Accurate interpretation of chest radiographs is central to the early diagnosis and management of pulmonary disorders. This study introduces an explainable deep learning framework that integrates biomedical signal fidelity analysis with transfer learning to enhance diagnostic reliability and transparency. Using the publicly available COVID-19 Radiography Dataset (21,165 chest X-ray images across four classes: COVID-19, Viral Pneumonia, Lung Opacity, and Normal), three architectures, namely baseline Convolutional Neural Network (CNN), ResNet-50, and EfficientNetB3, were trained and evaluated under varied class-balancing and hyperparameter configurations. Signal preservation was quantitatively verified using the Structural Similarity Index Measure (SSIM = 0.93 ± 0.02), ensuring that preprocessing retained key diagnostic features. Among all models, ResNet-50 achieved the highest classification accuracy (93.7%) and macro-AUC = 0.97 (class-balanced), whereas EfficientNetB3 demonstrated superior generalization with reduced parameter overhead. Gradient-weighted Class Activation Mapping (Grad-CAM) visualizations confirmed anatomically coherent activations aligned with pathological lung regions, substantiating clinical interpretability. The integration of signal fidelity metrics with explainable deep learning presents a reproducible and computationally efficient framework for medical image analysis. These findings highlight the potential of signal-aware transfer learning to support reliable, transparent, and resource-efficient diagnostic decision-making in radiology and other imaging-based medical domains.

It is unclear whether codetections of viruses in respiratory infections are associated with worse outcomes. We aimed to establish the association between viral codetections and severity in children under 5 years of age hospitalized with suspected viral pneumonia. We designed a case-control study. The population comprised children hospitalized at a tertiary-level institution between September 2019 and June 2022 with clinical suspicion of pneumonia and detection of at least 1 respiratory virus by nested Multiplex polymerase chain reaction (FilmArray Respiratory Panel, Biomerieux). The cases were children between 1 and 60 months of age who met the severity criteria: complicated pneumonia, intrapulmonary complications, vasoactive requirement or ventilatory support or death. Controls were patients who did not meet these criteria. We performed multivariate analysis using binary logistic regression models to calculate adjusted ORs (ORa) for relevant variables and potential confounders. We included 561 patients; 224 (39.9%) were infants, of whom 175 patients (31.2%) required intensive care and 1 died. Codetection was found in 223 children (39.8%). In the multivariate analysis, being younger than 1 year [ORa = 2.8; 95% confidence interval (CI): 1.8-4.5], and having a prior bronchiolitis (ORa = 3.6; 95% CI: 1.4-8.9), but not codetections (ORa = 1.4; 95% CI: 0.9-2.0), were independently associated with a greater severity. However, codetection was the only associated variable with worse severity in the subgroup of infants (ORa = 2.7, 95% CI: 1.1-6.8). The detection of 2 or more viruses in children <1 year of age with suspected viral pneumonia is associated with more severe disease.

Pneumonia is a respiratory disorder that involves inflammation of the air sacs of the lungs and is normally diagnosed through imaging of the chest using X-rays. This research proposes a deep learning-based classification system to classify chest X-ray images and identify pneumonia. The methodological framework adopted is the Cross-Industry Standard Process for Data Mining (CRISP-DM), in which experiments are run in a Jupyter Notebook using five-fold cross-validation. The data is made up of anterior-posterior chest X-ray photographs of children aged between one and five years of age at Guangzhou Women and Children Medical Center. A number of convolutional neural network models are assessed and compared with the proposed Improved DenseNet (ImDenseNet), including DenseNet, VGG16, and InceptionNet. According to experimental findings, ImDenseNet achieves 96.15% accuracy, 92.86% precision, and 92.94% recall, which are significantly better than those of the base models. The results show that the proposed architectural improvements improve feature discrimination and classification performance, and ImDenseNet is a credible solution for detecting pneumonia using chest X-ray images. Future research could focus on expanding ImDenseNet into a multi-class classifier to differentiate between bacterial and viral pneumonia. Pruning and quantization techniques can also be used to optimize the model for lightweight deployment on the edge or in clinical devices. Also, by incorporating explainable artificial intelligence (XAI) algorithms, clinical interpretability and confidence may be improved.

