(TS)2WM: Tumor Segmentation and Tract Statistics for Assessing White Matter Integrity with Applications to Glioblastoma Patients

A deep convolutional neural network for the automatic segmentation of glioblastoma brain tumor: Joint spatial pyramid module and attention mechanism network

This study proposes a deep convolutional neural network for the automatic segmentation of glioblastoma brain tumors, aiming sat replacing the manual segmentation method that is both time-consuming and labor-intensive. There are many challenges for automatic segmentation to finely segment sub-regions from multi-sequence magnetic resonance images because of the complexity and variability of glioblastomas, such as the loss of boundary information, misclassified regions, and subregion size. To overcome these challenges, this study introduces a spatial pyramid module and attention mechanism to the automatic segmentation algorithm, which focuses on multi-scale spatial details and context information. The proposed method has been tested in the public benchmarks BraTS 2018, BraTS 2019, BraTS 2020 and BraTS 2021 datasets. The Dice score on the enhanced tumor, whole tumor, and tumor core were respectively 79.90 %, 89.63 %, and 85.89 % on the BraTS 2018 dataset, respectively 77.14 %, 89.58 %, and 83.33 % on the BraTS 2019 dataset, and respectively 77.80 %, 90.04 %, and 83.18 % on the BraTS 2020 dataset, and respectively 83.48 %, 90.70 %, and 88.94 % on the BraTS 2021 dataset offering performance on par with that of state-of-the-art methods with only 1.90 M parameters. In addition, our approach significantly reduced the requirements for experimental equipment, and the average time taken to segment one case was only 1.48 s; these two benefits rendered the proposed network intensely competitive for clinical practice.

Applications of nanotechnologies for miRNA-based cancer therapeutics: current advances and future perspectives

MicroRNAs (miRNAs) are short (18-25 nt), non-coding, widely conserved RNA molecules responsible for regulating gene expression via sequence-specific post-transcriptional mechanisms. Since the human miRNA transcriptome regulates the expression of a number of tumor suppressors and oncogenes, its dysregulation is associated with the clinical onset of different types of cancer. Despite the fact that numerous therapeutic approaches have been designed in recent years to treat cancer, the complexity of the disease manifested by each patient has prevented the development of a highly effective disease management strategy. However, over the past decade, artificial miRNAs (i.e., anti-miRNAs and miRNA mimics) have shown promising results against various cancer types; nevertheless, their targeted delivery could be challenging. Notably, numerous reports have shown that nanotechnology-based delivery of miRNAs can greatly contribute to hindering cancer initiation and development processes, representing an innovative disease-modifying strategy against cancer. Hence, in this review, we evaluate recently developed nanotechnology-based miRNA drug delivery systems for cancer therapeutics and discuss the potential challenges and future directions, such as the promising use of plant-made nanoparticles, phytochemical-mediated modulation of miRNAs, and nanozymes.

Deep Learning Classifies Low- and High-Grade Glioma Patients with High Accuracy, Sensitivity, and Specificity Based on Their Brain White Matter Networks Derived from Diffusion Tensor Imaging

Classifying low-grade glioma (LGG) patients from high-grade glioma (HGG) is one of the most challenging tasks in planning treatment strategies for brain tumor patients. Previous studies derived several handcrafted features based on the tumor's texture and volume from magnetic resonance images (MRI) to classify LGG and HGG patients. The accuracy of classification was moderate. We aimed to classify LGG from HGG with high accuracy using the brain white matter (WM) network connectivity matrix constructed using diffusion tensor tractography. We obtained diffusion tensor images (DTI) of 44 LGG and 48 HGG patients using routine clinical imaging. Fiber tractography and brain parcellation were performed for each patient to obtain the fractional anisotropy, axial diffusivity, radial diffusivity, and mean diffusivity weighted connectivity matrices. We used a deep convolutional neural network (DNN) for classification and the gradient class activation map (GRAD-CAM) technique to identify the neural connectivity features focused on by the DNN. DNN could classify both LGG and HGG with 98% accuracy. The sensitivity and specificity values were above 0.98. GRAD-CAM analysis revealed a distinct WM network pattern between LGG and HGG patients in the frontal, temporal, and parietal lobes. Our results demonstrate that glioma affects the WM network in LGG and HGG patients differently.

