A multichannel EfficientNet deep learning-based stacking ensemble approach for lung disease detection using chest X-ray images

Ravi, Vinayakumar; Acharya, Vasundhara; Alazab, Mamoun

doi:10.1007/s10586-022-03664-6

Cited by 33 publications

(7 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It also manages to be effective and support hardware accelerators like GPUs. Performing pulmonary function testing on individuals with ILD is difficult during the present SARS-CoV-2 outbreak [28]. Specialists have advocated spirometry and video consultations at home.…”

Section: Methodsmentioning

confidence: 99%

Segmentation and Pre-processing of Interstitial Lung Disease using Deep Learning Model

Yadlapalli¹,

Bhavana

2022

SCPE

View full text Add to dashboard Cite

Medical image processing involves using and examining 3D human body images, which are most frequently acquired through a computed tomography scanner, to diagnose disorders. Medical image process- ing helps radiologists, engineers, and clinicians better comprehend the anatomy of specific patients or groups of patients. Due to recent advancements in deep learn ing techniques, the study of medical image analysis is now a quickly expanding area of research. Interstitial Lung Disease is a chronic lung disease that worsens with time. This condition cannot be completely treated when the lungs have been damaged. Early detection, on the other hand, aids in the control of the disease. It causes lung scarring as a result. The first methodology characterizes lung tissue utilizing first order statistics, grey live occurrence, run length matrices, and fractal analysis. It was suggested by Uppaluri et al in one instance. In the pre-processing step, patients' CT scans are presented using various color map models for better understanding of data-set. and also for determining the patients final Force Vital Capacity and Confidence values using a Pytorch model with leaky relu activation function. These variables can be used to determine whether a person has a disease. Segmentation is a crucial stage in employing a computer assisted diagnosis system to estimate interstitial lung disease. Accurate segmentation of aberrant lung is essential for a trustworthy computer-aided illness diagnosis. Using separate training, validation, and test sets, we proposed an efficient deep learning model using Unet architecture and Densenet121 to segment lungs with Interstitial Lung Disease. The proposed segmentation model distinguishes the exact lung region from the ct slice background. To train and evaluate the algo rithm, 176 sparsely annotated Computed Tomography scans were utilized. The training was completed in a supervised and end to end manner. Contrary to current approaches, the suggested method yields accurate segmentation results without the requirement for re-initialization. We were able to achieve an accuracy of 92.59 percent after training the proposed model with Nvidia's CUDA GPU.

show abstract

Section: Methodsmentioning

confidence: 99%

Segmentation and Pre-processing of Interstitial Lung Disease using Deep Learning Model

Yadlapalli¹,

Bhavana

2022

SCPE

View full text Add to dashboard Cite

show abstract

“…The adoption of FFNN achieved a noteworthy accuracy of 92.5%, contributing to the understanding of how architectural modifications impact the classification of medical images. Ravi et al [18] focused on prominent convolutional neural network architectures-GoogleNet, ResNet, and VGG-each renowned for its unique contributions to image classification. The remarkable accuracy of 95.05% achieved with the InceptionV3 architecture emphasizes the importance of selecting state-of-the-art models for achieving high-performance outcomes.…”

