An overview of the energy harvesting concept will be discussed in detail in this paper. Energy harvesting is a very promising method for the development of self-powered electronics. Many applications, such as the Internet of Things (IoT), smart environments, the military or agricultural monitoring depend on the use of sensor networks which require a large variety of small and scattered devices. The low-power operation of such distributed devices requires wireless energy to be obtained from their surroundings in order to achieve safe, self-sufficient and maintenance-free systems. The energy harvesting circuit is known to be an interface between piezoelectric and electro-strictive loads. A modern view of circuitry for energy harvesting is based on power conditioning principles that also involve AC-to-DC conversion and voltage regulation. Throughout the field of energy conversion, energy harvesting circuits often impose electric boundaries for devices, which are important for maximizing the energy that is harvested. The power conversion efficiency (PCE) is described as the ratio between the rectifier's output DC power and the antenna-based RF-input power (before its passage through the corresponding network).

Metamaterial Cell-Based Superstrate towards Bandwidth and Gain Enhancement of Quad-Band CPW-Fed Antenna for Wireless Applications.

A multiband coplanar waveguide (CPW)-fed antenna loaded with metamaterial unit cell for GSM900, WLAN, LTE-A, and 5G Wi-Fi applications is presented in this paper. The proposed metamaterial structure is a combination of various symmetric split-ring resonators (SSRR) and its characteristics were investigated for two major axes directions at (x and y-axis) wave propagation through the material. For x-axis wave propagation, it indicates a wide range of negative refractive index in the frequency span of 2-8.5 GHz. For y-axis wave propagation, it shows more than 2 GHz bandwidth of near-zero refractive index (NZRI) property. Two categories of the proposed metamaterial plane were applied to enhance the bandwidth and gain. The measured reflection coefficient (S11) demonstrated significant bandwidths increase at the upper bands by 4.92-6.49 GHz and 3.251-4.324 GHz, considered as a rise of 71.4% and 168%, respectively, against the proposed antenna without using metamaterial. Besides being high bandwidth achieving, the proposed antenna radiates bi-directionally with 95% as the maximum radiation efficiency. Moreover, the maximum measured gain reaches 6.74 dBi by a 92.57% improvement compared with the antenna without using metamaterial. The simulation and measurement results of the proposed antenna show good agreement.

Compact Ultra-Wideband Monopole Antenna Loaded with Metamaterial.

A printed compact monopole antenna based on a single negative (SNG) metamaterial is proposed for ultra-wideband (UWB) applications. A low-profile, key-shaped structure forms the radiating monopole and is loaded with metamaterial unit cells with negative permittivity and more than 1.5 GHz bandwidth of near-zero refractive index (NZRI) property. The antenna offers a wide bandwidth from 3.08 to 14.1 GHz and an average gain of 4.54 dBi, with a peak gain of 6.12 dBi; this is in contrast to the poor performance when metamaterial is not used. Moreover, the maximum obtained radiation efficiency is 97%. A reasonable agreement between simulation and experiments is realized, demonstrating that the proposed antenna can operate over a wide bandwidth with symmetric split-ring resonator (SSRR) metamaterial structures and compact size of 14.5 × 22 mm2 (0.148 λ0 × 0.226 λ0) with respect to the lowest operating frequency.

Left-Handed Metamaterial-Inspired Unit Cell for S-Band Glucose Sensing Application.

This paper presents an oval-shaped sensor design for the measurement of glucose concentration in aqueous solution. This unit cell sensing device is inspired by metamaterial properties and is analytically described for better parametric study. The mechanism of the sensor is a sensing layer with varying permittivity placed between two nozzle-shaped microstrip lines. Glucose aqueous solutions were characterized considering the water dielectric constant, from 55 to 87, and were identified with a transmission coefficient at 3.914 GHz optimal frequency with double negative (DNG) metamaterial properties. Consequently, the sensitivity of the sensor was estimated at 0.037 GHz/(30 mg/dL) glucose solution. The design and analysis of this sensor was performed using the finite integration technique (FIT)-based Computer Simulation Technology (CST) microwave studio simulation software. Additionally, parametric analysis of the sensing characteristics was conducted using experimental verification for the justification. The performance of the proposed sensor demonstrates the potential application scope for glucose level identification in aqueous solutions regarding qualitative analysis.

A Compact UWB Antenna with Independently Controllable Notch Bands.

A minimally-sized, triple-notched band ultra-wideband (UWB) antenna, useful for many applications, is designed, analyzed, and experimentally validated in this paper. A modified maple leaf-shaped main radiating element with partial ground is used in the proposed design. An E-shaped resonator, meandered slot, and U-shaped slot are implemented in the proposed design to block the co-existing bands. The E-shaped resonator stops frequencies ranging from 1.8⁻2.3 GHz (Advanced Wireless System (AWS1⁻AWS2) band), while the meandered slot blocks frequencies from 3.2⁻3.8 GHz (WiMAX band). The co-existing band ranging from 5.6⁻6.1 GHz (IEEE 802.11/HIPERLANband) is blocked by utilizing the U-shaped section in the feeding network. The notched bands can be independently controlled over a wide range of frequencies using specific parameters. The proposed antenna is suitable for many applications because of its flat gain, good radiation characteristics at both principal planes, uniform group delay, and non-varying transfer function ( S 21 ) for the entire UWB frequency range.

Performance Analysis of a Defected Ground-Structured Antenna Loaded with Stub-Slot for 5G Communication.

In this paper, a defected ground-structured antenna with a stub-slot configuration is proposed for future 5G wireless applications. A simple stub-slot configuration is used in the patch antenna to get the dual band frequency response in the 5G mid-band and the upper unlicensed frequency region. Further, a 2-D double period Electronic band gap (EBG) structure has been implemented as a defect in the metallic ground plane to get a wider impedance bandwidth. The size of the slots and their positions are optimized to get a considerably high impedance bandwidth of 12.49% and 4.49% at a passband frequency of 3.532 GHz and 6.835 GHz, respectively. The simulated and measured realized gain and reflection coefficients are in good agreement for both operating bandwidths. The overall antenna structure size is 33.5 mm × 33.5 mm. The antenna is fabricated and compared with experimental results. The proposed antenna shows a stable radiation pattern and high realized gain with wide impedance bandwidth using the EBG structure, which are necessary for the requirements of IoT applications offered by 5G technology.

Circularly Polarized Broadband Printed Antenna for Wireless Applications.

A simple, compact sickle-shaped printed antenna with a slotted ground plane is designed and developed for broadband circularly polarized (CP) radiation. The sickle-shaped radiator with a tapered feed line and circular slotted square ground plane are utilized to realize the wideband CP radiation feature. With optimized dimensions of 0.29λ × 0.29λ × 0.012λ at 2.22 GHz frequency for the realized antenna parameters, the measured results display that the antenna has a 10 dB impedance bandwidth of 7.70 GHz (126.85%; 2.22⁻9.92 GHz) and a 3 dB axial ratio (AR) bandwidth of 2.64 GHz (73.33%; 2.28⁻4.92 GHz). The measurement agrees well with simulation, which proves an excellent circularly polarized property. For verification, the mechanism of band improvement and circular polarization are presented, and the parametric study is carried out. Since, the proposed antenna is a simple design structure with broad impedance and AR bandwidth, which is a desirable feature as a candidate for various wireless communication systems. Because of the easy printed structure and scaling the dimension with broadband CP characteristics, the realized antenna does incorporate in a number of CP wireless communication applications.