High-grade gliomas are the most aggressive malignant brain tumors. Accurate pre-operative prognosis for this cohort can lead to better treatment planning. Conventional survival prediction based on clinical information is subjective and could be inaccurate. Recent radiomics studies have shown better prognosis by using carefully-engineered image features from magnetic resonance images (MRI). However, feature engineering is usually time consuming, laborious and subjective. Most importantly, the engineered features cannot effectively encode other predictive but implicit information provided by multi-modal neuroimages. We propose a two-stage learning-based method to predict the overall survival (OS) time of high-grade gliomas patient. At the first stage, we adopt deep learning, a recently dominant technique of artificial intelligence, to automatically extract implicit and high-level features from multi-modal, multi-channel preoperative MRI such that the features are competent of predicting survival time. Specifically, we utilize not only contrast-enhanced T1 MRI, but also diffusion tensor imaging (DTI) and resting-state functional MRI (rs-fMRI), for computing multiple metric maps (including various diffusivity metric maps derived from DTI, and also the frequency-specific brain fluctuation amplitude maps and local functional connectivity anisotropy-related metric maps derived from rs-fMRI) from 68 high-grade glioma patients with different survival time. We propose a multi-channel architecture of 3D convolutional neural networks (CNNs) for deep learning upon those metric maps, from which high-level predictive features are extracted for each individual patch of these maps. At the second stage, those deeply learned features along with the pivotal limited demographic and tumor-related features (such as age, tumor size and histological type) are fed into a support vector machine (SVM) to generate the final prediction result (i.e., long or short overall survival time). The experimental results demonstrate that this multi-model, multi-channel deep survival prediction framework achieves an accuracy of 90.66%, outperforming all the competing methods. This study indicates highly demanded effectiveness on prognosis of deep learning technique in neuro-oncological applications for better individualized treatment planning towards precision medicine.

Neuroimaging suggests that human consciousness relies on frequent access to default mode and dorsal attention brain networks.

Background: Pathological diagnosis of glioma subtypes is essential for treatment planning and prognosis. Standard histological diagnosis of glioma is based on postoperative hematoxylin and eosin stained slides by neuropathologists. With advancing artificial intelligence (AI), the aim of this study was to determine whether deep learning can be applied to glioma classification. Methods: A neuropathological diagnostic platform is designed comprising a slide scanner and deep convolutional neural networks (CNNs) to classify 5 major histological subtypes of glioma to assist pathologists. The CNNs were trained and verified on over 79 990 histological patch images from 267 patients. A logical algorithm is used when molecular profiles are available. Results: A new model of the squeeze-and-excitation block DenseNet with weighted cross-entropy (named SD-Net_WCE) is developed for the glioma classification task, which learns the recognizable features of glioma histology CNN-based independent diagnostic testing on data from 56 patients with 17 262 histological patch images demonstrated patch level accuracy of 86.5% and patient level accuracy of 87.5%. Histopathological classifications could be further amplified to integrated neuropathological diagnosis by 2 molecular markers (isocitrate dehydrogenase and 1p/19q). Conclusion: The model is capable of solving multiple classification tasks and can satisfactorily classify glioma subtypes. The system provides a novel aid for the integrated neuropathological diagnostic workflow of glioma.

