Non-local MRI denoising using random sampling

Deep Learning-Based Denoising of CEST MR Data: A Feasibility Study on Applying Synthetic Phantoms in Medical Imaging

Chemical Exchange Saturation Transfer (CEST) magnetic resonance imaging (MRI) provides a novel method for analyzing biomolecule concentrations in tissues without exogenous contrast agents. Despite its potential, achieving a high signal-to-noise ratio (SNR) is imperative for detecting small CEST effects. Traditional metrics such as Magnetization Transfer Ratio Asymmetry (MTRasym) and Lorentzian analyses are vulnerable to image noise, hampering their precision in quantitative concentration estimations. Recent noise-reduction algorithms like principal component analysis (PCA), nonlocal mean filtering (NLM), and block matching combined with 3D filtering (BM3D) have shown promise, as there is a burgeoning interest in the utilization of neural networks (NNs), particularly autoencoders, for imaging denoising. This study uses the Bloch-McConnell equations, which allow for the synthetic generation of CEST images and explores NNs efficacy in denoising these images. Using synthetically generated phantoms, autoencoders were created, and their performance was compared with traditional denoising methods using various datasets. The results underscored the superior performance of NNs, notably the ResUNet architectures, in noise identification and abatement compared to analytical approaches across a wide noise gamut. This superiority was particularly pronounced at elevated noise intensities in the in vitro data. Notably, the neural architectures significantly improved the PSNR values, achieving up to 35.0, while some traditional methods struggled, especially in low-noise reduction scenarios. However, the application to the in vivo data presented challenges due to varying noise profiles. This study accentuates the potential of NNs as robust denoising tools, but their translation to clinical settings warrants further investigation.

MR Image Denoising and Super-Resolution Using Regularized Reverse Diffusion

Patient scans from MRI often suffer from noise, which hampers the diagnostic capability of such images. As a method to mitigate such artifacts, denoising is largely studied both within the medical imaging community and beyond the community as a general subject. However, recent deep neural network-based approaches mostly rely on the minimum mean squared error (MMSE) estimates, which tend to produce a blurred output. Moreover, such models suffer when deployed in real-world situations: out-of-distribution data, and complex noise distributions that deviate from the usual parametric noise models. In this work, we propose a new denoising method based on score-based reverse diffusion sampling, which overcomes all the aforementioned drawbacks. Our network, trained only with coronal knee scans, excels even on out-of-distribution in vivo liver MRI data, contaminated with a complex mixture of noise. Even more, we propose a method to enhance the resolution of the denoised image with the same network. With extensive experiments, we show that our method establishes state-of-the-art performance while having desirable properties which prior MMSE denoisers did not have: flexibly choosing the extent of denoising, and quantifying uncertainty.

Image Denoising Using Non-Local Means (NLM) Approach in Magnetic Resonance (MR) Imaging: A Systematic Review

The non-local means (NLM) noise reduction algorithm is well known as an excellent technique for removing noise from a magnetic resonance (MR) image to improve the diagnostic accuracy. In this study, we undertook a systematic review to determine the effectiveness of the NLM noise reduction algorithm in MR imaging. A systematic literature search was conducted of three databases of publications dating from January 2000 to March 2020; of the 82 publications reviewed, 25 were included in this study. The subjects were categorized into four major frameworks and analyzed for each research result. Research in NLM noise reduction for MR images has been increasing worldwide; however, it was found to have slightly decreased since 2016. It was found that the NLM technique was most frequently used on brain images taken using the general MR imaging technique; these were most frequently performed during simultaneous real and simulated experimental studies. In particular, comparison parameters were frequently used to evaluate the effectiveness of the algorithm on MR images. The ultimate goal is to provide an accurate method for the diagnosis of disease, and our conclusion is that the NLM noise reduction algorithm is a promising method of achieving this goal.

Non-Local SVD Denoising of MRI Based on Sparse Representations

Magnetic Resonance (MR) Imaging is a diagnostic technique that produces noisy images, which must be filtered before processing to prevent diagnostic errors. However, filtering the noise while keeping fine details is a difficult task. This paper presents a method, based on sparse representations and singular value decomposition (SVD), for non-locally denoising MR images. The proposed method prevents blurring, artifacts, and residual noise. Our method is composed of three stages. The first stage divides the image into sub-volumes, to obtain its sparse representation, by using the KSVD algorithm. Then, the global influence of the dictionary atoms is computed to upgrade the dictionary and obtain a better reconstruction of the sub-volumes. In the second stage, based on the sparse representation, the noise-free sub-volume is estimated using a non-local approach and SVD. The noise-free voxel is reconstructed by aggregating the overlapped voxels according to the rarity of the sub-volumes it belongs, which is computed from the global influence of the atoms. The third stage repeats the process using a different sub-volume size for producing a new filtered image, which is averaged with the previously filtered images. The results provided show that our method outperforms several state-of-the-art methods in both simulated and real data.

A review on brain tumor diagnosis from MRI images: Practical implications, key achievements, and lessons learned

The successful early diagnosis of brain tumors plays a major role in improving the treatment outcomes and thus improving patient survival. Manually evaluating the numerous magnetic resonance imaging (MRI) images produced routinely in the clinic is a difficult process. Thus, there is a crucial need for computer-aided methods with better accuracy for early tumor diagnosis. Computer-aided brain tumor diagnosis from MRI images consists of tumor detection, segmentation, and classification processes. Over the past few years, many studies have focused on traditional or classical machine learning techniques for brain tumor diagnosis. Recently, interest has developed in using deep learning techniques for diagnosing brain tumors with better accuracy and robustness. This study presents a comprehensive review of traditional machine learning techniques and evolving deep learning techniques for brain tumor diagnosis. This review paper identifies the key achievements reflected in the performance measurement metrics of the applied algorithms in the three diagnosis processes. In addition, this study discusses the key findings and draws attention to the lessons learned as a roadmap for future research.

Adaptive anatomical preservation optimal denoising for radiation therapy daily MRI

Low-field magnetic resonance imaging (MRI) has recently been integrated with radiation therapy systems to provide image guidance for daily cancer radiation treatments. The main benefit of the low-field strength is minimal electron return effects. The main disadvantage of low-field strength is increased image noise compared to diagnostic MRIs conducted at 1.5 T or higher. The increased image noise affects both the discernibility of soft tissues and the accuracy of further image processing tasks for both clinical and research applications, such as tumor tracking, feature analysis, image segmentation, and image registration. An innovative method, adaptive anatomical preservation optimal denoising (AAPOD), was developed for optimal image denoising, i.e., to maximally reduce noise while preserving the tissue boundaries. AAPOD employs a series of adaptive nonlocal mean (ANLM) denoising trials with increasing denoising filter strength (i.e., the block similarity filtering parameter in the ANLM algorithm), and then detects the tissue boundary losses on the differences of sequentially denoised images using a zero-crossing edge detection method. The optimal denoising filter strength per voxel is determined by identifying the denoising filter strength value at which boundary losses start to appear around the voxel. The final denoising result is generated by applying the ANLM denoising method with the optimal per-voxel denoising filter strengths. The experimental results demonstrated that AAPOD was capable of reducing noise adaptively and optimally while avoiding tissue boundary losses. AAPOD is useful for improving the quality of MRIs with low-contrast-to-noise ratios and could be applied to other medical imaging modalities, e.g., computed tomography.