Auscultation-Based Pulmonary Disease Detection through Parallel Transformation and Deep Learning.

Respiratory diseases are among the leading causes of death, with many individuals in a population frequently affected by various types of pulmonary disorders. Early diagnosis and patient monitoring (traditionally involving lung auscultation) are essential for the effective management of respiratory diseases. However, the interpretation of lung sounds is a subjective and labor-intensive process that demands considerable medical expertise, and there is a good chance of misclassification. To address this problem, we propose a hybrid deep learning technique that incorporates signal processing techniques. Parallel transformation is applied to adventitious respiratory sounds, transforming lung sound signals into two distinct time-frequency scalograms: the continuous wavelet transform and the mel spectrogram. Furthermore, parallel convolutional autoencoders are employed to extract features from scalograms, and the resulting latent space features are fused into a hybrid feature pool. Finally, leveraging a long short-term memory model, a feature from the latent space is used as input for classifying various types of respiratory diseases. Our work is evaluated using the ICBHI-2017 lung sound dataset. The experimental findings indicate that our proposed method achieves promising predictive performance, with average values for accuracy, sensitivity, specificity, and F1-score of 94.16%, 89.56%, 99.10%, and 89.56%, respectively, for eight-class respiratory diseases; 79.61%, 78.55%, 92.49%, and 78.67%, respectively, for four-class diseases; and 85.61%, 83.44%, 83.44%, and 84.21%, respectively, for binary-class (normal vs. abnormal) lung sounds.

Intelligent Reflecting Surface-Based Non-LOS Human Activity Recognition for Next-Generation 6G-Enabled Healthcare System.

Human activity monitoring is a fascinating area of research to support autonomous living in the aged and disabled community. Cameras, sensors, wearables, and non-contact microwave sensing have all been suggested in the past as methods for identifying distinct human activities. Microwave sensing is an approach that has lately attracted much interest since it has the potential to address privacy problems caused by cameras and discomfort caused by wearables, especially in the healthcare domain. A fundamental drawback of the current microwave sensing methods such as radar is non-line-of-sight and multi-floor environments. They need precise and regulated conditions to detect activity with high precision. In this paper, we have utilised the publicly available online database based on the intelligent reflecting surface (IRS) system developed at the Communications, Sensing and Imaging group at the University of Glasgow, UK (references 39 and 40). The IRS system works better in the multi-floor and non-line-of-sight environments. This work for the first time uses algorithms such as support vector machine Bagging and Decision Tree on the publicly available IRS data and achieves better accuracy when a subset of the available data is considered along specific human activities. Additionally, the work also considers the processing time taken by the classier in training stage when exposed to the IRS data which was not previously explored.

Multiple Participants' Discrete Activity Recognition in a Well-Controlled Environment Using Universal Software Radio Peripheral Wireless Sensing.

Wireless sensing is the utmost cutting-edge way of monitoring different health-related activities and, concurrently, preserving most of the privacy of individuals. To meet future needs, multi-subject activity monitoring is in demand, whether it is for smart care centres or homes. In this paper, a smart monitoring system for different human activities is proposed based on radio-frequency sensing integrated with ensemble machine learning models. The ensemble technique can recognise a wide range of activity based on alterations in the wireless signal's Channel State Information (CSI). The proposed system operates at 3.75 GHz, and up to four subjects participated in the experimental study in order to acquire data on sixteen distinct daily living activities: sitting, standing, and walking. The proposed methodology merges subject count and performed activities, resulting in occupancy count and activity performed being recognised at the same time. To capture alterations owing to concurrent multi-subject motions, the CSI amplitudes collected from 51 subcarriers of the wireless signals were processed and merged. To distinguish multi-subject activity, a machine learning model based on an ensemble learning technique was designed and trained using the acquired CSI data. For maximum activity classes, the proposed approach attained a high average accuracy of up to 98%. The presented system has the ability to fulfil prospective health activity monitoring demands and is a viable solution towards well-being tracking.

Machine learning empowered COVID-19 patient monitoring using non-contact sensing: An extensive review.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which caused the coronavirus disease 2019 (COVID-19) pandemic, has affected more than 400 million people worldwide. With the recent rise of new Delta and Omicron variants, the efficacy of the vaccines has become an important question. The goal of various studies has been to limit the spread of the virus by utilizing wireless sensing technologies to prevent human-to-human interactions, particularly for healthcare workers. In this paper, we discuss the current literature on invasive/contact and non-invasive/non-contact technologies (including Wi-Fi, radar, and software-defined radio) that have been effectively used to detect, diagnose, and monitor human activities and COVID-19 related symptoms, such as irregular respiration. In addition, we focused on cutting-edge machine learning algorithms (such as generative adversarial networks, random forest, multilayer perceptron, support vector machine, extremely randomized trees, and k-nearest neighbors) and their essential role in intelligent healthcare systems. Furthermore, this study highlights the limitations related to non-invasive techniques and prospective research directions.

