Surface Reconstruction from Point Clouds: A Survey and a Benchmark.

Reconstruction of a continuous surface of two-dimensional manifold from its raw, discrete point cloud observation is a long-standing problem in computer vision and graphics research. The problem is technically ill-posed, and becomes more difficult considering that various sensing imperfections would appear in the point clouds obtained by practical depth scanning. In literature, a rich set of methods has been proposed, and reviews of existing methods are also provided. However, existing reviews are short of thorough investigations on a common benchmark. The present paper aims to review and benchmark existing methods in the new era of deep learning surface reconstruction. To this end, we contribute a large-scale benchmarking dataset consisting of both synthetic and real-scanned data; the benchmark includes object- and scene-level surfaces and takes into account various sensing imperfections that are commonly encountered in practical depth scanning. We conduct thorough empirical studies by comparing existing methods on the constructed benchmark, and pay special attention on robustness of existing methods against various scanning imperfections; we also study how different methods generalize in terms of reconstructing complex surface shapes. Our studies help identity the best conditions under which different methods work, and suggest some empirical findings. For example, while deep learning methods are increasingly popular in the research community, our systematic studies suggest that, surprisingly, a few classical methods perform even better in terms of both robustness and generalization; our studies also suggest that the practical challenges of misalignment of point sets from multi-view scanning, missing of surface points, and point outliers remain unsolved by all the existing surface reconstruction methods. We expect that the benchmark and our studies would be valuable both for practitioners and as a guidance for new innovations in future research. We make the benchmark publicly accessible at https://Gorilla-Lab-SCUT.github.io/SurfaceReconstructionBenchmark.

Cavitation is the determining mechanism for the atomization of high-viscosity liquid.

Piezoelectric atomization is becoming mainstream in the field of inhalation therapy due to its significant advantages. With the rapid development of high-viscosity gene therapy drugs, the demand for piezoelectric atomization devices is increasing. However, conventional piezoelectric atomizers with a single-dimensional energy supply are unable to provide the energy required to atomize high-viscosity liquids. To address this problem, our team has designed a flow tube internal cavitation atomizer (FTICA). This study focuses on dissecting the atomization mechanism of FTICA. In contrast to the widely supported capillary wave hypothesis, our study provides evidence in favor of the cavitation hypothesis, proving that cavitation is the key to atomizing high-viscosity liquids with FTICA. In order to prove that the cavitation is the key to atomizing in the structure of FTICA, the performance of atomization is experimented after changing the cavitation conditions by heating and stirring of the liquids.

Carnosine ameliorates postoperative cognitive dysfunction of aged rats by limiting astrocytes pyroptosis.

Postoperative cognitive dysfunction (POCD) is a common postoperative complication in elderly patients, and neuroinflammation is a key hallmark. Recent studies suggest that the NOD-like receptor family, pyrin domain containing 3 (NLRP3) inflammasome-mediated astrocytes pyroptosis is involved in the regulation of neuroinflammation in many neurocognitive diseases, while its role in POCD remains obscure. Carnosine is a natural endogenous dipeptide with anti-inflammatory and neuroprotective effects. To explore the effect of carnosine on POCD and its mechanism, we established a POCD model by exploratory laparotomy in 24-month-old male Sprague-Dawley rats. We found that the administrated of carnosine notably attenuated surgery-induced NLRP3 inflammasome activation and pyroptosis in astrocytes, central inflammation, and neuronal damage in the hippocampus of aged rats. In addition, carnosine dramatically ameliorated the learning and memory deficits of surgery-induced aged rats. Then in the in vitro experiments, we stimulated primary astrocytes with lipopolysaccharide (LPS) after carnosine pretreatment. The results also showed that the application of carnosine alleviated the activation of the NLRP3 inflammasome, pyroptosis, and inflammatory response in astrocytes stimulated by LPS. Taken together, these findings suggest that carnosine improves POCD in aged rats via inhibiting NLRP3-mediated astrocytes pyroptosis and neuroinflammation.

Method of generating speckle patterns for digital image correlation based on modified Conway's Game of Life.

