Search Results (376)

We propose an advanced electronic nose based on SE-RelationNet for COPD diagnosis with limited breath samples. The model integrates residual blocks, BiGRU layers, and squeeze–excitation attention mechanisms to enhance feature-extraction efficiency. Experimental results demonstrate exceptional performance with minimal samples: in 4-way 1-shot tasks, the model achieves 85.8% mean accuracy (F1-score = 0.852), scaling to 93.3% accuracy (F1-score = 0.931) with four samples per class. Ablation studies confirm that the 5-layer residual structure and single-hidden-layer BiGRU optimize stability (h_F1-score ≤ 0.011). Compared to SiameseNet and ProtoNet, SE-RelationNet shows superior accuracy (>15% improvement in 1-shot tasks). This technology enables COPD detection with as few as one breath sample, facilitating early intervention to mitigate lung cancer risks in COPD patients. Full article

Background: Anti-synthetase Syndrome (ASyS) is an idiopathic inflammatory myopathy characterized by muscle weakness and inflammatory infiltrates in muscles. Sjogren’s disease (SD) is an autoimmune condition primarily affecting exocrine glands. Both these conditions may present lung involvement. We describe a female patient with anti-synthetase/SD overlap syndrome and review the literature to identify published cases describing this overlap, aiming to better define its clinical, radiological, and serological features. Methods: The case description was based on a retrospective collection of clinical, laboratory, and imaging data related to the patient’s diagnostic process and clinical course. Data were anonymized and handled in accordance with the competent territorial Ethics Committee. A literature review was performed using the MEDLINE and Scopus databases by combining the keywords “Anti-Synthetase syndrome”, “Sjögren disease”, “Sjögren syndrome”, “Myositis”, and “Interstitial lung disease” (ILD). Published cases were selected if they met the 2016 EULAR/ACR criteria for SD and at least one of the currently proposed classification criteria for ASyS. Results: The described case concerns a 68-year-old woman with rapidly progressive ILD. The diagnosis of anti-synthetase/SD overlap syndrome was based on clinical, serological (anti-Ro52 and anti-PL7 antibodies), histological, and radiological findings. Despite immunosuppressive and antifibrotic treatment, the clinical course worsened, leading to a poor outcome. In addition, six relevant cases were identified in the literature. Clinical presentations, autoantibody profiles, radiological findings, and outcomes were highly heterogeneous. Among the reported cases, no standardized treatment protocols were adopted, reflecting the lack of consensus in managing this rare condition. Conclusions: In anti-synthetase/SD overlap syndrome, ILD may follow a rapidly progressive course. Early recognition can be challenging, especially in the absence of muscular involvement. This case-based review highlights the need for more standardized approaches to the diagnosis and management of this rare and complex overlap syndrome. Full article

Accurate classification of chest X-ray (CXR) images is crucial for diagnosing lung diseases in medical imaging. Existing deep learning models for CXR image classification face challenges in distinguishing non-lung features. In this work, we propose a new segmentation-assisted fusion-based classification method. The method involves two stages: first, we use a lightweight segmentation model, Partial Convolutional Segmentation Network (PCSNet) designed based on an encoder–decoder architecture, to accurately obtain lung masks from CXR images. Then, a fusion of the masked CXR image with the original image enables classification using the improved lightweight ShuffleNetV2 model. The proposed method is trained and evaluated on segmentation datasets including the Montgomery County Dataset (MC) and Shenzhen Hospital Dataset (SH), and classification datasets such as Chest X-Ray Images for Pneumonia (CXIP) and COVIDx. Compared with seven segmentation models (U-Net, Attention-Net, SegNet, FPNNet, DANet, DMNet, and SETR), five classification models (ResNet34, ResNet50, DenseNet121, Swin-Transforms, and ShuffleNetV2), and state-of-the-art methods, our PCSNet model achieved high segmentation performance on CXR images. Compared to the state-of-the-art Attention-Net model, the accuracy of PCSNet increased by 0.19% (98.94% vs. 98.75%), and the boundary accuracy improved by 0.3% (97.86% vs. 97.56%), while requiring 62% fewer parameters. For pneumonia classification using the CXIP dataset, the proposed strategy outperforms the current best model by 0.14% in accuracy (98.55% vs. 98.41%). For COVID-19 classification with the COVIDx dataset, the model reached an accuracy of 97.50%, the absolute improvement in accuracy compared to CovXNet was 0.1%, and clinical metrics demonstrate more significant gains: specificity increased from 94.7% to 99.5%. These results highlight the model’s effectiveness in medical image analysis, demonstrating clinically meaningful improvements over state-of-the-art approaches. Full article

