Wyniki wyszukiwania - BazTech

1

Multi-scale CNN-Swin transformer network with boundary supervision for multiclass biomarker segmentation in retinal OCT images

Fan Zhanpeng, Liu Xiaoming, Zhang Ying, Zhang Jia

Biocybernetics and Biomedical Engineering

|

2025

|

Vol. 45, iss. 3

340--356

EN

Retinal biomarker morphology is closely associated with a variety of chronic ophthalmic diseases, in which biomarker localization and segmentation in optical coherence tomography (OCT) play a key role in the diagnosis of retina-related diseases. Although great progress has been made in deep learning based OCT biomarker segmentation, several challenges still exist. Due to issues such as image noise or class imbalance, retinal biomarkers affect the model’s recognition of other biomarkers. Moreover, small biomarkers are prone to lose accuracy during downsampling. And most existing methods rely on convolutional neural networks, which make it challenging to obtain the global context due to locality of convolution. Benefiting from the Swin Transformer with powerful modeling capabilities, we propose MSCS-Net (Multi-scale CNN-Swin Network), a network for OCT biomarker segmentation, which effectively combines CNN and Swin Transformer and integrates them in parallel into a dual-encoder structure. Specifically, an edge detection path is added alongside to enhance the localization of biomarkers at the edges. For the Swin Transformer branch, considering the irregular distribution of most OCT biomarkers, a new windowing partition is performed in the Swin Transformer to capture the features more efficiently. Meanwhile, we design a Feature Dimensionality Reduction Module to extensively collect the information of small-scale biomarkers. To effectively integrate information from two scales, we design a Transformer Cross Fusion Module to finely fuse the global and local feature information from the two-branch encoders. We validate the proposed approach on local and public datasets, and the experimental results demonstrate the effectiveness of the proposed framework.

2

Advancing eye disease detection: A comprehensive study on computer-aided diagnosis with vision transformers and SHAP explainability techniques

Balaha Hossam Magdy, Hassan Asmaa El-Sayed, Ahmed Rawan Ayman, Balaha Magdy Hassan

Biocybernetics and Biomedical Engineering

|

2025

|

Vol. 45, iss. 1

23--33

EN

Eye diseases such as age-related macular degeneration (AMD) and diabetic retinopathy are common worldwide and affect millions of people. These conditions can cause severe vision problems and even lead to blindness if not treated promptly. Therefore, accurate and timely diagnosis is crucial to manage these diseases effectively and prevent irreversible vision loss. This study introduces a computer-aided diagnosis (CAD) framework for automatically detecting various eye diseases via advanced methodologies and datasets. The main focus is on classifying fundus images, which is essential for precise diagnosis and prognosis. By incorporating cutting-edge techniques such as Vision Transformers (ViTs), this study aims to improve the performance and interpretability of traditional Convolutional Neural Networks (CNNs). ViTs can capture complex patterns and long-range dependencies in fundus images, helping distinguish between different eye diseases and healthy conditions. Furthermore, the study integrates SHapley additive exPlanations (SHAP) explainability techniques to provide insights into the model’s decision-making process, enhancing trust and understanding of its predictions. The results demonstrate significant performance enhancements compared with the baseline models, with an overall accuracy of 95%. This method outperforms previous state-of-the-art methods by a considerable margin. Additionally, metrics such as precision, recall, intersection over union (IoU), and the Matthews correlation coefficient (MCC) show superior performance across various eye diseases, such as diabetic retinopathy, glaucoma, and age-related macular degeneration. These findings underscore the effectiveness and reliability of the proposed approach in automated eye disease detection, indicating its potential for clinical integration and widespread adoption in healthcare settings.

3

Classification System from Optical Coherence Tomography Using Transfer Learning

Asvial Muhamad, Silalahi Tobias Ivandito Margogo, Laagu Muh Asnoer

International Journal of Electronics and Telecommunications

|

2024

|

Vol. 70, No. 4

871-877

EN

This research aims to create a decision support system to identify retinal diseases using a four-class classification problem. To achieve this, the proposed system uses deep learning architecture to automatically recognize CNV, DME, and drusen from OCT images. The model employs two transfer learning architectures with several additional layers to classify retinal diseases. The purpose of model training, validation, and testing, the experiment uses 6,000 grayscale images labeled into four classes from the OCT data set. The Inception V3 model's proposed additional layer exhibits an increase in accuracy of 3.08% and a reduction in the loss by 0.3767. The experiment's results indicate that the Inception V3 model achieved an accuracy rate of 99.31%, and the VGG-16 model reached 98.83%, which outperformed other deep learning models using the OCT data set.