Respiratory Reactivation of Herpes Simplex Virus-1 Following Cardiac Surgery in Immunocompetent Individuals.

Vaccinations to Prevent Infections in Adult Individuals With CKD and After Kidney transplantation: A Review.

Frail older people have elevated risk of developing infection. Whether a history of infections is associated with frailty and mortality has not been fully explored. This study investigates the associations between infection history, frailty and their coexistence on all-cause mortality. Study used 1,399 participants (mean age 65.3 years; 51.5% females) of the Baltimore Longitudinal Study of Aging (BLSA) where infection history was established by ICD-9 codes. Frailty was assessed using a 44-item Frailty Index (FI). The association between infection history and frailty was explored using multiple linear regression. Their association with mortality was assessed using Cox proportional hazard models. A history of any infections, and specific infections, including urinary tract, viral infection, pneumonia, and Herpes virus were associated with higher FI (p ≤ 0.01). Plasma Leptin and BMI were positive mediators of the association between infection history and frailty. There was a significant association between infection history and FI progression (from FI < 0.12 to FI ≥ 0.12) in participants 65 years and older (HR 1.43, CI 95% 1.08-1.90, p = 0.02). Participants with both infections' history and high FI (FI ≥0.12) had higher risk of mortality (HR = 1.12; 95%CI: 1.05-2.10; p = 0.02) compared to those without history of infection and low FI. A history of infections may accelerate the rate of frailty progression, defined as the transition from robustness to pre-frailty/frailty, and has an independent effect with frailty on increasing the risk of death. The mediating role of Leptin suggests that obesity may be implicated in the relationship between history of infections and frailty.

Clinical characteristics of respiratory syncytial virus and human metapneumovirus pneumonia in adults: A multicenter retrospective cohort study in Japan.

Invasive fungal disease (IFD) has historically been observed predominantly in neutropenic patients, particularly those with hematological malignancies or undergoing allogeneic hematopoietic stem cell transplantation. However, clinical practice has found that an increasing proportion of patients with IFD have underlying diseases such as chronic obstructive pulmonary disease, severe viral pneumonia, and diabetes mellitus, rather than neutropenia. These patients have significantly different risk factors and degrees of immunosuppression to neutropenic patients, which makes assessment more difficult and leads to changes in how fungal infections are diagnosed. Based on the newly published "Guidelines for the diagnosis and treatment of invasive pulmonary fungal diseases (2025 edition)", this article systematically elaborates on the risk factors and their changes of IFD diagnosis, including classification, grading, and assessment methods. It also emphasizes the differences between neutropenic patients and their possible mechanisms. The aim is to provide a reference for correctly understanding the clinical guidelines for the diagnosis and treatment of IFD.

Yinhuang decoction alleviates RSV-induced viral pneumonia via TLR4/NF-κB/NLRP3 pathway with colloidal phase as dominant active fractions.

