Detection Of Lung Disease Research Articles

An Improving Lung Disease Detection by Combining Ensemble Deep Learning and Maximum Mean Discrepancy Transfer Learning

An Improving Lung Disease Detection by Combining Ensemble Deep Learning and Maximum Mean Discrepancy Transfer Learning

B-Lines in the Assessment of Interstitial Lung Disease Associated with Scleroderma: The Role of Handheld Devices.

Background: Timely detection and aggressive management of interstitial lung disease (ILD) in systemic sclerosis (SSc) are essential to improving outcomes and reducing risks of irreversible lung injury. Objective: to explore the usefulness of an ultraportable ultrasound device for the management of SSc-related ILD and to compare it with clinical and instrumental data. Methods: A total of 19 consecutive SSc patients underwent a comprehensive pulmonary evaluation: clinical, pulmonary function tests (PFTs) (spirometry, DLCO), lung CT (1.5 mm slice thickness reconstruction; HRCT), and lung ultrasound (LUS). A total score was calculated based on the number of color-coded B-lines recorded for each lung sliding. B-lines were analyzed against dyspnea, cough, Velcro, CT imaging (Warrick's score), and PFTs. Global and subgroup analysis were performed (diffuse versus limited cutaneous SSc, Warrick's < 7 versus >7). Results: Symptomatic lung involvement with varying degrees of dyspnea was reported in about 74% of cases (functional NYHA > 2 in more than half), chronic dry cough in one-third, Velcro rales in 42%. A total of 84.24% were classified as SSc with ILD on CT imaging. Statistically significant mild-to-moderate correlations between B-lines and clinical manifestations were demonstrated, as well as PFTs and Warrick's scores (more B-lines, lower pulmonary function, but higher extent and severity on CT) (p < 0.05); there were differences between SSc patients without and with ILD in terms of the number and distribution of B-lines (p < 0.05), as well as different B-lines patterns and numbers in diffuse versus limited SSc (p < 0.05). Conclusions: Ultraportable handheld LUS is a promising method suitable for the management (screening, early detection, and evaluation) of SSc patients.

Analysis of Lung Disease Prediction using Machine Learning Algorithms

Lung diseases are a notable in global health concern, requiring early diagnosis for better recovery and survival rates. Deep learning strategies, especially CNNs, have shown great promise in self learning lung disease diagnosis from medical images like chest X-rays. Ensemble learning methods using pretrained networks such as VGG16, InceptionV3, and MobileNetV2 have achieved up to 94% accuracy in identifying conditions like COVID-19, pneumonia, and lung opacity. Lightweight CNN models also performed well, with accuracy up to 89.89%. Traditional machine learning algorithms, including Random Forest and Logistic Regression, yielded accuracy rates between 88% and 90%. A hybrid deep learning approach, combining CNN based feature extraction with classifiers like AdaBoost, SVM, and Random Forest, improved classification accuracy by 3.1% and reduced computational complexity by 16.91%. This hybrid method highlights the main feature for integrating deep learning with traditional classifiers to enhance lung disease detection efficiency

A Framework for Detecting Pulmonary Diseases from Lung Sound Signals Using a Hybrid Multi-Task Autoencoder-SVM Model

Research focuses on the efficacy of Multi-Task Autoencoder (MTAE) models in signal classification due to their ability to handle many tasks while improving feature extraction. However, researchers have not thoroughly investigated the study of lung sounds (LSs) for pulmonary disease detection. This paper introduces a new framework that utilizes an MTAE model to detect lung diseases based on LS signals. The model integrates an autoencoder and a supervised classifier, simultaneously optimizing both classification accuracy and signal reconstruction. Furthermore, we propose a hybrid approach that combines an MTAE and a Support Vector Machine (MTAE-SVM) to enhance performance. We evaluated our model using LS signals from a publicly available database from King Abdullah University Hospital. The model attained an accuracy of 89.47% for four classes (normal, pneumonia, asthma, and chronic obstructive pulmonary disease) and 90.22% for three classes (normal, pneumonia, and asthma cases). Using the MTAE-SVM, the accuracy was further improved to 91.49% for four classes and 93.08% for three classes, respectively. The results indicate that the MTAE and MTAE-SVM have a considerable potential for detecting pulmonary diseases from lung sound signals. This could aid in the creation of more user-friendly and effective diagnostic tools.