Direct electrical stimulation, the transient ‘lesional’ method probing brain function, has been utilized in identifying the language cortex and preserving language function during epilepsy and neuro-oncological surgeries for about a century. However, comparison of functional maps of the language cortex across languages/continents based on cortical stimulation remains unclear. We conducted a retrospective multicentre study including four cohorts of direct electrical stimulation mapping from four centres across three continents, where three indigenous languages (English, French and Mandarin) are spoken. All subjects performed the two most common language tasks: number counting and picture naming during stimulation. All language sites were recorded and normalized to the same brain template. Next, Spearman’s correlation analysis was performed to explore the consistency of the distributions of the language cortex across centres, a kernel density estimation to localize the peak coordinates, and a hierarchical cluster analysis was performed to detect the crucial epicenters. A total of 598 subjects with 917 speech arrest sites (complete interruption of ongoing counting) and 423 anomia sites (inability to name or misnaming) were included. Different centres presented highly consistent distribution patterns for speech arrest (Spearman’s coefficient r ranged from 0.60 to 0.85, all pair-wise correlations P < 0.05), and similar patterns for anomia (Spearman’s coefficient r ranged from 0.37 to 0.80). The combinational speech arrest map was divided into four clusters: cluster 1 mainly located in the ventral precentral gyrus and pars opercularis, which contained the peak of speech arrest in the ventral precentral gyrus; cluster 2 in the ventral and dorsal precentral gyrus; cluster 3 in the supplementary motor area; cluster 4 in the posterior superior temporal gyrus and supramarginal gyrus. The anomia map revealed two clusters: one was in the posterior part of the superior and middle temporal gyri, which peaked at the posterior superior temporal gyrus; and the other within the inferior frontal gyrus, peaked at the pars triangularis. This study constitutes the largest series to date of language maps generated from direct electrical stimulation mapping. The consistency of data provides evidence for common language networks across languages, in the context of both speech and naming circuit. Our results not only clinically offer an atlas for language mapping and protection, but also scientifically provide better insight into the functional organization of language networks.

Glioblastoma (GBM) is the most common and deadly malignant brain tumor. For personalized treatment, an accurate pre-operative prognosis for GBM patients is highly desired. Recently, many machine learning-based methods have been adopted to predict overall survival (OS) time based on the pre-operative mono- or multi-modal imaging phenotype. The genotypic information of GBM has been proven to be strongly indicative of the prognosis; however, this has not been considered in the existing imaging-based OS prediction methods. The main reason is that the tumor genotype is unavailable pre-operatively unless deriving from craniotomy. In this paper, we propose a new deep learning-based OS prediction method for GBM patients, which can derive tumor genotype-related features from pre-operative multimodal magnetic resonance imaging (MRI) brain data and feed them to OS prediction. Specifically, we propose a multi-task convolutional neural network (CNN) to accomplish both tumor genotype and OS prediction tasks jointly. As the network can benefit from learning tumor genotype-related features for genotype prediction, the accuracy of predicting OS time can be prominently improved. In the experiments, multimodal MRI brain dataset of 120 GBM patients, with as many as four different genotypic/molecular biomarkers, are used to evaluate our method. Our method achieves the highest OS prediction accuracy compared to other state-of-the-art methods.

Background The marginal delineation of gliomas cannot be defined by conventional imaging due to their infiltrative growth pattern. Here we investigate the relationship between changes in glioma metabolism by proton magnetic resonance spectroscopic imaging (¹H-MRSI) and histopathological findings in order to determine an optimal threshold value of choline/N-acetyl-aspartate (Cho/NAA) that can be used to define the extent of glioma spread. Method Eighteen patients with different grades of glioma were examined using ¹H-MRSI. Needle biopsies were performed under the guidance of neuronavigation prior to craniotomy. Intraoperative magnetic resonance imaging (MRI) was performed to evaluate the accuracy of sampling. Haematoxylin and eosin, and immunohistochemical staining with IDH1, MIB-1, p53, CD34 and glial fibrillary acidic protein (GFAP) antibodies were performed on all samples. Logistic regression analysis was used to determine the relationship between Cho/NAA and MIB-1, p53, CD34, and the degree of tumour infiltration. The clinical threshold ratio distinguishing tumour tissue in high-grade (grades III and IV) glioma (HGG) and low-grade (grade II) glioma (LGG) was calculated. Results In HGG, higher Cho/NAA ratios were associated with a greater probability of higher MIB-1 counts, stronger CD34 expression, and tumour infiltration. Ratio threshold values of 0.5, 1.0, 1.5 and 2.0 appeared to predict the specimens containing the tumour with respective probabilities of 0.38, 0.60, 0.79, 0.90 in HGG and 0.16, 0.39, 0.67, 0.87 in LGG. Conclusions HGG and LGG exhibit different spectroscopic patterns. Using ¹H-MRSI to guide the extent of resection has the potential to improve the clinical outcome of glioma surgery.