4

Multi-scale local-global transformer with contrastive learning for biomarkers segmentation in retinal OCT images

Liu Xiaoming, Ding Yuanzhe, Zhang Ying, Tang Jinshan

Biocybernetics and Biomedical Engineering

|

2024

|

Vol. 44, no. 1

231--246

EN

Quantitative analysis of biomarkers in Optical Coherence Tomography (OCT) images plays an import role in the diagnosis and treatment of retinal diseases. However, biomarker segmentation in retinal OCT images is very hard due to the large variations in size and shape of retinal biomarkers, blurred boundaries, low contrast, and speckle interference. We proposed a novel Multi-scale Local-Global Transformer network (MsLGT-Net) for biomarker segmentation in retinal OCT images. The network combines the proposed Multi-scale Fusion Attention (MFA) module, Local-Global Transformer (LGT) module, and Contrastive Learning Enhancement (CLE) module to tackle the challenges of biomarker segmentation. Specifically, the proposed MFA module aims to enhance the network’s ability to learn multi-scale features of retinal biomarkers by effectively combining the local detail information and contextual semantic information of biomarkers at different scales, and improve the representation ability for different classes of biomarkers. The LGT module is designed to learn local and global information adaptively from multi-scale fused features to address the challenge of small biomarker segmentation. In addition, to distinguish features between different types of retinal biomarkers, we propose the CLE module to enhance the feature representation of different biomarkers. Our proposed method is validated on one public dataset and one local dataset. The experimental results show that the proposed method is more effective than other state-of-theart methods.

5

Automatic method of macular diseases detection using deep cnn-gru network in oct images

Powroznik Paweł, Skublewska-Paszkowska Maria, Rejdak Robert, Nowomiejska Katarzyna

Acta Mechanica et Automatica

|

2024

|

Vol. 18, no. 4

697--706

EN

The increasing development of Deep Learning mechanism allowed ones to create semi-fully or fully automated diagnosis software solutions for medical imaging diagnosis. The convolutional neural networks are widely applied for central retinal diseases classifi-cation based on OCT images. The main aim of this study is to propose a new network, Deep CNN-GRU for classification of early-stage and end-stages macular diseases as age-related macular degeneration and diabetic macular edema (DME). Three types of disorders have been taken into consideration: drusen, choroidal neovascularization (CNV), DME, alongside with normal cases. The created automatic tool was verified on the well-known Labelled Optical Coherence Tomography (OCT) dataset. For the classifier evaluation the following measures were calculated: accuracy, precision, recall, and F1 score. Based on these values, it can be stated that the use of a GRU layer directly connected to a convolutional network plays a pivotal role in improving previously achieved results. Additionally, the proposed tool was compared with the state-of-the-art of deep learning studies performed on the Labelled OCT dataset. The Deep CNN-GRU network achieved high performance, reaching up to 98.90% accuracy. The obtained results of classification performance place the tool as one of the top solutions for diagnosing retinal diseases, both early and late stage.

6

Common path optical coherence tomography with electronic feedback for improved sensitivity

Alsnaie Khalid

Optica Applicata

|

2023

|

Vol. 53, nr 3

495--504

EN

Optical coherence tomography (OCT) imaging has become a useful tool in medical diagnosis over the past 25 years, because of its ability to visualize intracellular structures at high resolution. The main objective of this work is to add electronic feedback to the optical coherence tomography setup to increase its sensitivity. Noise added to the measured interferogram obscures some details of examined tissue layered structure. Adjusting signal power level in such a way to improve signal-to-noise ratio can help to enhance image quality. Electronic feedback is added to enhance system sensitivity. A logarithmic amplifier is included in the OCT setup to automatically adapt signal level. Moreover, the resolution of the optical spectrum analyzer is controlled according to the farthest layer detected in the A-scan. These techniques are tested showing an improvement in obtained image of a human nail.

7

Podstawowe patologie rogówki. Cz. 4, Metody diagnostyczne – OCT rogówki

Seredyka-Burduk Małgorzata, Błoch Waldemar, Stępniewski Paweł, Burduk Jakub

Optyka

|

2022

|

Nr 4

56--59

PL

Wiele współczesnych aparatów OCT (ang. optical coherence tomography) posiada możliwość zobrazowania przedniego odcinka gałki ocznej, w tym rogówki. Niejednokrotnie można spotkać się z określeniem AS-OCT, co stanowi skrót od angielskiego określenia badania OCT przedniego odcinka (ang. anterior segment, OCT). Prawidłowo i starannie wykonane badanie OCT jest cennym narzędziem diagnostycznym w schorzeniach rogówki – pozwala nie tylko uzyskać ostateczne rozpoznanie, ale także monitorować stan rogówki podczas jej leczenia.

