Wyniki wyszukiwania - BazTech

1

AutoCovNet: Unsupervised feature learning using autoencoder and feature merging for detection of COVID-19 from chest X-ray images

Rashid Nayeeb, Hossain Md Adnan Faisal, Ali Mohammad, Islam Sukanya Mumtahina, Mahmud Tanvir, Fattah Shaikh Anowarul

Biocybernetics and Biomedical Engineering

|

2021

|

Vol. 41, no. 4

1685--1701

EN

With the onset of the COVID-19 pandemic, the automated diagnosis has become one of the most trending topics of research for faster mass screening. Deep learning-based approaches have been established as the most promising methods in this regard. However, the limitation of the labeled data is the main bottleneck of the data-hungry deep learning methods. In this paper, a two-stage deep CNN based scheme is proposed to detect COVID-19 from chest X-ray images for achieving optimum performance with limited training images. In the first stage, an encoder-decoder based autoencoder network is proposed, trained on chest X-ray images in an unsupervised manner, and the network learns to reconstruct the X-ray images. An encoder-merging network is proposed for the second stage that consists of different layers of the encoder model followed by a merging network. Here the encoder model is initialized with the weights learned on the first stage and the outputs from different layers of the encoder model are used effectively by being connected to a proposed merging network. An intelligent feature merging scheme is introduced in the proposed merging network. Finally, the encoder-merging network is trained for feature extraction of the X-ray images in a supervised manner and resulting features are used in the classification layers of the proposed architecture. Considering the final classification task, an EfficientNet-B4 network is utilized in both stages. An end to end training is performed for datasets containing classes: COVID-19, Normal, Bacterial Pneumonia, Viral Pneumonia. The proposed method offers very satisfactory performances compared to the state of the art methods and achieves an accuracy of 90:13% on the 4-class, 96:45% on a 3-class, and 99:39% on 2-class classification.

2

Klasyfikacja obiektów na podstawie ich zdjęć rentgenowskich

Nowosad Piotr, Charytanowicz Małgorzata

Journal of Computer Sciences Institute

|

2020

|

Vol. 15

206--213

EN

The main aim of the presented research was to assess the possibility of utilizing geometric features in object classifica-tion. Studies were conducted using X-ray images of kernels belonging to three different wheat varieties: Kama, Canadi-an and Rosa. As a part of the work, image processing methods were used to determine the main geometric grain parameters, including the kernel area, kernel perimeter, kernel length and kernel width. The results indicate significant differences between wheat varieties, and demonstrates the importance of their size and shape parameters in the classification process. The percentage of correctness of classification was about 92% when the k-Means algorithm was used. A classification rate of 93% was obtain using the K-Nearest Neighbour and Support Vector Machines. Herein, the Rosa variety was better recognized, whilst the Canadian and Kama varieties were less successfully differentiated.

PL

Głównym celem artykułu było zbadanie możliwości wykorzystania cech geometrycznych obiektów w procesie ich klasyfikacji. Materiał badawczy stanowiły zdjęcia rentgenowskie ziaren trzech odmian pszenicy: kama, kanadyjskiej i rosa. W ramach pracy opracowano metody pozwalające na wyznaczenie cech geometrycznych obiektów znajdujących się na obrazach cyfrowych, takich jak długość, szerokość, średnica, pole i obwód. Otrzymane wyniki wykazały istotne różnice pomiędzy parametrami charakteryzującymi kształt i wielkości poszczególnych odmian pszenicy i możliwość ich zastosowania w procesie klasyfikacji. Procent poprawnie zaklasyfikowanych ziaren za pomocą algorytmu k-średnich wynosił 92%. Nieco lepsze wyniki, rzędu 93%, uzyskano za pomocą metod K-najbliższych sąsiadów i wek-torów wspierających. Najlepiej rozróżnialną odmianą okazała się rosa w porównaniu do odmian kanadyjskiej i kama.

3

Obrazowanie rentgenowskie – nie tylko pomiar osłabienia promieniowania

Skrzyński Witold

Inżynier i Fizyk Medyczny

|

2020

|

Vol. 9, nr 1

15--18

PL

Medyczne obrazowanie rentgenowskie jest oparte na pomiarze osłabienia wiązki promieniowania przechodzącej przez tkanki pacjenta. Osłabienie wiązki nie jest jednak jedynym efektem oddziaływania promieniowania w tkankach. Oprócz osłabienia obserwujemy rozproszenie (także wsteczne), tworzenie promieniowania charakterystycznego (wymuszona fluorescencja), a nawet zmianę fazy i załamanie fali elektromagnetycznej. Artykuł zawiera krótki przegląd literatury dotyczącej możliwych zastosowań tych zjawisk w obrazowaniu medycznym.