The measurement accuracy of digital image correlation (DIC) is influenced by the quality of the speckle pattern. Although various models for generating random speckle patterns have been well discussed, obtaining appropriate speckle images with isotropic quality and performance could be a challenging issue in DIC. In this paper, we propose a novel (to our knowledge) method for generating speckle patterns based on modified Conway's game of life (GoL). By sequentially assembling the speckle patterns generated from the modified GoL, we produced the GoL speckle image. Then, verification and comparison experiments were conducted through pure in-plane translations. The results show that the generated speckle image which was resized with k s=6& k r=2 processing and subsequently fuzzified using a Gaussian filter, produces the best accuracy for DIC measurement. Furthermore, based on the rigid body in-plane rotation displacement tests in the physical experimental results of three different speckle images, the GoL speckle generated from our proposed method shows the smallest measurement error. This indicates that the proposed speckle patterns generating method could provide a new type of speckle pattern with better quality and accuracy.

A Mitochondria-Targeted Nitric Oxide Probe for Multimodality Imaging of Macrophage Immune Responses.

Nitric oxide (NO) is a crucial signal molecule closely linked to the biological immune response, especially in macrophage polarization. When activated, macrophages enter a pro-inflammatory state and produce NO, a marker for the M1 phenotype. In contrast, the anti-inflammatory M2 phenotype does not produce NO. We developed a mitochondria-targeted two-photon iridium-based complex (Ir-ImNO) probe that can detect endogenous NO and monitor macrophages' different immune response states using various imaging techniques, such as one- and two-photon phosphorescence imaging and phosphorescence lifetime imaging. Ir-ImNO was used to monitor the immune activation of macrophages in mice. This technology aims to provide a clear and comprehensive visualization of macrophage immune responses.

The effect of spread through air spaces on postoperative recurrence-free survival in patients with multiple primary lung cancers.

PURPOSE: The purpose of the study was to investigate the effect of spread through air spaces (STAS) on the postoperative prognosis of patients with multiple primary lung cancers staged from IA to IB based on tumor size. METHODS: Clinicopathological and follow-up data of 122 patients with multiple primary lung cancers diagnosed at stages IA-IB and surgically treated at the Department of Thoracic Surgery, Shenzhen people's Hospital from January 2019 to December 2021 were retrospectively analyzed. The study involved 42 males and 80 females. STAS status was used to divide them into two groups (87 cases in STAS (-) and 35 cases in STAS (+)). A logistic regression analysis, univariate and multivariate Cox regression analysis, and Kaplan-Meier curves (K-M) were used to determine how STAS affected recurrence-free survival (RFS) in patients. RESULTS: STAS (+) had a significantly higher recurrence rate than STAS (-). STAS was predicted by smoking history (P = 0.044), main tumor diameter (P = 0.02), and solid nodules on chest CT (P = 0.02). STAS incidence was not significantly different between lobectomy and sublobar resection groups (P = 0.17). Solid nodules on CT, tumor diameter, vascular invasion, pleural invasion, and STAS were significant predictors of recurrence in the univariate Cox regression analysis. Tumor diameter, pleural invasion and STAS were significant prognostic factors for recurrence in the multivariate Cox regression analysis. Furthermore, STAS (+) group was at greater risk of recurrence than STAS (-) group (34% vs. 0%, P < 0.05)。. CONCLUSION: Stage IA-IB multiple primary lung cancer patients with STAS (+) had a higher recurrence rate and a shorter overall survival rate.

Multi-modal learning for inpatient length of stay prediction.

Predicting inpatient length of stay (LoS) is important for hospitals aiming to improve service efficiency and enhance management capabilities. Patient medical records are strongly associated with LoS. However, due to diverse modalities, heterogeneity, and complexity of data, it becomes challenging to effectively leverage these heterogeneous data to put forth a predictive model that can accurately predict LoS. To address the challenge, this study aims to establish a novel data-fusion model, termed as DF-Mdl, to integrate heterogeneous clinical data for predicting the LoS of inpatients between hospital discharge and admission. Multi-modal data such as demographic data, clinical notes, laboratory test results, and medical images are utilized in our proposed methodology with individual "basic" sub-models separately applied to each different data modality. Specifically, a convolutional neural network (CNN) model, which we termed CRXMDL, is designed for chest X-ray (CXR) image data, two long short-term memory networks are used to extract features from long text data, and a novel attention-embedded 1D convolutional neural network is developed to extract useful information from numerical data. Finally, these basic models are integrated to form a new data-fusion model (DF-Mdl) for inpatient LoS prediction. The proposed method attains the best R2 and EVAR values of 0.6039 and 0.6042 among competitors for the LoS prediction on the Medical Information Mart for Intensive Care (MIMIC)-IV test dataset. Empirical evidence suggests better performance compared with other state-of-the-art (SOTA) methods, which demonstrates the effectiveness and feasibility of the proposed approach.

