Pneumonia Detection Using Chest Radiographs With Novel EfficientNetV2L Model

Artículo Materias > Ingeniería Universidad Europea del Atlántico > Investigación > Producción Científica
Fundación Universitaria Internacional de Colombia > Investigación > Artículos y libros
Universidad Internacional Iberoamericana México > Investigación > Producción Científica
Universidad Internacional Iberoamericana Puerto Rico > Investigación > Producción Científica
Universidad Internacional do Cuanza > Investigación > Producción Científica
Abierto Inglés Pneumonia is a potentially life-threatening infectious disease that is typically diagnosed through physical examinations and diagnostic imaging techniques such as chest X-rays, ultrasounds or lung biopsies. Accurate diagnosis is crucial as wrong diagnosis, inadequate treatment or lack of treatment can cause serious consequences for patients and may become fatal. The advancements in deep learning have significantly contributed to aiding medical experts in diagnosing pneumonia by assisting in their decision-making process. By leveraging deep learning models, healthcare professionals can enhance diagnostic accuracy and make informed treatment decisions for patients suspected of having pneumonia. In this study, six deep learning models including CNN, InceptionResNetV2, Xception, VGG16, ResNet50 and EfficientNetV2L are implemented and evaluated. The study also incorporates the Adam optimizer, which effectively adjusts the epoch for all the models. The models are trained on a dataset of 5856 chest X-ray images and show 87.78%, 88.94%, 90.7%, 91.66%, 87.98% and 94.02% accuracy for CNN, InceptionResNetV2, Xception, VGG16, ResNet50 and EfficientNetV2L, respectively. Notably, EfficientNetV2L demonstrates the highest accuracy and proves its robustness for pneumonia detection. These findings highlight the potential of deep learning models in accurately detecting and predicting pneumonia based on chest X-ray images, providing valuable support in clinical decision-making and improving patient treatment. metadata Ali, Mudasir; Shahroz, Mobeen; Akram, Urooj; Mushtaq, Muhammad Faheem; Carvajal-Altamiranda, Stefanía; Aparicio Obregón, Silvia; Díez, Isabel De La Torre y Ashraf, Imran mail SIN ESPECIFICAR, SIN ESPECIFICAR, SIN ESPECIFICAR, SIN ESPECIFICAR, stefania.carvajal@uneatlantico.es, silvia.aparicio@uneatlantico.es, SIN ESPECIFICAR, SIN ESPECIFICAR (2024) Pneumonia Detection Using Chest Radiographs With Novel EfficientNetV2L Model. IEEE Access, 12. pp. 34691-34707. ISSN 2169-3536

[img] Texto
Pneumonia_Detection_Using_Chest_Radiographs_With_Novel_EfficientNetV2L_Model.pdf
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Descargar (2MB)

Resumen

Pneumonia is a potentially life-threatening infectious disease that is typically diagnosed through physical examinations and diagnostic imaging techniques such as chest X-rays, ultrasounds or lung biopsies. Accurate diagnosis is crucial as wrong diagnosis, inadequate treatment or lack of treatment can cause serious consequences for patients and may become fatal. The advancements in deep learning have significantly contributed to aiding medical experts in diagnosing pneumonia by assisting in their decision-making process. By leveraging deep learning models, healthcare professionals can enhance diagnostic accuracy and make informed treatment decisions for patients suspected of having pneumonia. In this study, six deep learning models including CNN, InceptionResNetV2, Xception, VGG16, ResNet50 and EfficientNetV2L are implemented and evaluated. The study also incorporates the Adam optimizer, which effectively adjusts the epoch for all the models. The models are trained on a dataset of 5856 chest X-ray images and show 87.78%, 88.94%, 90.7%, 91.66%, 87.98% and 94.02% accuracy for CNN, InceptionResNetV2, Xception, VGG16, ResNet50 and EfficientNetV2L, respectively. Notably, EfficientNetV2L demonstrates the highest accuracy and proves its robustness for pneumonia detection. These findings highlight the potential of deep learning models in accurately detecting and predicting pneumonia based on chest X-ray images, providing valuable support in clinical decision-making and improving patient treatment.

Tipo de Documento: Artículo
Palabras Clave: Pneumonia detection, transfer learning, efficientnetv2l, data augmentation, chest X-rays
Clasificación temática: Materias > Ingeniería
Divisiones: Universidad Europea del Atlántico > Investigación > Producción Científica
Fundación Universitaria Internacional de Colombia > Investigación > Artículos y libros
Universidad Internacional Iberoamericana México > Investigación > Producción Científica
Universidad Internacional Iberoamericana Puerto Rico > Investigación > Producción Científica
Universidad Internacional do Cuanza > Investigación > Producción Científica
Depositado: 12 Abr 2024 23:30
Ultima Modificación: 12 Abr 2024 23:30
URI: https://repositorio.unincol.edu.co/id/eprint/11666

Acciones (logins necesarios)