Machine learning empowered COVID-19 patient monitoring using non-contact sensing:An extensive review / 药物分析学报

The severe acute respiratory syndrome coronavirus 2(SARS-CoV-2),which caused the coronavirus disease 2019(COVID-19)pandemic,has affected more than 400 million people worldwide.With the recent rise of new Delta and Omicron variants,the efficacy of the vaccines has become an important question.The goal of various studies has been to limit the spread of the virus by utilizing wireless sensing technologies to prevent human-to-human interactions,particularly for healthcare workers.In this paper,we discuss the current literature on invasive/contact and non-invasive/non-contact technologies(including Wi-Fi,radar,and software-defined radio)that have been effectively used to detect,diagnose,and monitor human activities and COVID-19 related symptoms,such as irregular respiration.In addition,we focused on cutting-edge machine learning algorithms(such as generative adversarial networks,random forest,multilayer perceptron,support vector machine,extremely randomized trees,and k-nearest neighbors)and their essential role in intelligent healthcare systems.Furthermore,this study highlights the limitations related to non-invasive techniques and prospective research directions.

CAFD: Context-Aware Fault Diagnostic Scheme towards Sensor Faults Utilizing Machine Learning.

Sensors' existence as a key component of Cyber-Physical Systems makes it susceptible to failures due to complex environments, low-quality production, and aging. When defective, sensors either stop communicating or convey incorrect information. These unsteady situations threaten the safety, economy, and reliability of a system. The objective of this study is to construct a lightweight machine learning-based fault detection and diagnostic system within the limited energy resources, memory, and computation of a Wireless Sensor Network (WSN). In this paper, a Context-Aware Fault Diagnostic (CAFD) scheme is proposed based on an ensemble learning algorithm called Extra-Trees. To evaluate the performance of the proposed scheme, a realistic WSN scenario composed of humidity and temperature sensor observations is replicated with extreme low-intensity faults. Six commonly occurring types of sensor fault are considered: drift, hard-over/bias, spike, erratic/precision degradation, stuck, and data-loss. The proposed CAFD scheme reveals the ability to accurately detect and diagnose low-intensity sensor faults in a timely manner. Moreover, the efficiency of the Extra-Trees algorithm in terms of diagnostic accuracy, F1-score, ROC-AUC, and training time is demonstrated by comparison with cutting-edge machine learning algorithms: a Support Vector Machine and a Neural Network.

Wireless Channel Modelling for Identifying Six Types of Respiratory Patterns With SDR Sensing and Deep Multilayer Perceptron.

Contactless or non-invasive technology has a significant impact on healthcare applications such as the prediction of COVID-19 symptoms. Non-invasive methods are essential especially during the COVID-19 pandemic as they minimise the burden on healthcare personnel. One notable symptom of COVID-19 infection is a rapid respiratory rate, which requires constant real-time monitoring of respiratory patterns. In this paper, Software Defined Radio (SDR) based Radio-Frequency sensing technique and supervised machine learning algorithm is employed to provide a platform for detecting and monitoring various respiratory: eupnea, biot, bradypnea, sighing, tachypnea, and kussmaul. The variations in Channel State Information produced by human respiratory were utilised to identify distinct respiratory patterns using fine-grained Orthogonal Frequency-Division Multiplexing signals. The proposed platform based on the SDR and the Deep Multilayer Perceptron classifier exhibits the ability to effectively detect and classify the afore-mentioned distinct respiratory with an accuracy of up to 99%. Moreover, the effectiveness of the proposed scheme in terms of diagnosis accuracy, precision, recall, F1-score, and confusion matrix is demonstrated by comparison with a state-of-the-art machine learning classifier: Random Forest.

DFT modeling of thermoelectric and optical features of novel MgxSn1-xSe (x = 6%, 12% & 18%).

