Convolutional neural networks for the classification of chest X-rays in the IoT era

Almezhghwi, Khaled; Serte, Sertan; Al‐Turjman, Fadi

doi:10.1007/s11042-021-10907-y

Cited by 37 publications

(21 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To ensure the regular training of the model, we set the stride of the first residual block in ResNet-18 to 2, and the rest of the network layers of the model remained unchanged. Please refer to the relevant literature for a more detailed description of the 2 classical model architectures applied to classification tasks, the details of which are not repeated here (64,65).…”

Section: Super-resolutionmentioning

confidence: 99%

Artificial intelligence-assisted multistrategy image enhancement of chest X-rays for COVID-19 classification

Sun¹,

Ren²,

Teng³

et al. 2023

Quant Imaging Med Surg

View full text Add to dashboard Cite

Background: The coronavirus disease 2019 (COVID-19) led to a dramatic increase in the number of cases of patients with pneumonia worldwide. In this study, we aimed to develop an AI-assisted multistrategy image enhancement technique for chest X-ray (CXR) images to improve the accuracy of COVID-19 classification. Methods: Our new classification strategy consisted of 3 parts. First, the improved U-Net model with a variational encoder segmented the lung region in the CXR images processed by histogram equalization. Second, the residual net (ResNet) model with multidilated-rate convolution layers was used to suppress the bone signals in the 217 lung-only CXR images. A total of 80% of the available data were allocated for training and validation. The other 20% of the remaining data were used for testing. The enhanced CXR images containing only soft tissue information were obtained. Third, the neural network model with a residual cascade was used for the super-resolution reconstruction of low-resolution bone-suppressed CXR images. The training and testing data consisted of 1,200 and 100 CXR images, respectively. To evaluate the new strategy, improved visual geometry group (VGG)-16 and ResNet-18 models were used for the COVID-19 classification task of 2,767 CXR images. The accuracy of the multistrategy enhanced CXR images was verified through comparative experiments with various enhancement images. In terms of quantitative verification, 8-fold cross-validation was performed on the bone suppression model. In terms of evaluating the COVID-19 classification, the CXR images obtained by the improved method were used to train 2 classification models.Results: Compared with other methods, the CXR images obtained based on the proposed model had better performance in the metrics of peak signal-to-noise ratio and root mean square error. The super-resolution CXR images of bone suppression obtained based on the neural network model were also anatomically close to the real CXR images. Compared with the initial CXR images, the classification accuracy rates of the internal and external testing data on the VGG-16 model increased by 5.09% and 12.81%, respectively, while the values increased by 3.51% and 18.20%, respectively, for the ResNet-18 model. The numerical results were better than those of the single-enhancement, double-enhancement, and no-enhancement CXR images. Conclusions:The multistrategy enhanced CXR images can help to classify COVID-19 more accurately than the other existing methods.

show abstract

Section: Super-resolutionmentioning

confidence: 99%

Artificial intelligence-assisted multistrategy image enhancement of chest X-rays for COVID-19 classification

Sun¹,

Ren²,

Teng³

et al. 2023

Quant Imaging Med Surg

View full text Add to dashboard Cite

show abstract

“…Solving the optimization problem involves calculating the inner product 𝑎 𝑖 𝑇 𝑎 𝑗 between all training vectors. The initial training consists of finding Bo and Lb using the following formula: Φ(Bo) =½ 𝐵𝑜 𝑇 𝐵𝑜 is minimalized, and for wholly (𝑎 𝑖 , 𝑏 𝑖 )}, 𝑏 𝑖 (𝐵𝑜 𝑇 𝑎 𝑖 + 𝐿𝑏) greater and equal to one [24,25].…”

Section: Support Vector Machinementioning

confidence: 99%

Enhanced feature selection algorithm for pneumonia detection

Abdullah¹,

Abedi²,

Hadi³

2023

PEN

View full text Add to dashboard Cite

Pneumonia is a type of lung disease that can be detected using X-ray images. The analysis of chest X-ray images is an active research area in medical image analysis and computer-aided radiology. This research aims to improve the accuracy and efficiency of radiologists' work by providing a technique for identifying and categorizing diseases. More attention should be given to applying machine learning approaches to develop a robust chest X-ray image classification method. The typical method for detecting Pneumonia is through chest X-ray images but analyzing these images can be complex and requires the expertise of a radiographer. This paper demonstrates the feasibility of detecting the disease using chest X-ray images as datasets and a Support Vector Machine combined with a Naive Bayesian classifier, with PCA and GA as feature selection methods. The selected features are essential for training many classifiers. The proposed system achieved an accuracy of 92.26%, using 91% of the principal component. The study's result suggests that using PCA and GA for feature selection in chest X-ray image classification can achieve a good accuracy of 97.44%. Further research is needed to explore the use of other data mining models and care components to improve the accuracy and effectiveness of the system.

show abstract

“…Recently, the multiclass support vector machine is gaining popularity in deep learning particularly in image classification tasks. There have been several research papers [19][20][21][22][23] in the past few years on how multiclass SVM (L2-SVM) is being applied to classify images in deep learning with a higher level of accuracy than softmax. This work presents the novel modified deep CNN model that uses L2-SVM in place of softmax.…”

Section: Related Workmentioning

confidence: 99%

Transfer learning supported accurate assessment of multiclass cervix type images

Natarajan

Lakshmi

2022

Proc Inst Mech Eng H

View full text Add to dashboard Cite

Cervical cancer predominately affects women compared to lung, breast and endometrial cancer. Premature stage identification and proper treatment of this cancer may lead to 100% survival rate. The cervix type is very prominent in the detailed diagnosis of cervical cancer. High expertise and experienced gynecologist are required for an accurate diagnosis of cervical cancer. To reduce their burden, a model is proposed, based on deep learning that automatically classifies the cervix types. This paper presents Modified Deep Convolutional Neural Networks namely Modified VGG16 (MVGG16), Modified VGG19 (MVGG19), Modified ResNet50 (MRN50), Modified InceptionV3 (MIV3), and Modified InceptionResNetV2 (MIRNV2) for the classification of cervix type images. These modified networks are implemented using a Multiclass Support Vector Machine classifier. The performance metrics are tabulated and compared with pre-trained models. The simulation results show that MIRNV2 achieves the best performance compared to other models with an overall Accuracy of 92.91% and a Kappa score of 0.88. MIRNV2 model also gives better classification accuracy of 96.62% for type 1, 93.58% for type 2, and 95.61% for type 3 cervix images. Hence, this facilitates the application of MIRNV2 as a diagnostic tool to assist the gynecologist in the classification of cervix type images.

show abstract

Convolutional neural networks for the classification of chest X-rays in the IoT era

Cited by 37 publications

References 30 publications

Artificial intelligence-assisted multistrategy image enhancement of chest X-rays for COVID-19 classification

Artificial intelligence-assisted multistrategy image enhancement of chest X-rays for COVID-19 classification

Enhanced feature selection algorithm for pneumonia detection

Transfer learning supported accurate assessment of multiclass cervix type images

Contact Info

Product

Resources

About