Multiple sclerosis lesion segmentation from brain MRI using U-Net based on wavelet pooling

Artificial intelligence applied to MRI data to tackle key challenges in multiple sclerosis

Artificial intelligence (AI) is the branch of science aiming at creating algorithms able to carry out tasks that typically require human intelligence. In medicine, there has been a tremendous increase in AI applications thanks to increasingly powerful computers and the emergence of big data repositories. Multiple sclerosis (MS) is a chronic autoimmune condition affecting the central nervous system with a complex pathogenesis, a challenging diagnostic process strongly relying on magnetic resonance imaging (MRI) and a high and largely unexplained variability across patients. Therefore, AI applications in MS have the great potential of helping us better support the diagnosis, find markers for prognosis to eventually design more powerful randomised clinical trials and improve patient management in clinical practice and eventually understand the mechanisms of the disease. This topical review aims to summarise the recent advances in AI applied to MRI data in MS to illustrate its achievements, limitations and future directions.

Clinical applications of deep learning in neuroinflammatory diseases: A scoping review

BACKGROUND

Deep learning (DL) is an artificial intelligence technology that has aroused much excitement for predictive medicine due to its ability to process raw data modalities such as images, text, and time series of signals.

OBJECTIVES

Here, we intend to give the clinical reader elements to understand this technology, taking neuroinflammatory diseases as an illustrative use case of clinical translation efforts. We reviewed the scope of this rapidly evolving field to get quantitative insights about which clinical applications concentrate the efforts and which data modalities are most commonly used.

METHODS

We queried the PubMed database for articles reporting DL algorithms for clinical applications in neuroinflammatory diseases and the radiology.healthairegister.com website for commercial algorithms.

RESULTS

The review included 148 articles published between 2018 and 2024 and five commercial algorithms. The clinical applications could be grouped as computer-aided diagnosis, individual prognosis, functional assessment, the segmentation of radiological structures, and the optimization of data acquisition. Our review highlighted important discrepancies in efforts. The segmentation of radiological structures and computer-aided diagnosis currently concentrate most efforts with an overrepresentation of imaging. Various model architectures have addressed different applications, relatively low volume of data, and diverse data modalities. We report the high-level technical characteristics of the algorithms and synthesize narratively the clinical applications. Predictive performances and some common a priori on this topic are finally discussed.

CONCLUSION

The currently reported efforts position DL as an information processing technology, enhancing existing modalities of paraclinical investigations and bringing perspectives to make innovative ones actionable for healthcare.

MBUTransNet: multi-branch U-shaped network fusion transformer architecture for medical image segmentation

PURPOSE

Recently, transformers have been adopted to computer vision applications and achieve great success in image segmentation. However by simply applying transformers to medical segmentation task it is hard to achieve much higher accuracy than by traditional U-shaped network structures, which are based on CNNs and has been extensively researched. On the other hand, CNN structure pays more attention to local information and ignores global information, which is very important for the medical image segmentation dataset with cell scattered background. This motivates us to explore the feasibility of using U-shape effective fusion transformer network architectures for medical image segmentation tasks.

METHODS

In this paper, we propose a multibranch U-shaped structure fusion transformer network (MBUTransNet), which consists of two distinct branches. In branch 1, Coordinate attention transformer is designed to extract long-term dependency information through weight coordinates. In branch 2, small U-net blocks and multiscale feature fusion block are proposed to replace convolution blocks of each layer and fuse the feature maps from different layers, respectively.

RESULTS

Our experiments demonstrate that the proposed MBUTransNet has achieved a 0.076 and 0.1269 improvement in DICE compared to the previous best method on MoNuSeg and Synapse multiorgan segmentation dataset, respectively, while the model parameters will be no significant increase.

CONCLUSION

Without bells and whistles, MBUTransNet achieves better performance on medical image datasets, including medical cell segmentation and abdominal organs segmentation. Compared with transformer-based methods, our proposed model also obtains quite competitive parameters.

How far MS lesion detection and segmentation are integrated into the clinical workflow? A systematic review

INTRODUCTION

Over the past few years, the deep learning community has developed and validated a plethora of tools for lesion detection and segmentation in Multiple Sclerosis (MS). However, there is an important gap between validating models technically and clinically. To this end, a six-step framework necessary for the development, validation, and integration of quantitative tools in the clinic was recently proposed under the name of the Quantitative Neuroradiology Initiative (QNI).

AIMS

Investigate to what extent automatic tools in MS fulfill the QNI framework necessary to integrate automated detection and segmentation into the clinical neuroradiology workflow.

METHODS

Adopting the systematic Cochrane literature review methodology, we screened and summarised published scientific articles that perform automatic MS lesions detection and segmentation. We categorised the retrieved studies based on their degree of fulfillment of QNI's six-steps, which include a tool's technical assessment, clinical validation, and integration.