Respiratory diseases represent a persistent global health challenge, underscoring the need for intelligent, accurate, and personalized diagnostic and therapeutic systems. Existing methods frequently suffer from limitations in diagnostic precision, lack of individualized treatment, and constrained adaptability to complex clinical scenarios. To address these challenges, our study introduces a modular AI-powered framework that integrates an audio-based disease classification model with simulated molecular biomarker profiles to evaluate the feasibility of future multimodal diagnostic extensions, alongside a synthetic-data-driven prescription recommendation engine. The disease classification model analyzes respiratory sound recordings and accurately distinguishes among eight clinical classes: bronchiectasis, pneumonia, upper respiratory tract infection (URTI), lower respiratory tract infection (LRTI), asthma, chronic obstructive pulmonary disease (COPD), bronchiolitis, and healthy respiratory state. The proposed model achieved a classification accuracy of 99.99% on a holdout test set, including 94.2% accuracy on pediatric samples. In parallel, the prescription module provides individualized treatment recommendations comprising drug, dosage, and frequency trained on a carefully constructed synthetic dataset designed to emulate real-world prescribing logic.The model achieved over 99% accuracy in medication prediction tasks, outperforming baseline models such as those discussed in research. Minimal misclassification in the confusion matrix and strong clinician agreement on 200 prescriptions (Cohen’s $\kappa$ = 0.91 [0.87–0.94] for drug selection, 0.78 [0.74–0.81] for dosage, 0.96 [0.93–0.98] for frequency) further affirm the system’s reliability. Adjusted clinician disagreement rates were 2.7% (drug), 6.4% (dosage), and 1.5% (frequency). SHAP analysis identified age and smoking as key predictors, enhancing model explainability. Dosage accuracy was 91.3%, and most disagreements occurred in renal-impaired and pediatric cases. However, our study is presented strictly as a proof-of-concept. The use of synthetic data and the absence of access to real patient records constitute key limitations. A trialed clinical deployment was conducted under a controlled environment with a positive rate of satisfaction from experts and users, but the proposed system must undergo extensive validation with de-identified electronic medical records (EMRs) and regulatory scrutiny before it can be considered for practical application. Nonetheless, the findings offer a promising foundation for the future development of clinically viable AI-assisted respiratory care tools. Full article

Background and Objectives: Cardiac transplantation is currently the elective treatment choice in end-stage heart failure, and cellular rejection is a predictive factor for morbidity and mortality after surgery. We proposed an evaluation of the clinicopathologic factors involved in the mechanism of rejection. Materials and Methods: This study included 146 patients who underwent transplantation at the Institute of Cardiovascular Diseases and Transplantation in Targu Mures between 2010 and 2023, and we evaluated the function and structure of the myocardium after surgery by using endomyocardial biopsy. Results: Overall, 120 men and 26 women underwent transplantation, with an approximately equal proportion under and over 40 years old (48.6% and 51.4%). Evaluating the degree of acute cellular rejection according to the International Society for Heart and Lung Transplantation classification showed that most of the patients presented with acute cellular rejection (ACR) and antibody-mediated rejection (AMR) grade 0, and most cases of ACR and AMR were reported with mild changes (13% or 10.3% patients). Therefore, the most frequent histopathologic diagnoses were similar to lesions unrelated to rejection (45.2% of patients) and ischemia–reperfusion lesions (25.3% patients), respectively. Conclusions: Although 82.2% of the transplanted cases showed no rejection (ISHLT score 0), non-rejection-related lesion-like changes were present in 45.2% of cases, and because more of the non-rejection-related criteria could be detected, it may be necessary to adjust the grading of the rejection criteria. The histopathologic changes that characterize rejection are primarily represented by the mononuclear inflammatory infiltrate; in our study, inflammatory changes were mostly mild (71.9%), with myocyte involvement in all cases. These changes are associated with and contribute to the maintenance of the rejection phenomenon. Full article