We explore influence of Mg alloying effect on electronic band structure dispersion and thermoelectric properties of tin chalcogenide materials. Based on density functional theory (DFT) within a framework of full potential linearized augmented plane wave method (FP-LAPW), we evaluate the energy band structure and optical properties of MgxSn1-xSe (x = 6%, 12% and 18%) materials. Moreover, we extend our calculations to simulate the electrical transport properties using Boltzmann transport theory. Within the approximations employed in our calculations the theoretically predicted band energy gap values and the temperature dependence of electrical transport properties of MgxSn1-xSe compounds revealed that the Mg-alloying have enhanced thermoelectric features. To verify the quality of calculations the comparison with the experimental absorption spectra are presented. The better thermoelectric performance in MgxSn1-xSe is expected to occur for all doping concentrations, however 18% Mg-doped material exhibits higher value of Seebeck coefficient and lower thermal conductivity which is suggestive that at higher Mg concentration the holes become dominant over electrons and hence make these materials to be more promising candidates for their use in thermoelectric power generation and in cooling devices.

Assessing climate change impacts on pearl millet under arid and semi-arid environments using CSM-CERES-Millet model.

Climate change adversely affects food security all over the world, especially in developing countries where the increasing population is confronting food insecurity and malnutrition. Crop models can assist stakeholders for assessment of climate change in current and future agricultural production systems. The aim of this study was to use of system analysis approach through CSM-CERES-Millet model to quantify climate change and its impact on pearl millet under arid and semi-arid climatic conditions of Punjab, Pakistan. Calibration and evaluation of CERES-Millet were performed with the field observations for pearl millet hybrid 86M86. Mid-century (2040-2069) climate change scenarios for representative concentration pathway (RCP) 4.5 and RCP 8.5 were generated based on an ensemble of selected five general circulation models (GCMs). The model was calibrated with optimum treatment (15-cm plant spacing and 200 kg N ha-1) using field observations on phenology, growth and grain yield. Thereafter, pearl millet cultivar was evaluated with remaining treatments of plant spacing and nitrogen during 2015 and 2016 in Faisalabad and Layyah. The CERES-Millet model was calibrated very well and predicted the grain yield with 1.14% error. Model valuation results showed that there was a close agreement between the observed and simulated values of grain yield with RMSE ranging from 172 to 193 kg ha-1. The results of future climate scenarios revealed that there would be an increase in Tmin (2.8 °C and 2.9 °C, respectively, for the semi-arid and arid environment) and Tmax (2.5 °C and 2.7 °C, respectively, for the semi-arid and arid environment) under RCP4.5. For RCP8.5, there would be an increase of 4 °C in Tmin for the semi-arid and arid environment and an increase of 3.7 °C and 3.9 °C in Tmax, respectively, for the semi-arid and arid environment. The impacts of climate changes showed that pearl millet yield would be reduced by 7 to 10% under RCPs 4.5 and 8.5 in Faisalabad and 10 to 13% in Layyah under RCP 4.5 and 8.5 for mid-century. So, CSM-CERES-Millet is a useful tool in assessing the climate change impacts.

Potential impacts of climate change and adaptation strategies for sunflower in Pakistan.

Growth, development, and economic yield of agricultural crops rely on moisture, temperature, light, and carbon dioxide concentration. However, the amount of these parameters is varying with time due to climate change. Climate change is factual and ongoing so, first principle of agronomy should be to identify climate change potential impacts and adaptation measures to manage the susceptibilities of agricultural sector. Crop models have ability to predict the crop's yield under changing climatic conditions. We used OILCROP-SUN model to simulate the influence of elevated temperature and CO2 on crop growth duration, maximum leaf area index (LAI), total dry matter (TDM), and achene yield of sunflower under semi-arid conditions of Pakistan (Faisalabad, Punjab). The model was calibrated and validated with the experimental data of 2012 and 2013, respectively. The simulation results showed that phenological events of sunflower were not changed at higher concentration of CO2 (430 and 550 ppm). However LAI, achene yield, and TDM increased by 0.24, 2.41, and 4.67% at 430 ppm and by 0.48, 3.09, and 9.87% at 550 ppm, respectively. Increased temperature (1 and 2 °C) reduced the sunflower duration to remain green that finally led to less LAI, achene yield, and TDM as compared to present conditions. However, the drastic effects of increased temperature on sunflower were reduced to some extent at 550 ppm CO2 concentration. Evaluation of different adaptation options revealed that 21 days earlier (as compared to current sowing date) planting of sunflower crop with increased plant population (83,333 plants ha-1) could reduce the yield losses due to climate change. Flowering is the most critical stage of sunflower to water scarcity. We recommended skipping second irrigation or 10% (337.5 mm) less irrigation water application to conserve moisture under possible water scarce conditions of 2025 and 2050.

Nitrogen and plant population change radiation capture and utilization capacity of sunflower in semi-arid environment.