Esketamine Prevents Postoperative Emotional and Cognitive Dysfunction by Suppressing Microglial M1 Polarization and Regulating the BDNF-TrkB Pathway in Ageing Rats with Preoperative Sleep Disturbance.

Postoperative depression (POD) and postoperative cognitive dysfunction (POCD) have placed heavy burden on patients' physical and mental health in recent years. Sleep disturbance before surgery is a common phenomenon that has been increasingly believed to affect patients' recovery, especially in aged patients, while little attention has been paid to sleep disruption before surgery and the potential mechanism remains ambiguous. Ketamine has been reported to attenuate POCD after cardiac surgery and elicit rapid-acting and sustained antidepressant actions. The present study aimed to clarify the effect of esketamine's (the S-enantiomer of ketamine) protective effects and possible mechanisms of action in POCD and POD. Our results showed that sleep disturbance before surgery exacerbated microglial M1 polarization and microglial BDNF-TrkB signalling dysfunction induced by surgery, resulting in postoperative emotional changes and cognitive impairments. Notably, treatment with esketamine reversed the behavioural abnormalities through inhibiting the M1 polarization of microglia and the inflammatory response thus improving BDNF-TrkB signalling in vivo and vitro. In addition, esketamine administration also reversed the impaired hippocampal synaptic plasticity which has been perturbed by sleep disturbance and surgery. These findings warrant further investigations into the interplay of esketamine and may provide novel ideas for the implication of preoperative preparations and the prevention of postoperative brain-related complications.

A Mitochondria-Localized Iridium(III) Complex for Simultaneous Two-Photon Phosphorescence Lifetime Imaging of Downstream Products N2O3 and ONOO- of Endogenous Nitric Oxide.

Nitric oxide (NO) serves as a ubiquitous and fundamental signaling molecule involved in intricate effects on both physiological and pathological processes. NO, biosynthesized by nitric oxide synthase (NOS) or generated from nitrite, can form nitrosation reagent N2O3 (4NO + O2 = 2N2O3) through its oxidation or quickly produce peroxynitrite anion ONOO- (NO + •O2- = ONOO-) by reacting with superoxide anion (•O2-). However, most of the existing luminescent probes for NO just focus on specificity and utilize only a single signal to distinguish products N2O3 or ONOO-. In most of the present work, they differentiate one product from another simply by fluorescence signal or fluorescence intensity, which is not enough to distinguish accurately the behavior of NO in living cells. Herein, a new mitochondria-targeted and two-photon near-infrared (NIR) phosphorescent iridium(III) complex, known as Ir-NBD, has been designed for accurate detection and simultaneous imaging of two downstream products of endogenous NO, i.e., N2O3 and ONOO-. Ir-NBD exhibits a rapid response to N2O3 and ONOO- in enhanced phosphorescence intensity, increased phosphorescence lifetime, and an exceptionally high two-photon cross-section, reaching values of 78 and 85 GM, respectively, after the reaction. Furthermore, we employed multiple imaging methods, phosphorescence intensity imaging, and phosphorescence lifetime imaging together to image even distinguish N2O3 and ONOO- by probe Ir-NBD. Thus, coupled with its excellent photometrics, Ir-NBD enabled the detection of the basal level of intracellular NO accurately by responding to N2O3 and ONOO- in the lipopolysaccharide-stimulated macrophage model in virtue of fluorescence signal and phosphorescence lifetime imaging, revealing precisely the endogenous mitochondrial NO distribution during inflammation in a cell environment.

A deep learning approach for inpatient length of stay and mortality prediction.

PURPOSE: Accurate prediction of the Length of Stay (LoS) and mortality in the Intensive Care Unit (ICU) is crucial for effective hospital management, and it can assist clinicians for real-time demand capacity (RTDC) administration, thereby improving healthcare quality and service levels. METHODS: This paper proposes a novel one-dimensional (1D) multi-scale convolutional neural network architecture, namely 1D-MSNet, to predict inpatients' LoS and mortality in ICU. First, a 1D multi-scale convolution framework is proposed to enlarge the convolutional receptive fields and enhance the richness of the convolutional features. Following the convolutional layers, an atrous causal spatial pyramid pooling (SPP) module is incorporated into the networks to extract high-level features. The optimized Focal Loss (FL) function is combined with the synthetic minority over-sampling technique (SMOTE) to mitigate the imbalanced-class issue. RESULTS: On the MIMIC-IV v1.0 benchmark dataset, the proposed approach achieves the optimum R-Square and RMSE values of 0.57 and 3.61 for the LoS prediction, and the highest test accuracy of 97.73% for the mortality prediction. CONCLUSION: The proposed approach presents a superior performance in comparison with other state-of-the-art, and it can effectively perform the LoS and mortality prediction tasks.