A dataset of Solicited Cough Sound for Tuberculosis Triage Testing

Cough is a common and commonly ignored symptom of lung disease. Cough is often perceived as difficult to quantify, frequently self-limiting, and non-specific. However, cough has a central role in the clinical detection of many lung diseases including tuberculosis (TB), which remains the leading infectious disease killer worldwide. TB screening currently relies on self-reported cough which fails to meet the World Health Organization (WHO) accuracy targets for a TB triage test. Artificial intelligence (AI) models based on cough sound have been developed for several respiratory conditions, with limited work being done in TB. To support the development of an accurate, point-of-care cough-based triage tool for TB, we have compiled a large multi-country database of cough sounds from individuals being evaluated for TB. The dataset includes more than 700,000 cough sounds from 2,143 individuals with detailed demographic, clinical and microbiologic diagnostic information. We aim to empower researchers in the development of cough sound analysis models to improve TB diagnosis, where innovative approaches are critically needed to end this long-standing pandemic.

Clinical utility of lung ultrasound for the detection of interstitial lung disease in patients with rheumatoid arthritis.

To establish the diagnostic value of lung ultrasound (LUS) in patients with rheumatoid arthritis (RA) for the detection of interstitial lung disease (ILD). A cross-sectional study was performed. Consecutive patients with RA (American College of Rheumatology/European League Against Rheumatism 2010 criteria) who had a chest high-resolution computed tomography (HRCT) performed within 12 months before inclusion, regardless of symptomatology, were included. Demographic, clinical, laboratory, and pharmacological data were recorded. Each patient underwent a LUS with assessment of B-lines (BL) and pleural irregularities (PI). HRCT was considered the gold standard for the confirmatory diagnosis of ILD. Receiver operating characteristic (ROC) curves were calculated to test the ability of LUS findings (BL and PI) in discriminating patients with ILD. A total of 104 RA patients were included, of which 21.8% had ILD. Patients with ILD had more BL (median 26 versus 1, p<0.001) and PI (median 16 versus 5, p<0.001) than patients without ILD. The diagnostic accuracy in ROC curves was: area under the curve (AUC) 0.88 and 95% confidence interval (CI) 0.78-0.93 for BL and AUC 0.82 and 95% CI 0.74-0.89 for PI. The best cut-off points for (ILD detection) discriminating the presence of significant interstitial lung abnormalities were 8 BL and 7 PI. The presence of 8 BL and/or 7 PI in the LUS showed an adequate cut-off value for discriminating the presence of significant interstitial lung abnormalities, evocative of ILD.

Detection and management of autoimmune disease-associated interstitial lung diseases.

Interstitial lung disease (ILD) causes significant morbidity and mortality in patients with systemic autoimmune rheumatic diseases. Patients at high risk of ILD should be screened using high-resolution CT (HRCT), but there is no consensus as to which risk factors-or combination of risk factors-should prompt referral for HRCT. The course of autoimmune disease-associated ILD is highly variable, and it may not mirror the activity of the underlying autoimmune disease. Patients require close monitoring with periodic pulmonary function testing and symptom assessment and with repeat HRCT considered based on clinical assessment. The relevance of clinical and radiologic signs of progression-and their implications for management-ideally should be discussed by a multidisciplinary team. Management of autoimmune disease-associated ILD may involve immunosuppressant and/or antifibrotic therapy in addition to supportive care. It is important that treatment decisions be individualized to the needs and wishes of the patient. Regular follow-up is important to monitor disease progression and manage the adverse effects related to treatment.

Experiment on Lung Disease Classification using YOLOv8

Lung diseases are a leading cause of illness and mortality worldwide. Accurate and timely diagnosis is crucial for improving patient outcomes, but manual interpretation of chest X-rays by radiologists may consume a significant amount of time and is susceptible to errors. This study leverages the state-of-the-art YOLOv8 deep learning model to develop an automated system for classifying lung diseases from chest X-ray images. The proposed approach utilizes a large, diverse dataset of 21,165 chest X-ray images categorized into four classes: COVID-19, viral pneumonia, lung opacity, and normal. The YOLOv8-cls model is fine-tuned using transfer learning and advanced data augmentation techniques. The system achieves a high accuracy of approximately 95% across all disease classes while maintaining real-time performance. Confusion matrix analysis demonstrates strong performance in correctly identifying each condition, with COVID-19, normal, viral pneumonia and lung opacity cases correctly identified 97%, 97%, 99% and 93% of the time respectively. The outcomes validate the adaptability and reliability of deep learning, specifically YOLOv8, for automated lung disease detection, offering the potential to improve clinical workflows and patient care by providing efficient screening tools for practitioners. The study addresses limitations of previous works by utilizing a large consolidated dataset, reporting comprehensive computational efficiency metrics, and extending classification complexity without substantially impacting accuracy or speed. Future research should focus on further optimizations and extending the curated image repositories to include under-represented patient groups to enhance the model's inclusiveness and robustness.

