Deep Learning for Pneumonia Classification in Chest Radiography Images using Wavelet Transform DOI Open Access
Amal Azeroual, Benayad Nsiri, Taoufiq Belhoussine Drissi

и другие.

WSEAS TRANSACTIONS ON INFORMATION SCIENCE AND APPLICATIONS, Год журнала: 2023, Номер 20, С. 245 - 253

Опубликована: Сен. 8, 2023

Chronic respiratory diseases constitute a prognostic severity factor for some illnesses. A case in point is pneumonia, lung infection, whose effective management requires highly accurate diagnosis and precise treatment. Categorizing pneumonia as positive or negative does go through process of classifying chest radiography images. This task plays crucial role medical diagnostics it facilitates the detection helps making timely treatment decisions. Deep learning has shown remarkable effectiveness various imaging applications, including recognition categorization The main aim this research to compare efficacy two convolutional neural network models first model was directly trained on original images, achieving training accuracy 0.9266, whereas second images transformed using wavelets achieved 0.94. demonstrated significantly superior results terms accuracy, sensitivity, specificity.

Язык: Английский

A CNN Transfer Learning-Based Automated Diagnosis of COVID-19 From Lung Computerized Tomography Scan Slices DOI
Jaspreet Kaur, Prabhpreet Kaur

New Generation Computing, Год журнала: 2023, Номер 41(4), С. 795 - 838

Опубликована: Окт. 5, 2023

Язык: Английский

Процитировано

3
Enhanced detonators detection in X-ray baggage inspection by image manipulation and deep convolutional neural networks DOI Creative Commons

Lynda Oulhissane,

Mostefa Merah,

Simona Moldovanu

и другие.

Scientific Reports, Год журнала: 2023, Номер 13(1)

Опубликована: Авг. 31, 2023

Detecting detonators is a challenging task because they can be easily mis-classified as being harmless organic mass, especially in high baggage throughput scenarios. Of particular interest the focus on automated security X-ray analysis for detection. The complex scenarios require increasingly advanced combinations of computer-assisted vision. We propose an extensive set experiments to evaluate ability Convolutional Neural Network (CNN) models detect detonators, when quality input images has been altered through manipulation. leverage recent advances field wavelet transforms and established CNN architectures-as both these used object Various methods image manipulation are further, performance detection evaluated. Both raw manipulated with Contrast Limited Adaptive Histogram Equalization (CLAHE), transform-based mixed CLAHE RGB-wavelet method were analyzed. results showed that significant number operations, such as: edges enhancements, color information or different frequency components provided by transforms, differentiate between almost similar features. It was found wavelet-based achieved higher performance. Overall, this illustrates potential combined use deep CNNs airport applications.

Язык: Английский

Процитировано

1
Deep Learning for Pneumonia Classification in Chest Radiography Images using Wavelet Transform DOI Open Access
Amal Azeroual, Benayad Nsiri, Taoufiq Belhoussine Drissi

и другие.

WSEAS TRANSACTIONS ON INFORMATION SCIENCE AND APPLICATIONS, Год журнала: 2023, Номер 20, С. 245 - 253

Опубликована: Сен. 8, 2023

Chronic respiratory diseases constitute a prognostic severity factor for some illnesses. A case in point is pneumonia, lung infection, whose effective management requires highly accurate diagnosis and precise treatment. Categorizing pneumonia as positive or negative does go through process of classifying chest radiography images. This task plays crucial role medical diagnostics it facilitates the detection helps making timely treatment decisions. Deep learning has shown remarkable effectiveness various imaging applications, including recognition categorization The main aim this research to compare efficacy two convolutional neural network models first model was directly trained on original images, achieving training accuracy 0.9266, whereas second images transformed using wavelets achieved 0.94. demonstrated significantly superior results terms accuracy, sensitivity, specificity.

Язык: Английский

Процитировано

0