Clinical decision support (CDS) systems have increasingly optimized care pathways. Integrating imaging findings into CDS remains a challenge due to unstructured radiology outputs. To evaluate the role of imaging-based decision support within an integrated clinicoradiological pathway for determining emergency department (ED) disposition using modality-based assessment. A computer-assisted reporting/decision support (CAR/DS) tool was developed to standardize chest radiograph and CT interpretations and integrated with a CDS system. 9,036 adult patients with suspected viral pneumonia presenting to the ED (07/2020 -12/2021) were analyzed as a use case. Associations between CAR/DS outputs and ED disposition were assessed using correlation and logistic regression. Agreement between radiograph- and CT-derived outputs was also examined. Most exams were negative [70.9% (6,408/9,036)] and 3.1% typical (276/9,036). Higher CAR/DS likelihood was independently associated with increased odds of admission [all p < 0.001; Groups: radiograph only (OR = 1.86), CT only (OR = 1.83), or both (CT OR = 1.97, radiograph OR = 1.71)]. Radiograph and CT outputs agreement was moderate (Spearman's ρ = 0.43, p < 0.001). Most negative radiographs were followed by a negative CT (60.7%, 389/642), only 23.8% (24/101) of typical radiographs had a subsequent typical CT; most typical radiographs were followed by an indeterminate (37.6%, 38/101) or atypical (28.7%, 29/101) CT. CAR/DS outputs integrated into CDS systems provide actionable information that independently predicts ED disposition. CT added value by excluding suspected pneumonia on radiographs. This exemplifies how imaging data can be standardized and seamlessly incorporated into broader decision pathways, with potential applicability well beyond pandemic-related use cases.

The role of corticosteroids in severe viral pneumonia: lessons from COVID-19 and influenza.

Elucidating the material basis and mechanism of Shegan Mahuang decoction in inhibiting influenza virus pneumonia based on UHPLC-HRMS, network pharmacology and experimental verification.

Middle East respiratory syndrome coronavirus (MERS-CoV) is an emerging coronavirus that can cause zoonotic disease in humans with lethal severe viral pneumonia. Dromedary camels are the source of zoonotic infection. As of November 2025, MERS-CoV has resulted in a total of 2630 reported cases, 37% of these being fatal. The number of reported human cases has been on a decreasing trend since 2016 and reached a nadir during the COVID-19 pandemic. The reason for the reduction of cases is unclear and may be multifactorial. We hypothesized that mutations accumulating in the virus spike protein may have reduced zoonotic potential. Here, we investigate the impact of recently emerged virus spike-protein mutations on virus replication competence using pseudoviruses and replication-competent recombinant viruses. We found that virus spike variants detected in 2019 and some from 2023 show a reduced cell entry, lower viral replication and reduced fitness in human primary alveolar epithelial cells and multiple cell lines. All the MERS-CoV spikes tested showed a cell-entry pathway preference via the cell-surface TMPRSS2 route. Mechanistically, we showed the V530A mutation in the 2019 spike sequence had a reduced human DPP4 binding phenotype. Our data highlighted MERS-CoV spike mutations can modulate viral fitness in human cells and provide new insights to understand recent MERS epidemiology.

Acute Respiratory Distress Syndrome (ARDS) is a critical condition leading to acute hypoxemic respiratory failure, primarily resulting from bacterial sepsis and viral pneumonia. Tuberculosis (TB), generally a chronic lung condition, infrequently manifests as an acute cause of ARDS; but, when it does, it correlates with elevated mortality, especially in TB-endemic areas. Early recognition poses difficulties because of nonspecific clinical and radiological findings, frequently leading to delayed diagnosis and suboptimal outcomes. We describe a 31-year-old previously healthy male who exhibited a 10-day history of high-grade fever, dry cough and progressive dyspnoea, culminating in acute deterioration over a 48-hour period. Upon admission, the patient presented with fever, tachypnoea and significant hypoxemia, exhibiting an oxygen saturation of 76% in room air. Arterial blood gas analysis indicated a PaO₂/FiO₂ ratio of less than 200, meeting the Berlin criteria for acute respiratory distress syndrome. Chest radiography revealed bilateral diffuse alveolar opacities, whereas computed tomography of the thorax displayed extensive ground-glass opacities accompanied by areas of consolidation and centrilobular nodules presenting a tree-in-bud pattern. Despite the administration of non-invasive positive pressure ventilation and empirical broad-spectrum antibiotics, clinical improvement was minimal. The microbiological assessment of induced sputum via cartridge-based nucleic acid amplification testing identified Mycobacterium tuberculosis without evidence of rifampicin resistance. Antitubercular therapy and systemic corticosteroids were commenced, leading to progressive clinical and radiological enhancement. Tuberculosis must be regarded as a significant differential diagnosis in patients exhibiting unexplained ARDS in endemic areas. Timely microbiological confirmation and the swift commencement of antitubercular therapy, coupled with suitable respiratory support, are critical for enhancing patient outcomes.

