Geometrical X-lets for Image Denoising

Khodabandeh, Zahra; Rabbani, Hossein; Mehri, Alireza

doi:10.1109/embc.2019.8856318

Cited by 4 publications

(5 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Darooei et al 16 determined that, for OCT B-scan segmentation using dice coefficient and Jaccard index, contourlet yields optimal results. Khodabandeh et al 28 observed diverse performance in noise reduction, where DTCW transform excels in Structural Similarity Index (SSIM), and 2D-DWT in Edge Preservation (EP) and Texture Preservation (TP). This suggests the effectiveness of different X-lets based on distinct criteria and image characteristics.…”

Section: Discussionmentioning

confidence: 99%

“…Given the focus of the research on OCT images, it is preferable to employ transforms that have demonstrated effective performance in this context. According to Khodabandeh et al 28 and taking into account the geometric structure of OCT images, primarily characterized by lines at zero and ± 45 angles, it appears that 2D discrete wavelet transform (2D-DWT), dual tree complex wavelet (DTCW) transform, and contourlet transform exhibit the capability to adequately decompose these images, offering desirable features. In their research, Circlet transform and Ellipselet transform also demonstrated commendable performance in certain evaluation parameters.…”

mentioning

confidence: 99%

“…In their research, Circlet transform and Ellipselet transform also demonstrated commendable performance in certain evaluation parameters. While Khodabandeh et al 28 offered a relatively comprehensive analysis of the effectiveness of these transforms in noise reduction application, it's essential to acknowledge that the application of X-let sub-bands in image classification task is distinctly different. In classification, the objective is to utilize all acquired basis functions simultaneously, requiring the management of different sizes to be incorporated together as input for the feature extraction model.…”

mentioning

confidence: 99%

“…In classification, the objective is to utilize all acquired basis functions simultaneously, requiring the management of different sizes to be incorporated together as input for the feature extraction model. Khodabandeh et al 28 did not encounter this challenge as all denoising processes, such as thresholding on the coefficients of X-lets, were conducted separately on each basis, eliminating the necessity to use them concurrently in model training. Furthermore, in that study, basis functions were not utilized as features or inputs for the models.…”

mentioning

confidence: 99%

See 3 more Smart Citations

A new convolutional neural network based on combination of circlets and wavelets for macular OCT classification

Arian,

Vard,

Kafieh

et al. 2023

Sci Rep

Self Cite

View full text Add to dashboard Cite

Artificial intelligence (AI) algorithms, encompassing machine learning and deep learning, can assist ophthalmologists in early detection of various ocular abnormalities through the analysis of retinal optical coherence tomography (OCT) images. Despite considerable progress in these algorithms, several limitations persist in medical imaging fields, where a lack of data is a common issue. Accordingly, specific image processing techniques, such as time–frequency transforms, can be employed in conjunction with AI algorithms to enhance diagnostic accuracy. This research investigates the influence of non-data-adaptive time–frequency transforms, specifically X-lets, on the classification of OCT B-scans. For this purpose, each B-scan was transformed using every considered X-let individually, and all the sub-bands were utilized as the input for a designed 2D Convolutional Neural Network (CNN) to extract optimal features, which were subsequently fed to the classifiers. Evaluating per-class accuracy shows that the use of the 2D Discrete Wavelet Transform (2D-DWT) yields superior outcomes for normal cases, whereas the circlet transform outperforms other X-lets for abnormal cases characterized by circles in their retinal structure (due to the accumulation of fluid). As a result, we propose a novel transform named CircWave by concatenating all sub-bands from the 2D-DWT and the circlet transform. The objective is to enhance the per-class accuracy of both normal and abnormal cases simultaneously. Our findings show that classification results based on the CircWave transform outperform those derived from original images or any individual transform. Furthermore, Grad-CAM class activation visualization for B-scans reconstructed from CircWave sub-bands highlights a greater emphasis on circular formations in abnormal cases and straight lines in normal cases, in contrast to the focus on irrelevant regions in original B-scans. To assess the generalizability of our method, we applied it to another dataset obtained from a different imaging system. We achieved promising accuracies of 94.5% and 90% for the first and second datasets, respectively, which are comparable with results from previous studies. The proposed CNN based on CircWave sub-bands (i.e. CircWaveNet) not only produces superior outcomes but also offers more interpretable results with a heightened focus on features crucial for ophthalmologists.

show abstract

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

A new convolutional neural network based on combination of circlets and wavelets for macular OCT classification

Arian,

Vard,

Kafieh

et al. 2023

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…Wavelet based denoising 5 11,[16][17][18]50 Threshold estimation 2 10,51 Shrinkage rules 2 17,19 Intra and inter scale dependencies based denoising 2 52,53 Image denoising based on extended versions of transform 6 14,21,51,[54][55][56] Block-matching and 3D filtering (BM3D) 4 15,25,57,58 image patches from a set of training images and uses this dictionary to denoise new images. This method can effectively reduce noise while preserving image details and structures.…”

Section: Number Of Studiesmentioning

confidence: 99%

CT image denoising methods for image quality improvement and radiation dose reduction

Sadia,

Chen,

Zhang

2024

J Applied Clin Med Phys

View full text Add to dashboard Cite

With the ever‐increasing use of computed tomography (CT), concerns about its radiation dose have become a significant public issue. To address the need for radiation dose reduction, CT denoising methods have been widely investigated and applied in low‐dose CT images. Numerous noise reduction algorithms have emerged, such as iterative reconstruction and most recently, deep learning (DL)‐based approaches. Given the rapid advancements in Artificial Intelligence techniques, we recognize the need for a comprehensive review that emphasizes the most recently developed methods. Hence, we have performed a thorough analysis of existing literature to provide such a review. Beyond directly comparing the performance, we focus on pivotal aspects, including model training, validation, testing, generalizability, vulnerability, and evaluation methods. This review is expected to raise awareness of the various facets involved in CT image denoising and the specific challenges in developing DL‐based models.

show abstract

A Lightweight Mimic Convolutional Auto-Encoder for Denoising Retinal Optical Coherence Tomography Images

Tajmirriahi

Kafieh

Amini

et al. 2021

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

Geometrical X-lets for Image Denoising

Cited by 4 publications

References 27 publications

A new convolutional neural network based on combination of circlets and wavelets for macular OCT classification

A new convolutional neural network based on combination of circlets and wavelets for macular OCT classification

CT image denoising methods for image quality improvement and radiation dose reduction

A Lightweight Mimic Convolutional Auto-Encoder for Denoising Retinal Optical Coherence Tomography Images

Contact Info

Product

Resources

About