Remove Noise Artifacts Research Articles

Detection of Brain Tumour Via Reversing Hexagonal Feature Pattern for Classifying Double-Modal Brain Images

The diagnosis of brain tumours (BT) is time-consuming and heavily dependent on the radiologists' abilities. Multiple algorithms have been developed for detecting and classifying BT that are both accurate and fast. Recent years have seen an increase in the popularity of deep learning, especially when it comes to developing automated systems that can diagnose and segment BT more accurately and with less time. In this paper, a novel Brain Hexagonal Pattern Network (BHPN) has been proposed to classify the MEG and PET images into normal, benign and malignant tumours. For pre-processing, a bilateral filter is employed to remove noise artifacts from the collected MEG and PET images. To remove the outer cortical and skull region, skull stripping is used, to be implemented to raise the volume of the training datasets. The pre-processed images are segmented using the Otsu threshold algorithm to segment the BT. These segmented tumours are taken as input to the Reversing Hexagonal algorithm to generate the hexagonal feature sets with and without a segmentation mask. In order to categorize tumours into normal, benign and malignant cases, a Spiking Dilated Convolutional Neural Network (SDCNN) classifier system is implemented. The classification accuracy of the Proposed BHPN approach is 99.54%. The Proposed BHPN approach improves the overall accuracy by 1.49%, 2.52%, and 3.93% better than hybrid deep autoencoder (DAE) and Bayesian Fuzzy Clustering (BFC), Deep CNN, and Neutrosophy and Convolutional Neural Network (NS-CNN) respectively.

Lesion Detection in Optical Coherence Tomography with Transformer-Enhanced Detector.

Optical coherence tomography (OCT) is an emerging imaging tool in healthcare with common applications in ophthalmology for the detection of retinal diseases and in dentistry for the early detection of tooth decay. Speckle noise is ubiquitous in OCT images, which can hinder diagnosis by clinicians. In this paper, a region-based, deep learning framework for the detection of anomalies is proposed for OCT-acquired images. The core of the framework is Transformer-Enhanced Detection (TED), which includes attention gates (AGs) to ensure focus is placed on the foreground while identifying and removing noise artifacts as anomalies. TED was designed to detect the different types of anomalies commonly present in OCT images for diagnostic purposes and thus aid clinical interpretation. Extensive quantitative evaluations were performed to measure the performance of TED against current, widely known, deep learning detection algorithms. Three different datasets were tested: two dental and one CT (hosting scans of lung nodules, livers, etc.). The results showed that the approach verifiably detected tooth decay and numerous lesions across two modalities, achieving superior performance compared to several well-known algorithms. The proposed method improved the accuracy of detection by 16-22% and the Intersection over Union (IOU) by 10% for both dentistry datasets. For the CT dataset, the performance metrics were similarly improved by 9% and 20%, respectively.

StruNet: Perceptual and low-rank regularized transformer for medical image denoising.

Various types of noise artifacts inevitably exist in some medical imaging modalities due to limitations of imaging techniques, which impair either clinical diagnosis or subsequent analysis. Recently, deep learning approaches have been rapidly developed and applied on medical images for noise removal or image quality enhancement. Nevertheless, due to complexity and diversity of noise distribution representations in different medical imaging modalities, most of the existing deep learning frameworks are incapable to flexibly remove noise artifacts while retaining detailed information. As a result, it remains challenging to design an effective and unified medical image denoising method that will work across a variety of noise artifacts for different imaging modalities without requiring specialized knowledge in performing the task. In this paper, we propose a novel encoder-decoder architecture called Swin transformer-based residual u-shape Network (StruNet), for medical image denoising. Our StruNet adopts a well-designed block as the backbone of the encoder-decoder architecture, which integrates Swin Transformer modules with residual block in parallel connection. Swin Transformer modules could effectively learn hierarchical representations of noise artifacts via self-attention mechanism in non-overlapping shifted windows and cross-window connection, while residual block is advantageous to compensate loss of detailed information via shortcut connection. Furthermore, perceptual loss and low-rank regularization are incorporated into loss function respectively in order to constrain the denoising results on feature-level consistency and low-rank characteristics. To evaluate the performance of the proposed method, we have conducted experiments on three medical imaging modalities including computed tomography (CT), optical coherence tomography (OCT) and optical coherence tomography angiography (OCTA). The results demonstrate that the proposed architecture yields a promising performance of suppressing multiform noise artifacts existing in different imagingmodalities.

ALBAE feature extraction based lung pneumonia and cancer classification.