The combination of nitrogen and plant population expresses the spatial distribution of crop plants. The spatial distribution influences canopy structure and development, radiation capture, accumulated intercepted radiation (Sa), radiation use efficiency (RUE), and subsequently dry matter production. We hypothesized that the sunflower crop at higher plant populations and nitrogen (N) rates would achieve early canopy cover, capture more radiant energy, utilize radiation energy more efficiently, and ultimately increase economic yield. To investigate the above hypothesis, we examined the influences of leaf area index (LAI) at different plant populations (83,333, 66,666, and 55,555 plants ha-1) and N rates (90, 120, and 150 kg ha-1) on radiation interception (Fi), photosynthetically active radiation (PAR) accumulation (Sa), total dry matter (TDM), achene yield (AY), and RUE of sunflower. The experimental work was conducted during 2012 and 2013 on sandy loam soil in Punjab, Pakistan. The sunflower crop captured more than 96% of incident radiant energy (mean of all treatments), 98% with a higher plant population (83,333 plants ha-1), and 97% with higher N application (150 kg ha-1) at the fifth harvest (60 days after sowing) during both study years. The plant population of 83,333 plants ha-1 with 150 kg N ha-1 ominously promoted crop, RUE, and finally productivity of sunflower (AY and TDM). Sunflower canopy (LAI) showed a very close and strong association with Fi (R 2 = 0.99 in both years), PAR (R 2 = 0.74 and 0.79 in 2012 and 2013, respectively), TDM (R 2 = 0.97 in 2012 and 0.91 in 2013), AY (R 2 = 0.95 in both years), RUE for TDM (RUETDM) (R 2 = 0.63 and 0.71 in 2012 and 2013, respectively), and RUE for AY (RUEAY) (R 2 = 0.88 and 0.87 in 2012 and 2013, respectively). Similarly, AY (R 2 = 0.73 in 2012 and 0.79 in 2013) and TDM (R 2 = 0.75 in 2012 and 0.84 in 2013) indicated significant dependence on PAR accumulation of sunflower. High temperature during the flowering stage in 2013 shortened the crop maturity duration, which reduced the LAI, leaf area duration (LAD), crop growth rate (CGR), TDM, AY, Fi, Sa, and RUE of sunflower. Our results clearly revealed that RUE was enhanced as plant population and N application rates were increased and biomass assimilation in semi-arid environments varied with radiation capture capacity of sunflower.

Optimizing irrigation and nitrogen for wheat through empirical modeling under semi-arid environment.

Nitrogen fertilizer availability to plants is strongly linked with water availability. Excessive or insufficient use of nitrogen can cause reduction in grain yield of wheat and environmental issues. The per capita per annum water availability in Pakistan has reduced to less than 1000 m3 and is expected to reach 800 m3 during 2025. Irrigating crops with 3 or more than 3 in. of depth without measuring volume of water is not a feasible option anymore. Water productivity and economic return of grain yield can be improved by efficient management of water and nitrogen fertilizer. A study was conducted at post-graduate agricultural research station, University of Agriculture Faisalabad, during 2012-2013 and 2013-2014 to optimize volume of water per irrigation and nitrogen application. Split plot design with three replications was used to conduct experiment; four irrigation levels (I300 = 300 mm, I240 = 240 mm, I180 = 180 mm, I120 = 120 mm for whole growing season at critical growth stages) and four nitrogen levels (N60 = 60 kg ha-1, N120 = 120 kg ha-1, N180 = 180 kg ha-1, and N240 = 240 kg ha-1) were randomized as main and sub-plot factors, respectively. The recorded data on grain yield was used to develop empirical regression models. The results based on quadratic equations and economic analysis showed 164, 162, 158, and 107 kg ha-1 nitrogen as economic optimum with I300, I240, I180, and I120 mm water, respectively, during 2012-2013. During 2013-2014, quadratic equations and economic analysis showed 165, 162, 161, and 117 kg ha-1 nitrogen as economic optimum with I300, I240, I180, and I120 mm water, respectively. The optimum irrigation level was obtained by fitting economic optimum nitrogen as function of total water. Equations predicted 253 mm as optimum irrigation water for whole growing season during 2012-2013 and 256 mm water as optimum for 2013-2014. The results also revealed that reducing irrigation from I300 to I240 mm during 2012-2013 and 2013-2014 did not reduce crop yield significantly (P < 0.01). The excessive nitrogen application ranged from 31.2 to 55.4% at N180 and N240 kg ha-1 for different levels of irrigation. It is concluded from study that irrigation and nitrogen relationship can be used for efficient management of irrigation and nitrogen and to reduce nitrogen losses. The empirical equations developed in this study can help farmers of semi-arid environment to calculate optimum level of irrigation and nitrogen for maximum economic return from wheat.