Objectives: Accurate differentiation between usual interstitial pneumonia (UIP) and nonspecific interstitial pneumonia (NSIP) is crucial for guiding treatment in interstitial lung diseases (ILDs). This study evaluates the efficacy of clinical, radiomic, and combined models in classifying UIP and NSIP using high-resolution computed tomography (HRCT) scans. Materials and Methods: A retrospective analysis was performed on 105 HRCT scans (UIP = 60, NSIP = 45) from Faisal Hospital and Research Center. Demographic and pulmonary function data formed the clinical model. Radiomic features, extracted using the pyRadiomics package, were refined using recursive feature elimination. A combined model was developed by integrating clinical and radiomic features to assess their complementary diagnostic value. Model performance was assessed via the area under the receiver operating characteristic curve (AUC). SHapley Additive exPlanations (SHAP) analysis, including both global feature importance and individual-level explanations, was used to interpret the model predictions. Results: The clinical model achieved an AUC of 0.62 with a sensitivity of 54% and a specificity of 78%. The radiomic model outperformed it with an AUC of 0.90 with a sensitivity and specificity above 85%. The combined model showed an AUC of 0.86 with a sensitivity of 88% and a specificity of 78%. SHAP analysis identified texture-based features, such as GLCM_Idmn and NGTDM_Contrast, as influential for classification. Conclusions: Radiomic features enhance classification accuracy for UIP and NSIP compared to clinical models. Integrating HCR into clinical workflows may reduce variability and improve diagnostic accuracy in ILD. Future studies should validate findings using larger, multicenter datasets. Full article

Asthma represents a heterogeneous disorder characterized by a dynamic balance between pro-inflammatory and anti-inflammatory forces, with allergic sensitization contributing substantially to airway hyperresponsiveness and remodeling. Central to its pathogenesis are cytokines such as IL-4, IL-5, IL-13, IL-17, and IL-33, which drive recruitment of eosinophils, neutrophils, and other effector cells, thereby precipitating episodic exacerbations in response to viral and environmental triggers. Conventional biomarkers, including blood and sputum eosinophil counts, IgE levels, and fractional exhaled nitric oxide, facilitate phenotypic classification and guide the emerging biologic era. Monoclonal antibodies targeting IgE (omalizumab) and IL-5 (mepolizumab, benralizumab, reslizumab, depemokimab) have demonstrated the ability to reduce exacerbation frequency and improve lung function, with newer agents such as depemokimab offering extended dosing intervals. Itepekimab, an anti-IL-33 antibody, effectively engages its target and mitigates tissue eosinophilia, while CM310-stapokibart, tralokinumab, and lebrikizumab inhibit IL-4/IL-13 signaling with variable efficacy depending on patient biomarkers. Comparative analyses of these biologics, encompassing affinity, dosing regimens, and trial outcomes, underscore the imperative of personalized therapy to optimize disease control in severe asthma. Full article

Ubiquitination is a dynamic and reversible post-translational modification mediated by ubiquitination regulators (UBRs), which plays an essential role in protein stability, cell differentiation and immunity. Dysregulation of UBRs can lead to destabilization of biological processes and may induce serious human diseases, including cancer. Many UBRs, such as E3 ubiquitin ligases and deubiquitinases (DUBs), have been identified as potential drug targets for cancer therapy. However, the potential clinical value of UBRs in lung adenocarcinoma (LUAD) remains to be elucidated. Here, we identified 17 hub UBRs from high-confidence protein–protein interaction networks of UBRs correlated with cancer hallmark-related pathways using four topological algorithms. The expression of hub UBRs is affected by copy number variation and post-transcriptional regulation, and their high expression is often detrimental to patient survival. Based on the expression profiles of hub UBRs, patients can be classified into two ubiquitination subtypes with different characteristics. These subtypes exhibit significant differences across multiple dimensions, including survival, expression level, mutation burden, female predominance, infiltration level, immune profile, and drug response. In addition, we established a scoring system for evaluating the ubiquitination status of individual LUAD patients, called the ubiquitination-related risk (UB_risk) score, and found that patients with low scores are more likely to gain advantages from immunotherapy. The results of this study emphasize the critical role of ubiquitination in the classification, tumor microenvironment and immunotherapy of LUAD. The construction of the UB_risk scoring system lays a research foundation for evaluating the ubiquitination status of individual LUAD patients and formulating precise treatment strategies from the ubiquitination level. Full article