Lung cancer is a deadly disease showing uncontrolled proliferation of malignant cells in the lungs. If the lung cancer is detected in early stages, it can be cured before critical stage. In recent years, new technologies have gained much attention in the healthcare industry however, the unpredictable appearance of tumors, finding their presence, determining its shape, size and high discrepancy in medical images are the challenging tasks. To overcome this issue a novel Ant lion-based Autoencoders (ALbAE) model is proposed for efficient classification of lung cancer and pneumonia. Initially Computed Tomography (CT) images are pre-processed using median filters to remove noise artifacts and improving the quality of the images. Consequently, the relevant features such as image edges, pixel rates of the images and blood clots are extracted by ant lion-based autoencoder (ALbAE) technique. Finally, in classification stage, the lung CT images are classified into three different categories such as normal lung, cancer affected lung and pneumonia affected lung using Random forest technique. The effectiveness of the implemented design is estimated by different parameters such as precision, recall, Accuracy and F1-measure. The proposed approach attains 97% accuracy; 98% of recall and F-measure rate is attained through the developed design and the proposed model gains 96% of precision score. Experimental outcomes show that the proposed model performs better than existing SVM, ELM, and MLP in classifying lung cancer and pneumonia.

Skin Lesion Classification towards Melanoma Detection Using EfficientNetB3

The rise of incidences of melanoma skin cancer is a global health problem. Skin cancer, if diagnosed at an early stage, enhances the chances of a patient’s survival. Building an automated and effective melanoma classification system is the need of the hour. In this paper, an automated computer-based diagnostic system for melanoma skin lesion classification is presented using fine-tuned EfficientNetB3 model over ISIC 2017 dataset. To improve classification results, an automated image pre-processing phase is incorporated in this study, it can effectively remove noise artifacts such as hair structures and ink markers from dermoscopic images. Comparative analyses of various advanced models like ResNet50, InceptionV3, InceptionResNetV2, and EfficientNetB0-B2 are conducted to corroborate the performance of the proposed model. The proposed system also addressed the issue of model overfitting and achieved a precision of 88.00%, an accuracy of 88.13%, recall of 88%, and F1-score of 88%.

Denoising Speech Signals with Hifi-Coulomb-GANs

Recorded speech signals often contain noise that affects the quality of the signal and reduces intelligibility. Several studies have used Generative Adversarial Networks (GANs) to remove noise artifacts and improve speech intelligibility. However, GANs can suffer from gradient vanishing or gradient explosion that can reduce their effectiveness in denoising. To mitigate gradient vanishing, we applied the CoulombGAN architecture to speech denoising using a model structure similar to Hifi-GAN, the current state of the art speech denoiser. We call this new model Hifi-CoGAN. We used a WaveNet generator to denoise signals, a PostNet for general cleanup, and a Multi-Resolution Discriminator to evaluate the signal quality relative to the clean signal. Our results show that Hifi-CoGAN was able to outperform Hifi-GAN in many of the narrowband signals (signals with a limited range of frequencies) in terms of the Short-Term Objective Intelligibility (STOI) and Perceptual Evaluation of Speech Quality (PESQ) metrics. However, the model did not perform as well as Hifi-GAN with wideband noise signals (signals with a wider range of frequencies) such as white noise, so future work must be done to improve the model for these noise signals.

Studying the Role of Cerebrovascular Changes in Different Compartments in Human Brains in Hypertension Prediction

Hypertension is a major cause of mortality of millions of people worldwide. Cerebral vascular changes are clinically observed to precede the onset of hypertension. The early detection and quantification of these cerebral changes would help greatly in the early prediction of the disease. Hence, preparing appropriate medical plans to avoid the disease and mitigate any adverse events. This study aims to investigate whether studying the cerebral changes in specific regions of human brains (specifically, the anterior, and the posterior compartments) separately, would increase the accuracy of hypertension prediction compared to studying the vascular changes occurring over the entire brain’s vasculature. This was achieved by proposing a computer-aided diagnosis system (CAD) to predict hypertension based on cerebral vascular changes that occur at the anterior compartment, the posterior compartment, and the whole brain separately, and comparing corresponding prediction accuracy. The proposed CAD system works in the following sequence: (1) an MRA dataset of 72 subjects was preprocessed to enhance MRA image quality, increase homogeneity, and remove noise artifacts. (2) each MRA scan was then segmented using an automatic adaptive local segmentation algorithm. (3) the segmented vascular tree was then processed to extract and quantify hypertension descriptive vascular features (blood vessels’ diameters and tortuosity indices) the change of which has been recorded over the time span of the 2-year study. (4) a classification module used these descriptive features along with corresponding differences in blood pressure readings for each subject, to analyze the accuracy of predicting hypertension by examining vascular changes in the anterior, the posterior, and the whole brain separately. Experimental results presented evidence that studying the vascular changes that take place in specific regions of the brain, specifically the anterior compartment reported promising accuracy percentages of up to 90%. However, studying the vascular changes occurring over the entire brain still achieve the best accuracy (of up to 100%) in hypertension prediction compared to studying specific compartments.