8

Denoising methods for improving automatic segmentation in OCT images of human eye

Stankiewicz A., Marciniak T., Dąbrowski A., Stopa M., Rakowicz P., Marciniak E.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2017

|

Vol. 65, nr 1

71--78

EN

This paper presents analysis of selected noise reduction methods used in optical coherence tomography (OCT) retina images (the socalled B-scans). The tested algorithms include median and averaging filtering, anisotropic diffusion, soft wavelet thresholding, and multiframe wavelet thresholding. Precision of the denoising process was evaluated based on the results of automated retina layers segmentation, since this stage (vital for ophthalmic diagnosis) is strongly dependent on the image quality. Experiments were conducted with a set of 3D low quality scans obtained from 10 healthy patients and 10 patients with vitreoretinal pathologies. Influence of each method on the automatic image segmentation for both groups of patients is thoroughly described. Manual annotations of investigated retina layers provided by ophthalmology experts served as reference data for evaluation of the segmentation algorithm.

9

A New Idea of Fast Three-Dimensional Median Filtering for Despeckling of Optical Coherence Tomography Images

Gocławski J., Sekulska-Nalewajko J.

Image Processing & Communications

|

2015

|

Vol. 20, no. 3

25--34

EN

Median filtering has been widely used in image processing for noise removal because it can significantly reduce the power of noise while limiting edge blurring. This filtering is still a challenging task in the case of three-dimensional images containing up to a billion of voxels, especially for large size filtering windows. The authors encountered the problem when applying median filter to speckle noise reduction in optical coherence tomography images acquired by the Spark OCT systems. In the paper a new approach to the GPU (Graphics Processing Unit) based median smoothing has been proposed, which uses two-step evaluation of local intensity histograms stored in the shared memory of a graphic device. The solution is able to output about 50 million voxels per second while processing the neighbourhood of 125 voxels by Quadro K6000 graphic card configured on the Kepler architecture.

10

OCT jako nowoczesna metoda oceny wpływu profilu krawędzi miękkiej soczewki kontaktowej na zdrowie oka

Kozakiewicz N.

Optyka

|

2013

|

Nr 2

42--47

PL

Nieodpowiednio dobrane soczewki kontaktowe (SK) przyczyniają się do dyskomfortu i zmian w fizjologii oka, prowadząc do poważnych powikłań, co może w konsekwencji przyczynić się do rezygnacji z ich dalszego noszenia. Istnieją przesłanki, że na prawidłowe dopasowanie SK istotny wpływ może mieć kształt ich krawędzi. Za pomocą koherentnej tomografii optycznej (OCT) dokonano rejestracji wpływu profilu krawędzi miękkich SK na nabłonek spojówki w trzech segmentach oka (S – skroniowym, N – nosowym oraz D – dolnym) podczas noszenia soczewek o krawędziach okrągłej, ostrej i typu dłuto. Stopień ingerencji SK w nabłonek spojówki porównano po dwóch i czterech tygodniach noszenia dziennego. Podczas kolejnych wizyt odnotowano wzrost nacisku krawędzi soczewek na nabłonek spojówki, szczególnie w segmencie dolnym. Eksperyment ten stanowi próbę porównania subiektywnej oceny dopasowania SK w lampie szczelinowej z zastosowaniem fluoresceiny z obiektywną metodą pomiaru za pomocą OCT.

EN

Poorly fitted contact lenses (CL) contribute to discomfort and changes in the physiology of an eye, leading to serious complications. It is also one of the primary reasons for discontinuing their wear. We speculate that lens edge design could affect a proper fit of CL. We used optical coherent tomography (OCT) to register the impact of lens edge shape on the conjunctival epithelium in three segments of an eye (S – temporal, N – nasal, and D – bottom) and using lenses with round, sharp and chisel edge. After two and four weeks of daily lens wear, we compared the level of impact on conjunctival epithelium. During further visits we observed a significant increase in edge impact, mostly in the bottom segment. The study attempted to compare subjective evaluation of CL proper fit using biomicroscope with fluorescein and objective OCT measurement proces.

11

Metody kontroli grubości warstw wierzchnich usuwanych podczas ablacji laserowej

Marczak J.