EN

X-ray medical imaging is based on measuring the attenuation of the radiation beam passing through the patient’s tissues. However, beam attenuation is not the only effect of radiation interaction in tissues. In addition to attenuation, we can also observe scattering (including backscattering), creation of characteristic radiation (X-ray fluoerescence), and even phasechange and refraction of the electromagnetic wave. The article contains a brief review of the literature on the possible applications of these phenomena in medical imaging.

4

Klasyfikacja obrazów w tomografii komputerowej z wykorzystaniem głębokich sieci neuronowych

Podolszańska Jolanta

Inżynier i Fizyk Medyczny

|

2020

|

Vol. 9, nr 4

235--237

PL

Głębokie uczenie jest podkategorią uczenia maszynowego, które polega na tworzeniu wielowarstwowych sieci neuronowych, naśladując tym samym wykonywanie zadań przez ludzki mózg. Algorytmy głębokiego uczenia są ułożone według rosnącej złożoności, dlatego możliwe jest stworzenie systemów do analizy dużych zbiorów danych. Proces uczenia odbywa się bez nadzoru, a program buduje samodzielnie zestaw cech do rozpoznania. Artykuł przybliża na czym polega owa klasyfikacja obrazu tomograficznego.

EN

Deep learning is a subcategory of machine learning, which involves the creation of multilayer neural networks, mimicking the performance of tasks by the human brain. Deep learning algorithms are arranged according to increasing complexity, so it is possible to create systems to analyze large data sets. The learning process takes place unsupervised, and the program builds a set of features to recognize. The article presents the classification of the tomographic image.

5

Computer-aided detection of COVID-19 from X-ray images using multi-CNN and Bayesnet classifier

Abraham Bejoy, Nair Madhu S.

Biocybernetics and Biomedical Engineering

|

2020

|

Vol. 40, no. 4

1436--1445

EN

Corona virus disease-2019 (COVID-19) is a pandemic caused by novel coronavirus. COVID-19 is spreading rapidly throughout the world. The gold standard for diagnosing COVID-19 is reverse transcription-polymerase chain reaction (RT-PCR) test. However, the facility for RT-PCR test is limited, which causes early diagnosis of the disease difficult. Easily available modalities like X-ray can be used to detect specific symptoms associated with COVID-19. Pre-trained convolutional neural networks are widely used for computer-aided detection of diseases from smaller datasets. This paper investigates the effectiveness of multi-CNN, a combination of several pre-trained CNNs, for the automated detection of COVID-19 from X-ray images. The method uses a combination of features extracted from multi-CNN with correlation based feature selection (CFS) technique and Bayesnet classifier for the prediction of COVID-19. The method was tested using two public datasets and achieved promising results on both the datasets. In the first dataset consisting of 453 COVID-19 images and 497 non-COVID images, the method achieved an AUC of 0.963 and an accuracy of 91.16%. In the second dataset consisting of 71 COVID-19 images and 7 non-COVID images, the method achieved an AUC of 0.911 and an accuracy of 97.44%. The experiments performed in this study proved the effectiveness of pre-trained multi-CNN over single CNN in the detection of COVID-19.

6

A deep learning approach to detect Covid-19 coronavirus with X-Ray images

Jain Govardhan, Mittal Deepti, Thakur Daksh, Mittal Madhup K.

Biocybernetics and Biomedical Engineering

|

2020

|

Vol. 40, no. 4

1391--1405

EN

Rapid and accurate detection of COVID-19 coronavirus is necessity of time to prevent and control of this pandemic by timely quarantine and medical treatment in absence of any vaccine. Daily increase in cases of COVID-19 patients worldwide and limited number of available detection kits pose difficulty in identifying the presence of disease. Therefore, at this point of time, necessity arises to look for other alternatives. Among already existing, widely available and low-cost resources, X-ray is frequently used imaging modality and on the other hand, deep learning techniques have achieved state-of-the-art performances in computer-aided medical diagnosis. Therefore, an alternative diagnostic tool to detect COVID-19 cases utilizing available resources and advanced deep learning techniques is proposed in this work. The proposed method is implemented in four phases, viz., data augmentation, preprocessing, stage-I and stage-II deep network model designing. This study is performed with online available resources of 1215 images and further strengthen by utilizing data augmentation techniques to provide better generalization of the model and to prevent the model overfitting by increasing the overall length of dataset to 1832 images. Deep network implementation in two stages is designed to differentiate COVID-19 induced pneumonia from healthy cases, bacterial and other virus induced pneumonia on X-ray images of chest. Comprehensive evaluations have been performed to demonstrate the effectiveness of the proposed method with both (i) training-validation-testing and (ii) 5- fold cross validation procedures. High classification accuracy as 97.77%, recall as 97.14% and precision as 97.14% in case of COVID-19 detection shows the efficacy of proposed method in present need of time. Further, the deep network architecture showing averaged accuracy/ sensitivity/specificity/precision/F1-score of 98.93/98.93/98.66/96.39/98.15 with 5-fold cross validation makes a promising outcome in COVID-19 detection using X-ray images.