Background: Currently, no information exists on the clinical course of anti-topoisomerase I antibody (ATA)-positive limited cutaneous systemic sclerosis (lcSSc). We aimed to evaluate the incidence of and time to the development of interstitial lung disease (ILD), pulmonary hypertension (PHT), scleroderma renal crisis (SRC), and maximal modified Rodnan skin score (max-mRSS) in patients with lcSSc and dcSSc, with and without ATA. Methods: This cohort study included 522 patients with systemic sclerosis (SSc). The incidence of and time to the development of ILD, PHT, SRC, and max-mRSS were assessed. Results: ATA-positive dcSSc (dcSSc-posATA) was the most common presentation among Thai patients (321 cases; 61.5%). The median time to the development of ILD was shorter than that in lcSSc-posATA, comparable to that in dcSSc-posATA (1.0 vs. 1.8 years, p = 0.21), and shorter than that in ATA-negative dcSSc (dcSSc-negATA) (1.0 vs. 4.8 years, p = 0.001). The time to max-mRSS in lcSSc-posATA was comparable to that in dcSSc-posATA (p = 0.17) but shorter than that in dcSSc-negATA (p < 0.001). Conclusions: Patients with lcSSc-posATA had a similar risk of ILD development and time to reach max-mRSS as those with dcSSc, regardless of the presence of ATA, but had earlier ILD development and max-mRSS compared to those with dcSSc-negATA. Their prognosis appeared to be better than that of dcSSc-posATA. Full article

Background/Objectives: Pulmonary lymphangitic carcinomatosis (PLC) is a rare, aggressive manifestation of metastatic cancer characterized by lymphatic infiltration of the lungs, typically indicating advanced disease and poor prognosis. Methods: This comprehensive narrative review evaluates the role of [18F]fluorodeoxyglucose ([18F]FDG) positron emission tomography (PET) imaging in assessing PLC. Results: Current evidence demonstrates that [18F]FDG PET/CT achieves high diagnostic accuracy, with sensitivity and specificity ranging from 86 to 97% and 84 to 100%, respectively, particularly when employing semiquantitative metrics such as peritumoral standardized uptake value (SUVmax) thresholds (e.g., ≥2.1). PET/CT surpasses high-resolution computed tomography (HRCT) in distinguishing PLC from mimics like pulmonary sarcoidosis by identifying distinct metabolic patterns: bronchovascular hypermetabolism in PLC versus subpleural nodular uptake in sarcoidosis. Prognostically, metabolic tumor burden (e.g., SUVmax × involved lobes) and novel cPLC classifications (localized to the ipsilateral or contralateral lung) independently predict progression-free survival. However, challenges persist, including non-specific tracer uptake in inflammatory conditions and variability in SUV measurements due to technical factors. Emerging digital PET/CT systems, with enhanced spatial resolution, may improve the detection of focal PLC and reduce false negatives. While [18F]FDG PET/CT is invaluable for whole-body staging, therapeutic monitoring and biopsy guidance, the standardization of protocols and multicenter validation of prognostic models are critical for clinical integration. Future research should explore novel tracers (e.g., PSMA for prostate cancer-related PLC) and machine learning approaches to refine diagnostic and prognostic accuracy. Conclusions: This review underscores the role and the transformative potential of [18F]FDG PET/CT in PLC management while advocating for rigorous standardization to maximize its clinical utility. Full article

The accurate and early detection of respiratory diseases is vital for effective diagnosis and treatment. This study presents a new approach for classifying lung sounds using a double denoising method combined with a 1D Convolutional Neural Network (CNN). The preprocessing uses Fast Fourier Transform to clean up sounds and High-Pass Filtering to improve the quality of breathing sounds by eliminating noise and low-frequency interruptions. The Short-Time Fourier Transform (STFT) extracts features that capture localised frequency variations, crucial for distinguishing normal and abnormal respiratory sounds. These features are input into the 1D CNN, which classifies diseases such as bronchiectasis, pneumonia, asthma, COPD, healthy, and URTI. The dual denoising method enhances signal clarity and classification performance. The model achieved 96% validation accuracy, highlighting its reliability in detecting respiratory conditions. The results emphasise the effectiveness of combining signal augmentation with deep learning for automated respiratory sound analysis, with future research focusing on dataset expansion and model refinement for clinical use. Full article