Volatile Organic Compounds (VOC) and Serum Leukotriene B4 between COPD Patients and COPD with Lung Cancer Patients

Introduction: Chronic obstructive pulmonary disease (COPD) is estimated to become the third leading cause of death worldwide in 2030. COPD can affect the lungs and cause chronic systemic inflammation. Leukotriene B4 (LTB4) is involved in COPD and lung cancer pathogenesis. There has been the development of non-invasive methods for detecting lung disease in the last few decades, such as the examination of volatile organic compounds (VOC). This study aimed to analyze the serum LTB4 and the difference of VOCs in exhaled breath of stable COPD patients and COPD with lung cancer patients. Methods: This case-control study recruited 20 stable COPD patients and 20 patients with COPD and lung cancer. An exhaled breath sample was collected in Tedlar bags and analyzed using an arrayed sensor breath analyzer to check the concentration of VOCs. Meanwhile, a venous blood sample was collected to examine the level of LTB4 using an ELISA kit. Independent t-test and Mann-Whitney test were used to analyze the data. Results: The carbon dioxide (CO2), nitrogen dioxide (NO2), carbon monoxide (CO), benzene (C6H6), and propane (C3H8) levels were significantly different (p &lt;0.05) in COPD-only patients compared to COPD with lung cancer patients. Serum LTB4 increased in both groups. Conclusion: CO2, CO, and C3H8 levels increased, but the NO2 level decreased in COPD patients with lung cancer compared to COPD-only patients. Serum LTB4 increased in COPD with lung cancer patients.

Data-driven classification and explainable-AI in the field of lung imaging.

Detecting lung diseases in medical images can be quite challenging for radiologists. In some cases, even experienced experts may struggle with accurately diagnosing chest diseases, leading to potential inaccuracies due to complex or unseen biomarkers. This review paper delves into various datasets and machine learning techniques employed in recent research for lung disease classification, focusing on pneumonia analysis using chest X-ray images. We explore conventional machine learning methods, pretrained deep learning models, customized convolutional neural networks (CNNs), and ensemble methods. A comprehensive comparison of different classification approaches is presented, encompassing data acquisition, preprocessing, feature extraction, and classification using machine vision, machine and deep learning, and explainable-AI (XAI). Our analysis highlights the superior performance of transfer learning-based methods using CNNs and ensemble models/features for lung disease classification. In addition, our comprehensive review offers insights for researchers in other medical domains too who utilize radiological images. By providing a thorough overview of various techniques, our work enables the establishment of effective strategies and identification of suitable methods for a wide range of challenges. Currently, beyond traditional evaluation metrics, researchers emphasize the importance of XAI techniques in machine and deep learning models and their applications in classification tasks. This incorporation helps in gaining a deeper understanding of their decision-making processes, leading to improved trust, transparency, and overall clinical decision-making. Our comprehensive review serves as a valuable resource for researchers and practitioners seeking not only to advance the field of lung disease detection using machine learning and XAI but also from other diverse domains.

Deep Learning Approach for Lung Disease Detection by Image Processing

Deep Learning Approach for Lung Disease Detection by Image Processing

The role of screening of patients proved to have gastroesophageal reflux disease by upper gastrointestinal endoscope for early detection of interstitial lung diseases