Inżynieria Materiałowa

|

2011

|

Vol. 32, nr 3

176-189

PL

W artykule przedstawiono metody kontroli grubości warstw wierzchnich - nawarstwień usuwanych z dzieł sztuki i obiektów zabytkowych w architekturze. Omówiono również wybrane wyniki eksperymentalne. Dla wielu materiałów, a zwłaszcza tych, z których wykonane są dzieła sztuki i zalegające na nich nawarstwienie, istnieje granica związana ze stopniem oczyszczenia obiektu, jak również ryzyko uszkodzenia powierzchni substancji zabytkowej. W konserwacji dzieł sztuki obowiązuj e, podobnie jak w medycynie, zasada "przede wszystkim nie szkodzić". Omówione w artykule metody kontroli wykorzystywano podczas prac konserwatorskich związanych z usuwaniem warstw wierzchnich - nawarstwień z dzieł sztuki.

EN

The paper presents and discusses the methods of removal control of top layers - encrustation on works of art and architectural monuments. Author presents also and discuss selected experimental results. A particular limit as well as minimal risk of substrate surface damage exist for many materials, especially original works of art materials and lying on it encrustation. Similarly to the medicine, a rule "first of all not to harm" is forced also in the conservation of the works of art. Discussed control methods were utilized during conservation works connected with the removal of top surface layers - - encrustation on works of art.

12

Optical Fourier Domain Interferometry as a novel tool for biological imaging

Rumiński D., Kowalczyk A., Wojtkowski M.

Challenges of Modern Technology

|

2011

|

Vol. 2, no. 1

16--18

EN

Purpose of this work is to introduce results of biological objects measurements with Optical Coherence Tomography system working with swept source laser (OFDI system). This specified source with spectrum centred around 1040 nm is dedicated to measure posterior segment of the eye. OCT with this wavelength is expected to be next generation for this sort of systems. Although commonly used 800 nm source enables imaging retina, deeper tissue layers are not available with this wavelength. Longer wavelengths in range of infrared spectrum (e.g. 1040 nm) are being weakly scattered which enables to go down to the choroid layer [1]. Examination this structure composed mostly from blood vessels might be very important for early diagnosis of eye diseases. In this paper system setup, swept source scheme and eye images would be introduced.

13

The perspectives of applying OCT technology to the investigations of selected biological materials and processes

Rowińska Z., Gocławski J., Sekulska-Nalewajko J.

Automatyka / Akademia Górniczo-Hutnicza im. Stanisława Staszica w Krakowie

|

2010

|

T. 14, z. 3/2

749-764

EN

Optical Coherence Tomography is 3D imaging technology, which can produce high resolution cross-sectional images of biological issues in vivo and in real time. OCT perfectly fills apparent gap in depth/resolution feature space existing between confocal microscopy and ultrasound methods. The paper explains the concept of OCT method and presents the review of some measurement systems currently offered by most advanced producers and some existing applications used in medicine and biology. The authors suggested the new proposals of applying OCT to some biological investigations e.g.: the estimation of topological changes of wheat roots cultivated in the presence of heavy metals, the evaluation of spatial changes of plant tissues caused by biotic and abiotic stressors.

PL

Optyczna tomografia koherencyjna (OCT) jest trójwymiarową technologią obrazowania, która umożliwia uzyskanie wysokiej rozdzielczości obrazów przekrojów obiektów biologicznych in vivo oraz w czasie rzeczywistym. OCT stanowi pośrednie ogniwo w technologiach obrazowania 3D pod względem głębokości wnikania w materiał oraz rozdzielczości, może być zlokalizowana pomiędzy mikroskopią konfokalną a ultrasonografią. Artykuł wyjaśnia koncepcję OCT oraz prezentuje przegląd stanowisk pomiarowych oferowanych przez czołowych producentów oraz niektóre zastosowania w medycynie i biologii. Autorzy wskazali również nowe możliwości wykorzystania tej technologii do pewnych badań w dziedzinie biologii, takich jak: ocena zmian topologii korzeni pszenicy hodowanej w obecności metali ciężkich, ocena przestrzennych zmian strukturalnych tkanek roślin pod wpływem biotycznych i abiotycznych czynników stresowych.

14

Badanie niejednorodności warstwy ceramiki PLZT metodami optycznej interferometrii niskokoherentnej

Strąkowski M., Kosmowski B. B., Pluciński J.

Elektronika : konstrukcje, technologie, zastosowania

|

2008

|

Vol. 49, nr 11

83-86

PL

W artykule omówiono metodę badania wewnętrznej struktury materiałów ceramicznych z wykorzystaniem interferometrii niskokoherentnej. Przedstawiono układ pomiarowy optycznej tomografii koherentnej, wykorzystujący niskokoherentny interferometr do badania niejednorodności występujących w strukturze ośrodków rozpraszających. Zaprezentowano wyniki pomiarów grubości i niejednorodności ceramicznych warstw PLZT. Przeprowadzono rozważania nad możliwościami wykorzystania niskokoherentnych interferometrycznych technik pomiarowych do kontroli i optymalizacji procesów wytwarzania ceramicznych warstw PLZT.