Dense & Attention Convolutional Neural Networks for Toe Walking Recognition.

Idiopathic toe walking (ITW) is a gait disorder where children's initial contacts show limited or no heel touch during the gait cycle. Toe walking can lead to poor balance, increased risk of falling or tripping, leg pain, and stunted growth in children. Early detection and identification can facilitate targeted interventions for children diagnosed with ITW. This study proposes a new one-dimensional (1D) Dense & Attention convolutional network architecture, which is termed as the DANet, to detect idiopathic toe walking. The dense block is integrated into the network to maximize information transfer and avoid missed features. Further, the attention modules are incorporated into the network to highlight useful features while suppressing unwanted noises. Also, the Focal Loss function is enhanced to alleviate the imbalance sample issue. The proposed approach outperforms other methods and obtains a superior performance. It achieves a test recall of 88.91% for recognizing idiopathic toe walking on the local dataset collected from real-world experimental scenarios. To ensure the scalability and generalizability of the proposed approach, the algorithm is further validated through the publicly available datasets, and the proposed approach achieves an average precision, recall, and F1-Score of 89.34%, 91.50%, and 92.04%, respectively. Experimental results present a competitive performance and demonstrate the validity and feasibility of the proposed approach.

Serine-arginine protein kinase 1 (SRPK1) promotes EGFR-TKI resistance by enhancing GSK3ß Ser9 autophosphorylation independent of its kinase activity in non-small-cell lung cancer.

Resistance to epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) is a major challenge for clinicians and patients with non-small cell lung cancer (NSCLC). Serine-arginine protein kinase 1 (SRPK1) is a key oncoprotein in the EGFR/AKT pathway that participates in tumorigenesis. We found that high SRPK1 expression was significantly associated with poor progression-free survival (PFS) in patients with advanced NSCLC undergoing gefitinib treatment. Both in vitro and in vivo assays suggested that SRPK1 reduced the ability of gefitinib to induce apoptosis in sensitive NSCLC cells independently of its kinase activity. Moreover, SRPK1 facilitated binding between LEF1, ß-catenin and the EGFR promoter region to increase EGFR expression and promote the accumulation and phosphorylation of membrane EGFR. Furthermore, we verified that the SRPK1 spacer domain bound to GSK3ß and enhanced its autophosphorylation at Ser9 to activate the Wnt pathway, thereby promoting the expression of Wnt target genes such as Bcl-X. The correlation between SRPK1 and EGFR expression was confirmed in patients. In brief, our research suggested that the SRPK1/GSK3ß axis promotes gefitinib resistance by activating the Wnt pathway and may serve as a potential therapeutic target for overcoming gefitinib resistance in NSCLC.

Design and Analysis of a Cardioid Flow Tube Valveless Piezoelectric Pump for Medical Applications.

Piezoelectric pumps play an important role in modern medical technology. To improve the flow rate of valveless piezoelectric pumps with flow tube structures and promote the miniaturization and integration of their designs, a cardioid flow tube valveless piezoelectric pump (CFTVPP) is proposed in this study. The symmetric dual-bend tube design of CFTVPP holds great potential in applications such as fluid mixing and heat dissipation systems. The structure and working principle of the CFTVPP are analyzed, and flow resistance and velocity equations are established. Furthermore, the flow characteristics of the cardioid flow tube (CFT) are investigated through computational fluid dynamics, and the output performance of valveless piezoelectric pumps with different bend radii is studied. Experimental results demonstrate that CFTVPP exhibits the pumping effect, with a maximum vibration amplitude of 182.5 µm (at 22 Hz, 100 V) and a maximum output flow rate of 5.69 mL/min (at 25 Hz, 100 V). The results indicate that a smaller bend radius of the converging bend leads to a higher output flow rate, while the performance of valveless piezoelectric pumps with different diverging bends shows insignificant differences. The CFTVPP offers advantages such as a high output flow rate, low cost, small size for easy integration, and ease of manufacturing.

Weakly-supervised learning method for the recognition of potato leaf diseases.