BackgroundGastroesophageal reflux disease (GERD) is a frequent digestive illness in North America, causing symptoms and complications. It affects 18.1–27.28% of people and is associated with decreased quality of life. Treatment improves quality of life, but treatment costs are higher. The study aims to assess the role of screening of patients proven to have GERD by upper gastrointestinal endoscope for early detection of interstitial lung diseases.Subjects and methodsThis research was cross-sectional research. The study will be conducted at the Chest Diseases Department of Ain Shams University hospitals. This prospective cross-sectional research will be performed on 90 randomly selected cases diagnosed with GERD by upper gastrointestinal endoscopy. For 1 year.ResultThe findings of chest HRCT for enrolled patients are 15 patients (16.7%) showed positive CT findings suggestive of interstitial lung disease, whereas 75 patients (83.3%) had normal CT scans. The most common CT finding was atelectasis reported in nine patients (10%). a statistically significant correlation between age and fibrosis and reticular infiltration, female gender, and reticular infiltration. no statistically significant association between CT findings and each of the chest symptoms and GERD symptoms.ConclusionCT chest screen of patients diagnosed with GERD by upper gastrointestinal endoscopy plays a favorable role in the early detection of ILD to help in early management and treatment them. In patients diagnosed with GERD by upper gastrointestinal endoscopy, we revealed that: reticular infiltration is more prevalent in female gender. Reticular infiltration and fibrosis increase with older age. Reticular infiltration and atelectasis increase in patients who have a hiatus hernia. Cough, dyspnea, and wheezes increase in older age.

Proteomic Risk Score of Increased Respiratory Susceptibility: A Multi-Cohort Study.

Accelerated decline in lung function is associated with incident COPD, hospitalizations and death. However, identifying this trajectory with longitudinal spirometry measurements is challenging in clinical practice. To determine whether a proteomic risk score trained on accelerated decline in lung function can assess risk of future respiratory disease and mortality. In CARDIA, a population-based cohort starting in young adulthood, longitudinal measurements of FEV1 percent predicted (up to six timepoints over 30 years) were used to identify accelerated and normal decline trajectories. Protein aptamers associated with an accelerated decline trajectory were identified with multivariable logistic regression followed by LASSO regression. The proteomic respiratory susceptibility score was derived based on these circulating proteins and applied to the UK Biobank and COPDGene studies to examine associations with future respiratory morbidity and mortality. Higher susceptibility score was independently associated with all-cause mortality (UKBB: HR 1.56, 95%CI 1.50-1.61; COPDGene: HR 1.75, 95%CI 1.63-1.88), respiratory mortality (UKBB: HR 2.39, 95% CI 2.16-2.64; COPDGene: HR 1.83, 95%CI 1.33-2.51), incident COPD (UKBB: HR 1.84, 95%CI 1.71-1.98), incident respiratory exacerbation (COPDGene: OR 1.11, 95%CI 1.03-1.20), and incident exacerbation requiring hospitalization (COPDGene: OR 1.18, 95%CI 1.08-1.28). A proteomic signature of increased respiratory susceptibility identifies people at risk of respiratory death, incident COPD, and respiratory exacerbations. This susceptibility score is comprised of proteins with well-known and novel associations with lung health and holds promise for the early detection of lung disease without requiring years of spirometry measurements.

Subjective symptoms are triggers for the detection of immune checkpoint inhibitor-induced interstitial lung disease and associate with disease severity: a single-center retrospective study

BackgroundInterstitial lung disease (ILD) is one of the most common fatal immune-related adverse events (irAEs). ILD development adversely affects the continuation of anticancer drug therapy, including immune checkpoint inhibitor (ICI) therapy and prognosis. There are no established useful clinical indicators for the early detection of ILD. Furthermore, the factors that lead the attending physician to suspect ICI-induced ILD (ICI-ILD) remain unclear. This study aimed to investigate the ICI-ILD detection based on subjective symptoms and their relationship with disease severity in patients receiving anti-PD-1/PD-L1 antibody.MethodsThis was a retrospective observational study. We enrolled the patients who received anti-PD-1/PD-L1 antibody at Kyoto University Hospital between September 2014 and April 2021. Patients who developed ICI-ILD were stratified into two distinct groups based on factors that triggered the suspicion of ILD development. The “Subjective symptoms” group was defined as patients in whom ILD was detected based on subjective symptoms. Conversely, the “Routine examinations” group was defined as patients in whom ILD was suspected based on scheduled routine examinations. The severity of ILD in each group was assessed and its association with changes in the respiratory symptoms was examined.ResultsOf 926 patients who received anti-PD-1/PD-L1 antibody, 51 patients (5.5%) developed ICI-ILD. The incidence of ICI-ILD in patients with lung cancer was significantly higher than that in patients with other cancers (P < 0.001). Among the patients with ICI-ILD, 27 patients (52.9%) were classified into the “Subjective symptoms” group. The “Subjective symptoms” group exhibited a significantly higher proportion of Grade 3–5 ICI-ILD cases than the “Routine examinations” group (76.2% vs. 23.8%, P = 0.010). At the last visit, before the suspected onset of ILD, 21 of the 27 patients (77.8%) had no symptoms or no change in the respiratory symptoms.ConclusionSubjective symptoms triggered the suspicion of Grade 3–5 ICI-ILD. Enhanced monitoring and patient education could be essential for the early detection of ICI-ILD because ILD may develop rapidly. Our findings might help to manage ICI-ILD in clinical practice.