7

The test system for modelling imaging methods using X-Ray technologies

Mizak Wojciech, Mężyk Jordan, Jóźwiak Grzegorz

Journal of Machine Construction and Maintenance - Problemy Eksploatacji

|

2019

|

no. 1

93--100

EN

X-Ray imaging is a tool used for non-destructive inspection of the internal structure of products and measurements of geometric dimensions of products of any shape. The non-destructive nature of measurement techniques using X-Rays has enormous potential for wide application in many industries. At the Institute for Sustainable Technologies - National Research Institute in Radom, a universal research system for modelling X-Ray imaging methods was designed and manufactured, which was intended for performing automatic inspection of products using X-Rays. The developed system will be used in research and development works carried out jointly with industrial partners to develop innovative product inspection systems. The article describes the mechanical structure of the device, the control system, the ranges of parameters at which it is possible to carry out inspections of products using X-Rays, as well as exemplary test results.

PL

Obrazowanie rentgenowskie jest narzędziem wykorzystywanym do nieniszczącej kontroli wewnętrznej struktury wyrobów oraz pomiarów wymiarów geometrycznych produktów o dowolnych kształtach. Nieniszczący charakter technik pomiarowych wykorzystujących promieniowanie rentgenowskie ma ogromny potencjał do szerokiego zastosowania w wielu gałęziach przemysłu. W Instytucie Technologii Eksploatacji - Państwowym Instytucie Badawczym w Radomiu zaprojektowano i wytworzono uniwersalny system badawczy modelowania metod obrazowania X-Ray, przeznaczony do wykonywania automatycznej inspekcji wyrobów z wykorzystaniem promieniowania rentgenowskiego. Opracowany system zostanie wykorzystany w pracach badawczo-rozwojowych realizowanych wspólnie z zakładami przemysłowymi w celu opracowania innowacyjnych systemów inspekcji wyrobów. W artykule opisano konstrukcję mechaniczną urządzenia, system sterowania, podano zakresy parametrów, przy których możliwe jest prowadzenie inspekcji wyrobów z wykorzystaniem promieniowania rentgenowskiego, a także przedstawiono przykładowe wyniki badań.

8

Dynamic comparator design in 28 nm CMOS

Kaczmarczyk Piotr, Kmon Piotr

International Journal of Microelectronics and Computer Science

|

2018

|

Vol. 9, nr 4

149--154

EN

The paper presents a dynamic comparator design in 28 nm CMOS process. The proposed comparator is a main block of an asynchronous analog-to-digital converter used in a multichannel integrated circuit dedicated for X-ray imaging systems. We provide comparator’s main parameters analysis, i.e. voltage offsets, power consumption, response delay, and input-referred noise in terms of its dimensioning and biasing. The final circuit occupies 5×5 μm2 of area, consumes 17.1 fJ for single comparison with 250 ps of propagation delay, and allows to work with 4 GHz clock signal.

9

Trends in hybrid pixel detectors for X-ray imaging using deep submicron VLSI technology

Drozd A., Satława T.

Challenges of Modern Technology

|

2014

|

Vol. 5, no. 4

3--7

EN

The article covers the latest developments in pixel detectors used for X-ray imaging as well as the description of the practical solution – a multichannel integrated circuit dedicated to X-ray imaging. In general introduction a wide range of pixel detector applications is presented. The main part focuses on the challenges and new solutions for the field of X-ray imaging, including 3D integration, silicon-on-insulator and submicron technologies. Since minimization of a pixel size together with implementing more functionality are important issues in the detectors’ and integrated circuits’ design, the aspects of channel-to-channel uniformity and additional effects like charge sharing between pixels are taken into consideration. In the last section, the Authors present the application specific integrated circuit designed in 40 nm technology dedicated to X-ray detection and future prospects are discussed.