Ver Objeto Ver Objeto

<a class="ep_document_link" href="/17140/1/s41598-025-90616-w.pdf"><img class="ep_doc_icon" alt="[img]" src="/style/images/fileicons/text.png" border="0"/></a>

en

open

Efficient CNN architecture with image sensing and algorithmic channeling for dataset harmonization

The process of image formulation uses semantic analysis to extract influential vectors from image components. The proposed approach integrates DenseNet with ResNet-50, VGG-19, and GoogLeNet using an innovative bonding process that establishes algorithmic channeling between these models. The goal targets compact efficient image feature vectors that process data in parallel regardless of input color or grayscale consistency and work across different datasets and semantic categories. Image patching techniques with corner straddling and isolated responses help detect peaks and junctions while addressing anisotropic noise through curvature-based computations and auto-correlation calculations. An integrated channeled algorithm processes the refined features by uniting local-global features with primitive-parameterized features and regioned feature vectors. Using K-nearest neighbor indexing methods analyze and retrieve images from the harmonized signature collection effectively. Extensive experimentation is performed on the state-of-the-art datasets including Caltech-101, Cifar-10, Caltech-256, Cifar-100, Corel-10000, 17-Flowers, COIL-100, FTVL Tropical Fruits, Corel-1000, and Zubud. This contribution finally endorses its standing at the peak of deep and complex image sensing analysis. A state-of-the-art deep image sensing analysis method delivers optimal channeling accuracy together with robust dataset harmonization performance.

Producción Científica

Khadija Kanwal mail , Khawaja Tehseen Ahmad mail , Aiza Shabir mail , Li Jing mail , Helena Garay mail helena.garay@uneatlantico.es, Luis Eduardo Prado González mail uis.prado@uneatlantico.es, Hanen Karamti mail , Imran Ashraf mail ,

Kanwal

<a href="/17392/1/journal.pone.0317863.pdf" class="ep_document_link"><img class="ep_doc_icon" alt="[img]" src="/style/images/fileicons/text.png" border="0"/></a>

en

open

Deep image features sensing with multilevel fusion for complex convolution neural networks & cross domain benchmarks

Efficient image retrieval from a variety of datasets is crucial in today's digital world. Visual properties are represented using primitive image signatures in Content Based Image Retrieval (CBIR). Feature vectors are employed to classify images into predefined categories. This research presents a unique feature identification technique based on suppression to locate interest points by computing productive sum of pixel derivatives by computing the differentials for corner scores. Scale space interpolation is applied to define interest points by combining color features from spatially ordered L2 normalized coefficients with shape and object information. Object based feature vectors are formed using high variance coefficients to reduce the complexity and are converted into bag-of-visual-words (BoVW) for effective retrieval and ranking. The presented method encompass feature vectors for information synthesis and improves the discriminating strength of the retrieval system by extracting deep image features including primitive, spatial, and overlayed using multilayer fusion of Convolutional Neural Networks(CNNs). Extensive experimentation is performed on standard image datasets benchmarks, including ALOT, Cifar-10, Corel-10k, Tropical Fruits, and Zubud. These datasets cover wide range of categories including shape, color, texture, spatial, and complicated objects. Experimental results demonstrate considerable improvements in precision and recall rates, average retrieval precision and recall, and mean average precision and recall rates across various image semantic groups within versatile datasets. The integration of traditional feature extraction methods fusion with multilevel CNN advances image sensing and retrieval systems, promising more accurate and efficient image retrieval solutions.

Producción Científica

Jyotismita Chaki mail , Aiza Shabir mail , Khawaja Tehseen Ahmed mail , Arif Mahmood mail , Helena Garay mail helena.garay@uneatlantico.es, Luis Eduardo Prado González mail uis.prado@uneatlantico.es, Imran Ashraf mail ,

Chaki

<a class="ep_document_link" href="/17450/1/ejaz-et-al-2025-fundus-image-classification-using-feature-concatenation-for-early-diagnosis-of-retinal-disease.pdf"><img class="ep_doc_icon" alt="[img]" src="/style/images/fileicons/text.png" border="0"/></a>

en

open

Fundus image classification using feature concatenation for early diagnosis of retinal disease

Background Deep learning models assist ophthalmologists in early detection of diseases from retinal images and timely treatment. Aim Owing to robust and accurate results from deep learning models, we aim to use convolutional neural network (CNN) to provide a non-invasive method for early detection of eye diseases. Methodology We used a hybridized CNN with deep learning (DL) based on two separate CNN blocks, to identify multiple Optic Disc Cupping, Diabetic Retinopathy, Media Haze, and Healthy images. We used the RFMiD dataset, which contains various categories of fundus images representing different eye diseases. Data augmenting, resizing, coping, and one-hot encoding are used among other preprocessing techniques to improve the performance of the proposed model. Color fundus images have been analyzed by CNNs to extract relevant features. Two CCN models that extract deep features are trained in parallel. To obtain more noticeable features, the gathered features are further fused utilizing the Canonical Correlation Analysis fusion approach. To assess the effectiveness, we employed eight classification algorithms: Gradient boosting, support vector machines, voting ensemble, medium KNN, Naive Bayes, COARSE- KNN, random forest, and fine KNN. Results With the greatest accuracy of 93.39%, the ensemble learning performed better than the other algorithms. Conclusion The accuracy rates suggest that the deep learning model has learned to distinguish between different eye disease categories and healthy images effectively. It contributes to the field of eye disease detection through the analysis of color fundus images by providing a reliable and efficient diagnostic system.