An optimal brain tumor segmentation algorithm for clinical MRI dataset with low resolution and non-contiguous slices

Background

Segmenting brain tumor and its constituent regions from magnetic resonance images (MRI) is important for planning diagnosis and treatment. In clinical routine often an experienced radiologist delineates the tumor regions using multimodal MRI. But this manual segmentation is prone to poor reproducibility and is time consuming. Also, routine clinical scans are usually of low resolution. To overcome these limitations an automated and precise segmentation algorithm based on computer vision is needed.

Methods

We investigated the performance of three widely used segmentation methods namely region growing, fuzzy C means and deep neural networks (deepmedic). We evaluated these algorithms on the BRATS 2018 dataset by choosing randomly 48 patients data (high grade, n = 24 and low grade, n = 24) and on our routine clinical MRI brain tumor dataset (high grade, n = 15 and low grade, n = 28). We measured their performance using dice similarity coefficient, Hausdorff distance and volume measures.

Results

Region growing method performed very poorly when compared to fuzzy C means (fcm) and deepmedic network. Dice similarity coefficient scores for FCM and deepmedic algorithms were close to each other for BRATS and clinical dataset. The accuracy was below 70% for both these methods in general.

Conclusion

Even though the deepmedic network showed very high accuracy in BRATS challenge for brain tumor segmentation, it has to be custom trained for the low resolution routine clinical scans. It also requires large training data to be used as a stand-alone algorithm for clinical applications. Nevertheless deepmedic may be a better algorithm for brain tumor segmentation when compared to region growing or FCM.

Tumor-associated alterations in white matter connectivity have prognostic significance in MGMT-unmethylated glioblastoma

PURPOSE

We investigated the prognostic significance of tumor-associated white matter (TA-WM) tracts in glioblastoma (GBM) using magnetic resonance-diffusion tensor imaging (MR-DTI). We hypothesized that (1) TA-WM tracts harbor microscopic disease not targeted through surgery or radiotherapy (RT), and (2) the greater the extent of TA-WM involvement, the worse the survival outcomes.

METHODS

We studied a retrospective cohort of 76 GBM patients. TA-WM tracts were identified by MR-DTI fractional anisotropy (FA) maps. For each patient, 22 TA-WM tracts were analyzed and each tract was graded 1-3 based on FA. A TA-WM score (TA-WMS) was computed based on number of involved tracts and corresponding FA grade of involvement. Kaplan-Meier statistics were utilized to determine survival outcomes, log-rank test was used to compare survival between groups, and Cox regression was utilized to determine prognostic variables.

RESULTS

For the MGMT-unmethylated cohort, there was a decrease in OS for increasing TA-WMS (median OS 16.5 months for TA-WMS 0-4; 13.6 months for TA-WMS 5-8; 7.3 months for TA-WMS > 9; p = 0.0002). This trend was not observed in the MGMT-methylated cohort. For MGMT-unmethylated patients with TA-WMS > 6 and involvement of tracts passing through brainstem or contralateral hemisphere, median OS was 8.3 months versus median OS 14.1 months with TA-WMS > 6 but not involving aforementioned critical tracts (p = 0.003 log-rank test). For MGMT-unmethylated patients, TA-WMS was predictive of overall survival in multivariate analysis (HR = 1.14, 95% CI 1.03-1.27, p = 0.012) while age, gender, and largest tumor dimension were non-significant.

CONCLUSION

Increased TA-WMS and involvement of critical tracts are associated with decreased overall survival in MGMT-unmethylated GBM.

Magnetic resonance image-based brain tumour segmentation methods: A systematic review

Background

Image segmentation is an essential step in the analysis and subsequent characterisation of brain tumours through magnetic resonance imaging. In the literature, segmentation methods are empowered by open-access magnetic resonance imaging datasets, such as the brain tumour segmentation dataset. Moreover, with the increased use of artificial intelligence methods in medical imaging, access to larger data repositories has become vital in method development.