As a crucial food crop, potatoes are highly consumed worldwide, while they are also susceptible to being infected by diverse diseases. Early detection and diagnosis can prevent the epidemic of plant diseases and raise crop yields. To this end, this study proposed a weakly-supervised learning approach for the identification of potato plant diseases. The foundation network was applied with the lightweight MobileNet V2, and to enhance the learning ability for minute lesion features, we modified the existing MobileNet-V2 architecture using the fine-tuning approach conducted by transfer learning. Then, the atrous convolution along with the SPP module was embedded into the pre-trained networks, which was followed by a hybrid attention mechanism containing channel attention and spatial attention submodules to efficiently extract high-dimensional features of plant disease images. The proposed approach outperformed other compared methods and achieved a superior performance gain. It realized an average recall rate of 91.99% for recognizing potato disease types on the publicly accessible dataset. In practical field scenarios, the proposed approach separately attained an average accuracy and specificity of 97.33% and 98.39% on the locally collected image dataset. Experimental results present a competitive performance and demonstrate the validity and feasibility of the proposed approach.

Therapeutic Drug Monitoring of Ceftazidime-Avibactam Concentrations in Carbapenem-Resistant K. pneumoniae-Infected Patients With Different Kidney Statuses.

Aims: Carbapenem-resistant K. pneumoniae (CRKP) is the most common carbapenem-resistant Enterobacteriaceae with high mortality. Ceftazidime-avibactam (CAZ-AVI) has exhibited excellent in vitro activity in vivo against CRKP. However, the efficacy of CAZ-AVI in KPC-producing CRKP-infected patients with different kidney statuses varies, such as renal insufficiency, normal renal function, and augmented renal clearance (ARC). We explored the use of therapeutic drug monitoring (TDM) to evaluate the concentration and efficacy of CAZ-AVI in CRKP-infected patients with different kidney statuses. Methods: Serum concentrations for CAZ and AVI were determined by the high-performance liquid chromatography method. Bacterial identification, routine susceptibility testing, renal function index, and others were performed in standard protocols in the hospital's clinical laboratories. Results: In the two patients with ARC, in case 1, CAZ-AVI 2.5g q6h was used with good efficacy, and the concentrations were up to the pharmacokinetics/pharmacodynamics targets. In Case 2, 2.5 g q8h was used with invalid effectiveness, and AVI Cmin was only 0.797 mg/l, which is lower than the PK/PD target. Case 3 was renal insufficiency using CAZ-AVI 1.25 q8h, and case 4 was normal renal function using 2.5 g q8h. Their concentrations were both up to the PK/PD targets. Conclusion: TDM results demonstrated that CAZ-AVI steady-state plasma concentration varies among patients with different kidney statuses, providing evidence for the utility of TDM of CAZ-AVI in individualized drug dose adjustment. ARC patients may need more CAZ-AVI daily doses than the standard dose.

Tomato Oxalyl-CoA Synthetase Degrades Oxalate and Affects Fruit Quality.

Acyl activating enzyme 3 (AAE3) encodes oxalyl-CoA synthetase involved in oxalate degradation. In this study, we investigated the role of AAE3 (SlAAE3) in the fruit quality of tomato (Solanum lycopersicum). The purified recombinant SlAAE3 protein from Escherichia coli exhibited a high activity toward oxalate, with a K m of 223.8 ± 20.03 µm and V max of 7.908 ± 0.606 µmol mg-1 protein min-1. Transient expression of SlAAE3-green fluorescent protein (GFP) fusion proteins suggests that SlAAE3 is a soluble protein without specific subcellular localization. The expression of SlAAE3 is both tissue- and development-dependent, and increased during fruit ripping. The Slaae3 knockout mutants had improved fruit quality as evidenced by the increased sugar-acid ratio and mineral nutrient content. To find the mechanism by which SlAAE3 affects fruit quality, transcriptome, and metabolome were employed on SlAAE3 over-expressed line and wide type fruits. The transcriptomic and metabolic profiles indicated that SlAAE3 in fruits mainly functions at 20 days post-anthesis (20 DPA) and mature green (MG) stages, resulting in up-regulation of amino acid derivatives, nucleotides, and derivatives, but down-regulation of lipid compounds. However, differentially expressed genes (DEGs) were mainly enriched at redox pathways. Taken together, both in vivo and in vitro results suggest that SlAAE3-encoded protein acts as an oxalyl-CoA synthetase, which also participates in redox metabolism. These data provide a further understanding of the mechanism by which SlAAE3 participates in tomato fruit quality.

PM2.5 promotes NSCLC carcinogenesis through translationally and transcriptionally activating DLAT-mediated glycolysis reprograming.