RESULTS

We found 156 studies; 146/156 (94%) fullfilled the first QNI step, 155/156 (99%) the second, 8/156 (5%) the third, 3/156 (2%) the fourth, 5/156 (3%) the fifth and only one the sixth.

CONCLUSIONS

To date, little has been done to evaluate the clinical performance and the integration in the clinical workflow of available methods for MS lesion detection/segmentation. In addition, the socio-economic effects and the impact on patients' management of such tools remain almost unexplored.

Multi-class semantic segmentation of breast tissues from MRI images using U-Net based on Haar wavelet pooling

MRI images used in breast cancer diagnosis are taken in a lying position and therefore are inappropriate for reconstructing the natural breast shape in a standing position. Some studies have proposed methods to present the breast shape in a standing position using an ordinary differential equation of the finite element method. However, it is difficult to obtain meaningful results because breast tissues have different elastic moduli. This study proposed a multi-class semantic segmentation method for breast tissues to reconstruct breast shapes using U-Net based on Haar wavelet pooling. First, a dataset was constructed by labeling the skin, fat, and fibro-glandular tissues and the background from MRI images taken in a lying position. Next, multi-class semantic segmentation was performed using U-Net based on Haar wavelet pooling to improve the segmentation accuracy for breast tissues. The U-Net effectively extracted breast tissue features while reducing image information loss in a subsampling stage using multiple sub-bands. In addition, the proposed network is robust to overfitting. The proposed network showed a mIOU of 87.48 for segmenting breast tissues. The proposed networks demonstrated high-accuracy segmentation for breast tissue with different elastic moduli to reconstruct the natural breast shape.

Unsupervised anomaly detection in brain MRI: Learning abstract distribution from massive healthy brains

PURPOSE

To develop a general unsupervised anomaly detection method based only on MR images of normal brains to automatically detect various brain abnormalities.

MATERIALS AND METHODS

In this study, a novel method based on three-dimensional deep autoencoder network is proposed to automatically detect and segment various brain abnormalities without being trained on any abnormal samples. A total of 578 normal T2w MR volumes without obvious abnormalities were used for model training and validation. The proposed 3D autoencoder was evaluated on two different datasets (BraTs dataset and in-house dataset) containing T2w volumes from patients with glioblastoma, multiple sclerosis and cerebral infarction. Lesions detection and segmentation performance were reported as AUC, precision-recall curve, sensitivity, and Dice score.

RESULTS

In anomaly detection, AUCs for three typical lesions were as follows: glioblastoma, 0.844; multiple sclerosis, 0.858; cerebral infarction, 0.807. In anomaly segmentation, the mean Dice for glioblastomas was 0.462. The proposed network also has the ability to generate an anomaly heatmap for visualization purpose.

CONCLUSION

Our proposed method was able to automatically detect various brain anomalies such as glioblastoma, multiple sclerosis, and cerebral infarction. This work suggests that unsupervised anomaly detection is a powerful approach to detect arbitrary brain abnormalities without labeled samples. It has the potential to support diagnostic workflow in radiology as an automated tool for computer-aided image analysis.

Texture and Materials Image Classification Based on Wavelet Pooling Layer in CNN

Convolutional Neural Networks (CNNs) have recently been proposed as a solution in texture and material classification in computer vision. However, inside CNNs, the internal layers of pooling often cause a loss of information and, therefore, is detrimental to learning the architecture. Moreover, when considering images with repetitive and essential patterns, the loss of this information affects the performance of subsequent stages, such as feature extraction and analysis. In this paper, to solve this problem, we propose a classification system with a new pooling method called Discrete Wavelet Transform Pooling (DWTP). This method is based on the image decomposition into sub-bands, in which the first level sub-band is considered as its output. The objective is to obtain approximation and detail information. As a result, this information can be concatenated in different combinations. In addition, wavelet pooling uses wavelets to reduce the size of the feature map. Combining these methods provides acceptable classification performance for three databases (CIFAR-10, DTD, and FMD). We argue that this helps eliminate overfitting and that the learning graphs reflect that the datasets show learning generalization. Therefore, our results indicate that our method based on wavelet analysis is feasible for texture and material classification. Moreover, in some cases, it outperforms traditional methods.

Hippocampus Segmentation Method Based on Subspace Patch-Sparsity Clustering in Noisy Brain MRI

Since the hippocampus is of small size, low contrast, and irregular shape, a novel hippocampus segmentation method based on subspace patch-sparsity clustering in brain MRI is proposed to improve the segmentation accuracy, which requires that the representation coefficients in different subspaces should be as sparse as possible, while the representation coefficients in the same subspace should be as average as possible. By restraining the coefficient matrix with the patch-sparse constraint, the coefficient matrix contains a patch-sparse structure, which is helpful to the hippocampus segmentation. The experimental results show that our proposed method is effective in the noisy brain MRI data, which can well deal with hippocampus segmentation problem.