EN

In this article we describe an optical method for inside materials investigation, which is based on low-coherence interferometry (LCI). The experimental LCI system applied in optical coherence tomography (OCT) for non-destructive and non-contact examination of materials inner inhomogeteities is presented. The test results of PLZT ceramics investigation are shown as well as brieff discussion about OCT capability for on-line control of PLZT ceramics manufacturing process.

15

Niskokoherentny, interferometryczny system do badania właściwości materiałów ceramicznych

Jędrzejewska-Szczerska M., Strąkowski M., Hypszer R., Kosmowski B.B.

Elektronika : konstrukcje, technologie, zastosowania

|

2007

|

Vol. 48, nr 3

39-41

PL

Przedstawiono światłowodowy system dwuwymiarowej wizualizacji niejednorodnych struktur warstw ceramicznych. Omówiono projekt i fizyczną realizację układu wykorzystującego optyczną, niskokoherentną reflektometrię optyczną, która umożliwia nieinwazyjne i bezkontaktowe obrazowanie wewnętrznych struktur różnych materiałów silnie rozpraszających promieniowanie. Przedstawiono przykładowe wyniki pomiarów wewnętrznych warstw ceramiki LSFO. Dodatkowo przeanalizowano możliwości poprawy rozdzielczości pomiaru systemu OCT przez zastosowanie syntezowanych źródeł promieniowania.

EN

In this paper we present a fiber based Optical Coherence Tomography system for two-dimensional visualization of LSFO ceramic subsurface inhomogeneities. The OCT system is capable for nondestructive and non-contact investigation of highly scattering materials like ceramics. The experimental results of investigation of LSFO ceramic structures has been presented. The usefulness of synthesized light source for improving the measurement resolution has been described.

16

Zastosowanie optycznej tomografii niskokoherentnej do 2-wymiarowego obrazowania cienkich warstw polimerów

Strąkowski M., Jędrzejewska-Szczerska M., Maciejewski M., Hypszer R., Pluciński J., Kosmowski B.

Elektronika : konstrukcje, technologie, zastosowania

|

2006

|

Vol. 47, nr 5

67-68

PL

Przedstawiono zasadę działania i podstawowe właściwości systemu OCT (Optical Coherence Tomography). Na przykładzie folii polietylenowej przedyskutowano możliwości wykorzystania systemu do badanie wewnętrznej struktury warstwowej obiektów niebiologicznych. Omówiono wyniki przeprowadzonych badań nad warstwowymi materiałami przeźroczystymi.

EN

An Optical Low-Coherence Tomography (OCT) is a novel optical measurement technique, which enables non-destructive and non-contact investigation of multilayer structures. In this paper the authors present a laboratory OCT system for surface and subsurface investigation of scattering technical objects such as polymer layers. Preliminary test results on subsurface technical objects investigation using OCT system have been presented and discussed.

17

Światłowodowa koherentna tomografia optyczna (OCT) w zastosowaniach technicznych

Maciejewski M., Strąkowski M., Pluciński J., Kosmowski B. B., Jędrzejewska-Szczerska M., Hypszer R.

Elektronika : konstrukcje, technologie, zastosowania

|

2005

|

Vol. 46, nr 11

24-25

PL

Koherentna tomografia optyczna (ang. Optical Coherence Tomography - OCT) jest szybko rozwijającą się metodą obrazowania przekrojów wewnętrznych struktur. Pierwotnie używana do celów biomedycznych, może znaleźć szerokie zastosowanie do badań materiałowych. Jest to uniwersalna technika pomiarowa charakteryzująca się dużą rozdzielczością (rzędu pojedyńczych mikrometrów) połączoną z możliwością obrazowania optycznych ośrodków silnie rozpraszających światło. W artykule przedstawiono możliwe zastosowania OCT do pomiarów technicznych oraz uzyskane wstępne wyniki badań przeprowadzonych na próbkach materiałów warstwowych silnie rozpraszających światło.

EN

Optical coherence tomography (OCT) is rapidly developing method for internal structures imaging. First used in biomedical applications, it can be also applicable for technical materials investigation. It is universal measuring technique with high resolution (single micrometers) combined with possibility of imaging highly scattering media. The article presents possible optical coherence tomography applications to technical measurements and preliminary measurements of highly scattering layered materials.