Pneumonia is a leading cause of antibiotic use for children. This study aimed to assess nationwide antibiotic prescribing patterns for pediatric pneumonia in South Korea between 2016 and 2023, a period encompassing both epidemic and nonepidemic periods of Mycoplasma pneumoniae pneumonia (MPP). Using national claims data from the Health Insurance Review and Assessment Service, we analyzed antibiotic prescriptions for patients under 18 years diagnosed with bacterial pneumonia (International Classification of Diseases, 10th Revision codes; viral pneumonia was excluded). Antibiotic use was measured in days of therapy per 1000 patient-days (inpatients) or per 1000 patients (outpatients). Among 8.7 million inpatient days and 3.5 million outpatient visits, the 2-4 years group accounted for the highest burden (497.1 inpatient days and 161.6 outpatient visits per 1000 population). Macrolides were the most prescribed class in both inpatients [805.1 days of therapy (DOT)/1000 patient-days] and outpatients (4898.0 DOT/1000 patients), except in infants, where third-generation cephalosporins predominated. Use of third-generation cephalosporins increased sharply from 2022 and became the most prescribed class in 2023. Tertiary hospitals had higher use of tetracyclines (26.8 DOT/1000 patient-days) and quinolones but lower use of beta-lactams than other facility types. Antibiotic prescribing peaked in the fourth quarter of each year. Among children in South Korea diagnosed with bacterial pneumonia, macrolides were most frequently prescribed, and prescriptions for third-generation cephalosporins showed an increasing trend. These findings underscore challenges in antimicrobial stewardship and need for strengthened nationwide antimicrobial stewardship policies for pediatric pneumonia tailored to age, setting, and seasonal trends.

Resource-limited settings continue to face challenges in the identification of COVID-19, bacterial pneumonia, viral pneumonia, and normal lung conditions because of the overlap of CT appearance and inter-observer variability. We justify a hybrid architecture of deep learning which combines hand-designed descriptors (Histogram of Oriented Gradients, Local Binary Patterns) and a 20-layer Convolutional Neural Network with dual self-attention. Handcrafted features were then trained with Support Vector Machines, and ensemble averaging was used to integrate the results with the CNN. The confidence level of 0.7 was used to mark suspicious cases to be reviewed manually. On a balanced dataset of 14,000 chest CT scans (3500 per class), the model was trained and cross-validated five-fold on a patient-wise basis. It had 97.43% test accuracy and a macro F1-score of 0.97, which was statistically significant compared to standalone CNN (92.0%), ResNet-50 (90.0%), multiscale CNN (94.5%), and ensemble CNN (96.0%). A further 2–3% enhancement was added by the self-attention module that targets the diagnostically salient lung regions. The predictions that were below the confidence limit amounted to only 5 percent, which indicated reliability and clinical usefulness. The framework provides an interpretable and scalable method of diagnosing multiclass lung disease, especially applicable to be deployed in healthcare settings with limited resources. The further development of the work will involve the multi-center validation, optimization of the model, and greater interpretability to be used in the real world.

Mesenchymal stromal cells engineered with ACE2 for treatment of receptor-mediated viral lung injury.

Prodigiosin targeting alveolar macrophages to mitigate the inflammatory response induced by H9N2 viral infection.

Xuanbai Chengqi decoction targets sphingolipid metabolism: A novel strategy to preserve pulmonary endothelial barrier integrity in severe viral pneumonia.

Herpes Simplex Virus Pneumonia in Immunocompetent Patients Undergoing Cardiac Surgery: Case Series.