BACKGROUND: Airborne fine particulate matter (PM2.5) has been associated with lung cancer development and progression in never smokers. However, the molecular mechanisms underlying PM2.5-induced lung cancer remain largely unknown. The aim of this study was to explore the mechanisms by which PM2.5 regulated the carcinogenesis of non-small cell lung cancer (NSCLC). METHODS: Paralleled ribosome sequencing (Ribo-seq) and RNA sequencing (RNA-seq) were performed to identify PM2.5-associated genes for further study. Quantitative real time-PCR (qRT-PCR), Western blot, and immunohistochemistry (IHC) were used to determine mRNA and protein expression levels in tissues and cells. The biological roles of PM2.5 and PM2.5-dysregulated gene were assessed by gain- and loss-of-function experiments, biochemical analyses, and Seahorse XF glycolysis stress assays. Human tissue microarray analysis and 18F-FDG PET/CT scans in patients with NSCLC were used to verify the experimental findings. Polysome fractionation experiments, chromatin immunoprecipitation (ChIP), and dual-luciferase reporter assay were implemented to explore the molecular mechanisms. RESULTS: We found that PM2.5 induced a translation shift towards glycolysis pathway genes and increased glycolysis metabolism, as evidenced by increased L-lactate and pyruvate concentrations or higher extracellular acidification rate (ECAR) in vitro and in vivo. Particularly, PM2.5 enhanced the expression of glycolytic gene DLAT, which promoted glycolysis but suppressed acetyl-CoA production and enhanced the malignancy of NSCLC cells. Clinically, high expression of DLAT was positively associated with tumor size, poorer prognosis, and SUVmax values of 18F-FDG-PET/CT scans in patients with NSCLC. Mechanistically, PM2.5 activated eIF4E, consequently up-regulating the expression level of DLAT in polysomes. PM2.5 also stimulated transcription factor Sp1, which further augmented transcription activity of DLAT promoter. CONCLUSIONS: This study demonstrated that PM2.5-activated overexpression of DLAT and enhancement in glycolysis metabolism contributed to the tumorigenesis of NSCLC, suggesting that DLAT-associated pathway may be a therapeutic target for NSCLC.

Time-to-event modeling for hospital length of stay prediction for COVID-19 patients.

Providing timely patient care while maintaining optimal resource utilization is one of the central operational challenges hospitals have been facing throughout the pandemic. Hospital length of stay (LOS) is an important indicator of hospital efficiency, quality of patient care, and operational resilience. Numerous researchers have developed regression or classification models to predict LOS. However, conventional models suffer from the lack of capability to make use of typically censored clinical data. We propose to use time-to-event modeling techniques, also known as survival analysis, to predict the LOS for patients based on individualized information collected from multiple sources. The performance of six proposed survival models is evaluated and compared based on clinical data from COVID-19 patients.

An interpretable multi-task system for clinically applicable COVID-19 diagnosis using CXR.

BACKGROUND: With the emergence of continuously mutating variants of coronavirus, it is urgent to develop a deep learning model for automatic COVID-19 diagnosis at early stages from chest X-ray images. Since laboratory testing is time-consuming and requires trained laboratory personal, diagnosis using chest X-ray (CXR) is a befitting option. OBJECTIVE: In this study, we proposed an interpretable multi-task system for automatic lung detection and COVID-19 screening in chest X-rays to find an alternate method of testing which are reliable, fast and easily accessible, and able to generate interpretable predictions that are strongly correlated with radiological findings. METHODS: The proposed system consists of image preprocessing and an unsupervised machine learning (UML) algorithm for lung region detection, as well as a truncated CNN model based on deep transfer learning (DTL) to classify chest X-rays into three classes of COVID-19, pneumonia, and normal. The Grad-CAM technique was applied to create class-specific heatmap images in order to establish trust in the medical AI system. RESULTS: Experiments were performed with 15,884 frontal CXR images to show that the proposed system achieves an accuracy of 91.94% in a test dataset with 2,680 images including a sensitivity of 94.48% on COVID-19 cases, a specificity of 88.46% on normal cases, and a precision of 88.01% on pneumonia cases. Our system also produced state-of-the-art outcomes with a sensitivity of 97.40% on public test data and 88.23% on a previously unseen clinical data (1,000 cases) for binary classification of COVID-19-positive and COVID-19-negative films. CONCLUSION: Our automatic computerized evaluation for grading lung infections exhibited sensitivity comparable to that of radiologist interpretation in clinical applicability. Therefore, the proposed solution can be used as one element of patient evaluation along with gold-standard clinical and laboratory testing.