Abstract. We propose a novel, multi-task, fully convolutional network (FCN) architecture for automatic segmentation of brain tumor. This network extracts multi-level contextual information by concatenating hierarchical feature representations extracted from multimodal MR images along with their symmetric-difference images. It achieves improved segmentation performance by incorporating boundary information directly into the loss function. The proposed method was evaluated on the BRATS13 and BRATS15 datasets and compared with competing methods on the BRATS13 testing set. Segmented tumor boundaries obtained were better than those obtained by single-task FCN and by FCN with CRF. The method is among the most accurate available and has relatively low computational cost at test time.

We sought to investigate, whether texture analysis of diffusional kurtosis imaging (DKI) enhanced by support vector machine (SVM) analysis may provide biomarkers for gliomas staging and detection of the IDH mutation. First-order statistics and texture feature extraction were performed in 37 patients on both conventional (FLAIR) and mean diffusional kurtosis (MDK) images and recursive feature elimination (RFE) methodology based on SVM was employed to select the most discriminative diagnostic biomarkers. The first-order statistics demonstrated significantly lower MDK values in the IDH-mutant tumors. This resulted in 81.1% accuracy (sensitivity = 0.96, specificity = 0.45, AUC 0.59) for IDH mutation diagnosis. There were non-significant differences in average MDK and skewness among the different tumour grades. When texture analysis and SVM were utilized, the grading accuracy achieved by DKI biomarkers was 78.1% (sensitivity 0.77, specificity 0.79, AUC 0.79); the prediction accuracy for IDH mutation reached 83.8% (sensitivity 0.96, specificity 0.55, AUC 0.87). For the IDH mutation task, DKI outperformed significantly the FLAIR imaging. When using selected biomarkers after RFE, the prediction accuracy achieved 83.8% (sensitivity 0.92, specificity 0.64, AUC 0.88). These findings demonstrate the superiority of DKI enhanced by texture analysis and SVM, compared to conventional imaging, for gliomas staging and prediction of IDH mutational status.

No abstract

Improving Ki67 assessment concordance by the use of an artificial intelligence-empowered microscope: a multi-institutional ring study Aims: The nuclear proliferation biomarker Ki67 plays potential prognostic and predictive roles in breast cancer treatment. However, the lack of interpathologist consistency in Ki67 assessment limits the clinical use of Ki67. The aim of this article was to report a solution utilising an artificial intelligence (AI)-empowered microscope to improve Ki67 scoring concordance. Methods and results: We developed an AI-empowered microscope in which the conventional microscope was equipped with AI algorithms, and AI results were provided to pathologists in real time through augmented reality. We recruited 30 pathologists with various experience levels from five institutes to assess the Ki67 labelling index on 100 Ki67-stained slides from invasive breast cancer patients. In the first round, pathologists conducted visual assessment on a conventional microscope; in the second round, they were assisted with reference cards; and in the third round, they were assisted with an AI-empowered microscope. Experienced pathologists had better reproducibility and accuracy [intraclass correlation coefficient (ICC) = 0.864, mean error = 8.25%] than inexperienced pathologists (ICC = 0.807, mean error = 11.0%) in visual assessment. Moreover, with reference cards, inexperienced pathologists (ICC = 0.836, mean error = 10.7%) and experienced pathologists (ICC = 0.875, mean error = 7.56%) improved their reproducibility and accuracy. Finally, both experienced pathologists (ICC = 0.937, mean error = 4.36%) and inexperienced pathologists (ICC = 0.923, mean error = 4.71%) improved the reproducibility and accuracy significantly with the AI-empowered microscope. Conclusion:The AI-empowered microscope allows seamless integration of the AI solution into the clinical workflow, and helps pathologists to obtain higher consistency and accuracy for Ki67 assessment.

In this paper, we present a novel and efficient method for brain tumor (and sub regions) segmentation in multimodal MR images based on a fully convolutional network (FCN) that enables end-to-end training and fast inference. Our structure consists of a downsampling path and three upsampling paths, which extract multi-level contextual information by concatenating hierarchical feature representation from each upsampling path. Meanwhile, we introduce a symmetry-driven FCN by the proposal of using symmetry difference images. The model was evaluated on Brain Tumor Image Segmentation Benchmark (BRATS) 2013 challenge dataset and achieved the state-of-the-art results while the computational cost is less than competitors.

No abstract

No abstract

No abstract