Purpose

To determine what automated brain tumour segmentation techniques can medical imaging specialists and clinicians use to identify tumour components, compared to manual segmentation.

Methods

We conducted a systematic review of 572 brain tumour segmentation studies during 2015-2020. We reviewed segmentation techniques using T1-weighted, T2-weighted, gadolinium-enhanced T1-weighted, fluid-attenuated inversion recovery, diffusion-weighted and perfusion-weighted magnetic resonance imaging sequences. Moreover, we assessed physics or mathematics-based methods, deep learning methods, and software-based or semi-automatic methods, as applied to magnetic resonance imaging techniques. Particularly, we synthesised each method as per the utilised magnetic resonance imaging sequences, study population, technical approach (such as deep learning) and performance score measures (such as Dice score).

Statistical tests

We compared median Dice score in segmenting the whole tumour, tumour core and enhanced tumour.

Results

We found that T1-weighted, gadolinium-enhanced T1-weighted, T2-weighted and fluid-attenuated inversion recovery magnetic resonance imaging are used the most in various segmentation algorithms. However, there is limited use of perfusion-weighted and diffusion-weighted magnetic resonance imaging. Moreover, we found that the U-Net deep learning technology is cited the most, and has high accuracy (Dice score 0.9) for magnetic resonance imaging-based brain tumour segmentation.

Conclusion

U-Net is a promising deep learning technology for magnetic resonance imaging-based brain tumour segmentation. The community should be encouraged to contribute open-access datasets so training, testing and validation of deep learning algorithms can be improved, particularly for diffusion- and perfusion-weighted magnetic resonance imaging, where there are limited datasets available.

BiTr-Unet: a CNN-Transformer Combined Network for MRI Brain Tumor Segmentation

Convolutional neural networks (CNNs) have achieved remarkable success in automatically segmenting organs or lesions on 3D medical images. Recently, vision transformer networks have exhibited exceptional performance in 2D image classification tasks. Compared with CNNs, transformer networks have an appealing advantage of extracting long-range features due to their self-attention algorithm. Therefore, we propose a CNN-Transformer combined model, called BiTr-Unet, with specific modifications for brain tumor segmentation on multi-modal MRI scans. Our BiTr-Unet achieves good performance on the BraTS2021 validation dataset with median Dice score 0.9335, 0.9304 and 0.8899, and median Hausdor_ distance 2.8284, 2.2361 and 1.4142 for the whole tumor, tumor core, and enhancing tumor, respectively. On the BraTS2021 testing dataset, the corresponding results are 0.9257, 0.9350 and 0.8874 for Dice score, and 3, 2.2361 and 1.4142 for Hausdorff distance. The code is publicly available at https://github.com/JustaTinyDot/BiTr-Unet.

A Two-Stage Cascade Model with Variational Autoencoders and Attention Gates for MRI Brain Tumor Segmentation

Automatic MRI brain tumor segmentation is of vital importance for the disease diagnosis, monitoring, and treatment planning. In this paper, we propose a two-stage encoder-decoder based model for brain tumor subregional segmentation. Variational autoencoder regularization is utilized in both stages to prevent the overfitting issue. The second-stage network adopts attention gates and is trained additionally using an expanded dataset formed by the first-stage outputs. On the BraTS 2020 validation dataset, the proposed method achieves the mean Dice score of 0.9041, 0.8350, and 0.7958, and Hausdorff distance (95%) of 4.953 , 6.299, 23.608 for the whole tumor, tumor core, and enhancing tumor, respectively. The corresponding results on the BraTS 2020 testing dataset are 0.8729, 0.8357, and 0.8205 for Dice score, and 11.4288, 19.9690, and 15.6711 for Hausdorff distance. The code is publicly available at https://github.com/shu-hai/two-stage-VAE-Attention-gate-BraTS2020.