Section: Literature Reviewmentioning

confidence: 99%

Transformative Approach for Prediction of Lung Disease Using Lung X-ray Images

Mehra,

Singh,

Kumar

et al. 2024

Preprint

View full text Add to dashboard Cite

Lung diseases pose a significant global health challenge, underscoring the critical need for prompt and precise diagnoses to facilitate effective treatments and enhance patient outcomes. In this research paper, we introduce an innovative method for the prediction of lung diseases by harnessing the capabilities of deep learning techniques, thereby streamlining and augmenting the diagnostic process. The Investigation commences by assembling an extensive dataset of chest Xray images sourced from diverse origins, encompassing both normal and diseased instances. Subsequently, we employ a specialized Pix2Pix Generative Adversarial Network architecture tailored for image classification. This network is meticulously trained on the comprehensive dataset, fine-tuning its abilities to discern distinctive features associated with a range of lung diseases, including pneumonia, tuberculosis, and lung cancer. The Empirical findings underscore the efficacy of the approach in diagnosing lung diseases, showcasing notable levels of accuracy, sensitivity, and specificity. Furthermore, we employ interpretability techniques to pinpoint the regions within the X-ray images that significantly contribute to the predictions, bolstering the transparency and credibility of the Model. This research presents a promising avenue for automating the diagnosis of lung diseases, with the potential to reduce human error and enhance patient care significantly. It holds the promise of aiding in early detection and intervention, potentially saving lives and alleviating the burden of lung diseases on healthcare systems worldwide. The integration of DenseNet architecture into our predictive model significantly enhances the accuracy and efficiency of diagnosing lung diseases from X-ray images. DenseNet’s interconnected layers facilitate collaborative learning, enabling the model to discern intricate details crucial for precise classifications. The adoption of DenseNet not only amplifies diagnostic precision but also paves the way for future strides in medical image analysis, particularly in advancing respiratory health diagnostics. Along with this under the investigation comprising the chest X-ray images in JPEG format, it is systematically categorized into train, test, and val directories, each further subdivided into folders representing Pneumonia and Normal classes. These images depict anterior-posterior chest radiographs obtained from pediatric patients aged one to five years at the Guangzhou Women and Children’s Medical Center, forming an integral part of routine clinical care for this specific demographic. To ensure the dataset’s integrity, a rigorous initial screening process is applied to exclude any low-quality or unreadable scans. Subsequently, two expert physicians meticulously evaluate and grade the diagnostic quality of the remaining images. Only those that successfully pass this comprehensive scrutiny are deemed suitable for the subsequent training of the artificial intelligence system. This meticulous curation underscores the reliability and high quality of the dataset, emphasizing its potential for advancing medical image analysis within the context of pediatric chest radiographs.

show abstract

“…One member of the family is EfficientNet B2, which is larger and stronger than EfficientNet B0 but smaller than EfficientNet B3 and superior [25][26]. The result of EfficientNet B2 for image feature extraction is represented by (1408) [27]. The MBConvBlock module, used to improve model efficiency, comprises several key elements.…”

Section: • Efficientnet B2mentioning

confidence: 99%

Optimizing Handwritten Arabic Character Recognition: Feature Extraction, Concatenation, and PSO-Based Feature Selection.

Rouabhi,

Belbachir,

Tlemsani

2023

Preprint

View full text Add to dashboard Cite

Recognizing and writing Arabic characters correctly is a difficult task, especially for non-native speakers. With the growing trend toward digital teaching and distance learning, there is a need for efficient and accurate automatic recognition of Arabic characters. The work investigates three distinct approaches to address this challenge. Firstly, features are extracted from handwritten Arabic character images using two pre-trained models: EfficientNet B2 and DenseNet 201. These extracted features are then employed for classification utilizing various classifiers. Secondly, the features obtained from both models are concatenated to enhance classification performance by leveraging the unique representations learned by each model. Lastly, the Particle Swarm Optimization (PSO) algorithm is integrated into the feature selection process, identifying the most relevant features that contribute to improved classification accuracy. The best test accuracies obtained were 85.94% in the first experiment, 88.73% in the second, and 90.20% in the third.

show abstract

A multichannel EfficientNet deep learning-based stacking ensemble approach for lung disease detection using chest X-ray images

Cited by 33 publications

References 58 publications

Segmentation and Pre-processing of Interstitial Lung Disease using Deep Learning Model

Segmentation and Pre-processing of Interstitial Lung Disease using Deep Learning Model

Transformative Approach for Prediction of Lung Disease Using Lung X-ray Images

Optimizing Handwritten Arabic Character Recognition: Feature Extraction, Concatenation, and PSO-Based Feature Selection.

Contact Info

Product

Resources

About