Background/Objectives: Lung cancer is one of the most prevalent malignant diseases in humans. Numerous studies have demonstrated the significance of [¹⁸F]fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) in the staging of this condition. Methods: The pictorial evaluation is based on a recent study comparing preoperative imaging with postoperative histopathological findings following thoracic surgery. It confirmed the value of PET/CT in assessing primary tumor extent and metastatic lymph node involvement; but also revealed discrepancies in primary tumor (T) and lymph nodes (N) classification in 25% and 14% of patients, respectively. Results: The aim of this pictorial review is to highlight and further analyze the causes of inaccurate staging, identify potential diagnostic pitfalls, and provide practical recommendations to help avoid misinterpretation of PET/CT findings. Additionally, the impact of the newly introduced ninth edition of the International Association for the Study of Lung Cancer (IASLC) primary tumor, lymph nodes, and metastasis (TNM) staging system for lung cancer is discussed. Conclusions: In this pictorial review, we presented various sources of error in preoperative staging observed at our institution. Awareness of these potential pitfalls may aid in improving staging accuracy and distinguishing physiological or reactive (benign) processes from pathological findings. Full article

Background/Objectives: HER2-positive breast cancer is an aggressive subtype with an established responsiveness to HER2-targeted therapies like ado-trastuzumab emtansine (T-DM1). However, inter-patient variability in treatment response and toxicity remains a challenge. Hormonal status, particularly menopausal state, may influence breast cancer behavior, therapeutic tolerance, and outcomes, yet data on its effect in patients treated with T-DM1 are scarce. This study aimed to evaluate whether menopausal status independently affects treatment response, side effects, and survival outcomes in HER2-positive breast cancer patients receiving T-DM1, accounting for the confounding role of age. Methods: This retrospective cohort study included 98 female patients with HER2-positive breast cancer treated with T-DM1: 53 premenopausal and 45 postmenopausal. The clinical characteristics, metastatic patterns, treatment history, T-DM1 outcomes, and toxicities were recorded. The statistical analysis included chi-square, t-tests, Mann–Whitney U tests, and Spearman’s correlations. Partial correlation analyses were conducted to isolate the effect of menopausal status by controlling for age. Results: The postmenopausal patients showed higher rates of lung metastasis (42.2% vs. 20.8%) and mortality (60.0% vs. 39.6%) than premenopausal patients. However, no significant differences were found in the T-DM1 response or toxicity profiles. After adjusting for age, menopausal status had no independent association with the treatment outcomes or side effects. Age was the dominant factor influencing performance status, metastatic burden, and mortality risk. Conclusions: Menopausal status affects disease presentation but not T-DM1 efficacy or toxicity when age is accounted for. Treatment decisions should consider age and clinical profile rather than menopausal classification alone when managing HER2-positive breast cancer with T-DM1. Full article

Accurate classification of lung diseases is vital for timely diagnosis and effective treatment of respiratory conditions such as COPD, pneumonia, asthma, and lung cancer. Traditional diagnostic approaches often suffer from limited consistency and elevated false-positive rates, highlighting the demand for more dependable automated systems. To address this challenge, we introduce LSE-Net, an end-to-end deep learning framework that combines precise lung segmentation using an optimized U-Net++ with robust classification powered by an ensemble of DenseNet121 and ResNet50. Leveraging structured hyperparameter tuning and patient-level evaluation, LSE-Net achieves 92.7% accuracy, 96.7% recall, and an F1-score of 94.0%, along with improved segmentation performance (DSC = 0.59 ± 0.01, IoU = 0.523 ± 0.07). These results demonstrate LSE-Net’s ability to reduce diagnostic uncertainty, enhance classification precision, and provide a practical, high-performing solution for real-world clinical deployment in lung disease assessment. Full article

Background/Objectives: Accurate classification of COVID-19 from chest X-rays is critical but remains limited by overlapping features with other lung diseases and the suboptimal performance of current methods. This study addresses the diagnostic gap by introducing a novel hybrid framework for precise segmentation and classification of lung conditions. Methods: The approach combines multi-level thresholding with the advanced metaheuristic optimization algorithms animal migration optimization (AMO), electromagnetism-like optimization (EMO), and the harmony search algorithm (HSA) to enhance image segmentation. A convolutional neural network (CNN) is then employed to classify segmented images into COVID-19, viral pneumonia, or normal categories. Results: The proposed method achieved high diagnostic performance, with 99% accuracy, 99% sensitivity, and 99.5% specificity, confirming its robustness and effectiveness in clinical image classification tasks. Conclusions: This study offers a novel and technically integrated solution for the automated diagnosis of COVID-19 and related lung conditions. The method’s high accuracy and computational efficiency demonstrate its potential for real-world deployment in medical diagnostics. Full article