Lung Ultrasound and Ultrasound Score: A Useful Tool in Neonatal Intensive Care Units for the Diagnosis and Therapeutic Management of Newborns With Respiratory Pathology.

Pulmonary ultrasound has become a fundamental tool for the early detection and management of major neonatal lung diseases in neonatal intensive care units (NICUs). The advantages of this imaging investigation include its rapid execution and information acquisition, non-invasive nature, early diagnosis establishment, dynamic monitoring, and usefulness in therapeutic management. Regarding therapeutic management, the lung ultrasound (LUS) score is used as a basic tool for indicating surfactant administration. Performing and interpreting pulmonary ultrasounds requires an experienced clinician capable of recognizing anatomical structures, understanding the limitations of the technique, and correlating the obtained data with the patient's clinical picture. A series of diagnostic characteristics of pulmonary ultrasonography for neonatal lung pathologies have been described, making pulmonary ultrasound a useful tool in establishing differential diagnoses. This study evaluates the effectiveness of ultrasonography in determining the severity of lung pathologies in newborns and its impact on therapeutic decision-making, including surfactant administration and continuous positive airway pressure (CPAP) support. Newborns admitted to the NICUwith various respiratory conditions underwent LUS scoring. The study analyzed the relationship between LUS scoring and the severity of conditions such as pneumonia, respiratory distress syndrome, meconium aspiration syndrome, transient tachypnea of the newborn, and pneumothorax. The correlation between LUS scoring, surfactant administration, and CPAP requirements was also examined.

Exploring explainable AI features in the vocal biomarkers of lung disease

This review delves into the burgeoning field of explainable artificial intelligence (XAI) in the detection and analysis of lung diseases through vocal biomarkers. Lung diseases, often elusive in their early stages, pose a significant public health challenge. Recent advancements in AI have ushered in innovative methods for early detection, yet the black-box nature of many AI models limits their clinical applicability. XAI emerges as a pivotal tool, enhancing transparency and interpretability in AI-driven diagnostics. This review synthesizes current research on the application of XAI in analyzing vocal biomarkers for lung diseases, highlighting how these techniques elucidate the connections between specific vocal features and lung pathology. We critically examine the methodologies employed, the types of lung diseases studied, and the performance of various XAI models. The potential for XAI to aid in early detection, monitor disease progression, and personalize treatment strategies in pulmonary medicine is emphasized. Furthermore, this review identifies current challenges, including data heterogeneity and model generalizability, and proposes future directions for research. By offering a comprehensive analysis of explainable AI features in the context of lung disease detection, this review aims to bridge the gap between advanced computational approaches and clinical practice, paving the way for more transparent, reliable, and effective diagnostic tools.

Optimization of vision transformer-based detection of lung diseases from chest X-ray images

BackgroundRecent advances in Vision Transformer (ViT)-based deep learning have significantly improved the accuracy of lung disease prediction from chest X-ray images. However, limited research exists on comparing the effectiveness of different optimizers for lung disease prediction within ViT models. This study aims to systematically evaluate and compare the performance of various optimization methods for ViT-based models in predicting lung diseases from chest X-ray images.MethodsThis study utilized a chest X-ray image dataset comprising 19,003 images containing both normal cases and six lung diseases: COVID-19, Viral Pneumonia, Bacterial Pneumonia, Middle East Respiratory Syndrome (MERS), Severe Acute Respiratory Syndrome (SARS), and Tuberculosis. Each ViT model (ViT, FastViT, and CrossViT) was individually trained with each optimization method (Adam, AdamW, NAdam, RAdam, SGDW, and Momentum) to assess their performance in lung disease prediction.ResultsWhen tested with ViT on the dataset with balanced-sample sized classes, RAdam demonstrated superior accuracy compared to other optimizers, achieving 95.87%. In the dataset with imbalanced sample size, FastViT with NAdam achieved the best performance with an accuracy of 97.63%.ConclusionsWe provide comprehensive optimization strategies for developing ViT-based model architectures, which can enhance the performance of these models for lung disease prediction from chest X-ray images.