Automated Skin Lesion Preprocessing Model of Dermoscopic Images Towards Melanoma Detection

Melanoma is lethal form of skin cancer. For unbiased diagnosis of melanoma skin cancer an automated and computerized image analysis system is prime requirement. Clinicians examine the digital photographs taken from dermoscopy lens to detect skin cancer. The images captured usually contains various noise artifacts like low contrast hair objects, ink markers, vignette effect, ebony frames. These artifacts often affect results of skin lesion segmentation and lesion classification analysis. In this paper, we propose an automated image pre-processing model to remove noise artifacts like thin & thick hair objects, surgical ink markers, dark vignette effect, ebony frames from the images. To segment required lesion region and detect lesion edges we applied Otsu thresholding and Canny edge technique. We also applied these techniques on preprocessed image and original image with noise artifacts to check the efficiency of our model. We discuss various methodologies and results acquired at each phase. Based on the experimental results, the proposed model can be used successfully in removal of noise artifacts from dermoscopic images. In future, it can be used to build a reliable and accurate computer aided diagnostic systems.

Adaptive switching filter for impulse noise removal in digital content

In this paper, a new fuzzy-rule-based impulse noise denoising method that removes unwanted artifacts and reconstructs original images is proposed. The proposed method is based on the fuzzy rule selective approach, which has four sub-methods. The proposed method can effectively remove noise artifacts that are caused by various levels of impulse noise. Simulation results show that the presented method outperforms conventional methods. The proposed method can be directly applied to various consumer electronics displays.

Exploiting sparsity and low-rank structure for the recovery of multi-slice breast MRIs with reduced sampling error

It has been shown that, magnetic resonance images (MRIs) with sparsity representation in a transformed domain, e.g. spatial finite-differences (FD), or discrete cosine transform (DCT), can be restored from undersampled k-space via applying current compressive sampling theory. The paper presents a model-based method for the restoration of MRIs. The reduced-order model, in which a full-system-response is projected onto a subspace of lower dimensionality, has been used to accelerate image reconstruction by reducing the size of the involved linear system. In this paper, the singular value threshold (SVT) technique is applied as a denoising scheme to reduce and select the model order of the inverse Fourier transform image, and to restore multi-slice breast MRIs that have been compressively sampled in k-space. The restored MRIs with SVT for denoising show reduced sampling errors compared to the direct MRI restoration methods via spatial FD, or DCT. Compressive sampling is a technique for finding sparse solutions to underdetermined linear systems. The sparsity that is implicit in MRIs is to explore the solution to MRI reconstruction after transformation from significantly undersampled k-space. The challenge, however, is that, since some incoherent artifacts result from the random undersampling, noise-like interference is added to the image with sparse representation. These recovery algorithms in the literature are not capable of fully removing the artifacts. It is necessary to introduce a denoising procedure to improve the quality of image recovery. This paper applies a singular value threshold algorithm to reduce the model order of image basis functions, which allows further improvement of the quality of image reconstruction with removal of noise artifacts. The principle of the denoising scheme is to reconstruct the sparse MRI matrices optimally with a lower rank via selecting smaller number of dominant singular values. The singular value threshold algorithm is performed by minimizing the nuclear norm of difference between the sampled image and the recovered image. It has been illustrated that this algorithm improves the ability of previous image reconstruction algorithms to remove noise artifacts while significantly improving the quality of MRI recovery.

Off-line general handwritten word recognition using an approximate BEAM matching algorithm

A recognition system for general isolated off-line handwritten words using an approximate segment-string matching algorithm is described. The fundamental paradigm employed is a character-based segment-then-recognize/match strategy. An additional user supplied contextual information in the form of a lexicon guides a graph search to estimate the most likely word image identity. This system is designed to operate robustly in the presence of document noise, poor handwriting, and lexicon errors. A pre-processing step is initially applied to the image to remove noise artifacts and normalize the handwriting. An oversegmentation approach is used to improve the likelihood of capturing the individual characters embedded in the word. A directed graph is constructed that contains many possible interpretations of the word image, many implausible. The most likely graph path and associated confidence is computed for each lexicon word to produce a final lexicon ranking. Experiments highlighting the characteristics of this algorithm are given.