Producción Científica

Sara Ejaz mail , Hafiz U Zia mail , Fiaz Majeed mail , Umair Shafique mail , Stefanía Carvajal-Altamiranda mail stefania.carvajal@uneatlantico.es, Vivian Lipari mail vivian.lipari@uneatlantico.es, Imran Ashraf mail ,

Ejaz

<a class="ep_document_link" href="/15983/1/Food%20Science%20%20%20Nutrition%20-%202025%20-%20Tanveer%20-%20Novel%20Transfer%20Learning%20Approach%20for%20Detecting%20Infected%20and%20Healthy%20Maize%20Crop.pdf"><img class="ep_doc_icon" alt="[img]" src="/style/images/fileicons/text.png" border="0"/></a>

en

open

Novel Transfer Learning Approach for Detecting Infected and Healthy Maize Crop Using Leaf Images

Maize is a staple crop worldwide, essential for food security, livestock feed, and industrial uses. Its health directly impacts agricultural productivity and economic stability. Effective detection of maize crop health is crucial for preventing disease spread and ensuring high yields. This study presents VG-GNBNet, an innovative transfer learning model that accurately detects healthy and infected maize crops through a two-step feature extraction process. The proposed model begins by leveraging the visual geometry group (VGG-16) network to extract initial pixel-based spatial features from the crop images. These features are then further refined using the Gaussian Naive Bayes (GNB) model and feature decomposition-based matrix factorization mechanism, which generates more informative features for classification purposes. This study incorporates machine learning models to ensure a comprehensive evaluation. By comparing VG-GNBNet's performance against these models, we validate its robustness and accuracy. Integrating deep learning and machine learning techniques allows VG-GNBNet to capitalize on the strengths of both approaches, leading to superior performance. Extensive experiments demonstrate that the proposed VG-GNBNet+GNB model significantly outperforms other models, achieving an impressive accuracy score of 99.85%. This high accuracy highlights the model's potential for practical application in the agricultural sector, where the precise detection of crop health is crucial for effective disease management and yield optimization.

Producción Científica

Muhammad Usama Tanveer mail , Kashif Munir mail , Ali Raza mail , Laith Abualigah mail , Helena Garay mail helena.garay@uneatlantico.es, Luis Eduardo Prado González mail uis.prado@uneatlantico.es, Imran Ashraf mail ,

Tanveer

<a class="ep_document_link" href="/16759/1/nutrients-17-00529.pdf"><img class="ep_doc_icon" alt="[img]" src="/style/images/fileicons/text.png" border="0"/></a>

en

open

Chronotype and Cancer: Emerging Relation Between Chrononutrition and Oncology from Human Studies

Fasting–feeding timing is a crucial pattern implicated in the regulation of daily circadian rhythms. The interplay between sleep and meal timing underscores the importance of maintaining circadian alignment in order to avoid creating a metabolic environment conducive to carcinogenesis following the molecular and systemic disruption of metabolic performance and immune function. The chronicity of such a condition may support the initiation and progression of cancer through a variety of mechanisms, including increased oxidative stress, immune suppression, and the activation of proliferative signaling pathways. This review aims to summarize current evidence from human studies and provide an overview of the potential mechanisms underscoring the role of chrononutrition (including time-restricted eating) on cancer risk. Current evidence shows that the morning chronotype, suggesting an alignment between physiological circadian rhythms and eating timing, is associated with a lower risk of cancer. Also, early time-restricted eating and prolonged nighttime fasting were also associated with a lower risk of cancer. The current evidence suggests that the chronotype influences cancer risk through cell cycle regulation, the modulation of metabolic pathways and inflammation, and gut microbiota fluctuations. In conclusion, although there are no clear guidelines on this matter, emerging evidence supports the hypothesis that the role of time-related eating (i.e., time/calorie-restricted feeding and intermittent/periodic fasting) could potentially lead to a reduced risk of cancer.

Producción Científica

Justyna Godos mail , Walter Currenti mail , Raffaele Ferri mail , Giuseppe Lanza mail , Filippo Caraci mail , Evelyn Frias-Toral mail , Monica Guglielmetti mail , Cinzia Ferraris mail , Vivian Lipari mail vivian.lipari@uneatlantico.es, Stefanía Carvajal Altamiranda mail stefania.carvajal@uneatlantico.es, Fabio Galvano mail , Sabrina Castellano mail , Giuseppe Grosso mail ,

Godos