Comparative sensitivity of early cystic fibrosis lung disease detection tools in school aged children

BackgroundEffective detection of early lung disease in cystic fibrosis (CF) is critical to understanding early pathogenesis and evaluating early intervention strategies. We aimed to compare ability of several proposed sensitive functional tools to detect early CF lung disease as defined by CT structural disease in school aged children. Methods50 CF subjects (mean±SD 11.2 ± 3.5y, range 5–18y) with early lung disease (FEV1≥70 % predicted: 95.7 ± 11.8 %) performed spirometry, Multiple breath washout (MBW, including trapped gas assessment), oscillometry, cardiopulmonary exercise testing (CPET) and simultaneous spirometer-directed low-dose CT imaging. CT data were analysed using well-evaluated fully quantitative software for bronchiectasis and air trapping (AT). ResultsCT bronchiectasis and AT occurred in 24 % and 58 % of patients, respectively. Of the functional tools, MBW detected the highest rates of abnormality: Scond 82 %, MBWTG RV 78 %, LCI 74 %, MBWTG IC 68 % and Sacin 51 %. CPET VO2peak detected slightly higher rates of abnormality (9 %) than spirometry-based FEV1 (2 %). For oscillometry AX (14 %) performed better than Rrs (2 %) whereas Xrs and R5-19 failed to detect any abnormality. LCI and Scond correlated with bronchiectasis (r = 0.55–0.64, p < 0.001) and AT (r = 0.73–0.74, p < 0.001). MBW-assessed trapped gas was detectable in 92 % of subjects and concordant with CT-assessed AT in 74 %. ConclusionsSignificant structural and functional deficits occur in early CF lung disease, as detected by CT and MBW. For MBW, additional utility, beyond that offered by LCI, was suggested for Scond and MBW-assessed gas trapping. Our study reinforces the complementary nature of these tools and the limited utility of conventional oscillometry and CPET in this setting.

Visualization and Segmentation of Lung Tissue in DICOM-format CT Images

Abstract In this project, two tools were developed as a preliminary step towards the creation of future Computer-Aided Diagnosis systems for the automatic detection of lung diseases. The first methodology aimed to extract information from DICOM-format files using Python and the Jupyter Notebook programming environment. The implemented process enables the generation of images in various anatomical planes, using the pixel matrix in Hounsfield units. Additionally, a custom tool was developed for anonymizing and visualizing file information. The techniques employed are independent of the grayscale image generated and are not reliant on the type of window used. The second tool introduces a lung segmentation technique for chest CT images. The segmentation uses the pixel matrix in Hounsfield units to initially identify lung tissue based on its linear X-ray attenuation. Subsequently, the Watershed algorithm based on markers is employed as the primary foundation of the technique. Various morphological operations and transformations on axial slices are applied, resulting in successful lung segmentation.

Advanced Lung Disease Detection and Classification Using Ge-U-Net-ODLwith Gabor Filters and Entropy-Based Feature Selection

Lung X-ray images play a crucial role in the detection and diagnosis of various lung diseases,including pneumonia, tuberculosis, and lung cancer. These images provide a non-invasive method for visualizing lung structures, allowing radiologists and machine learning models to identify abnormalities such as nodules, masses, or fluid accumulation. With the advancement of deep learning techniques, lung X-ray images are now used in automated systems that can detect and classify diseases with high accuracy. By applying sophisticated algorithms like GE-U-Net-ODL, Gabor filters, and entropy-based feature extraction, these images are analyzed pixel by pixel to enhance feature representation and improve diagnostic precision. This paper presents a novel approach for lung disease detection and classification using the GE-U-Net-ODL model, which integrates advanced preprocessing techniques and deep learning architectures. The study leverages the NIH Chest X-ray dataset and employs a variety of feature extraction and selection methods, including Gabor filters, entropy-based techniques, and multi-scale inputs. A detailed comparative analysis of different model configurations demonstrates that the GE-U-Net-ODL model with Transfer Learning achieves the highest classification accuracy of 95.0%, alongside superior precision (93.5%), recall (94.0%), and F1-score (93.7%). Other configurations, such as those utilizing data augmentation and hybrid filters, also showed notable performance improvements. The research underscores the model's effectiveness in enhancing diagnostic accuracy for lung diseases while balancing training and inference times.