RESUMO
Gliomas are primary brain tumors that originate from glial cells. Classification and grading of these tumors is critical to prognosis and treatment planning. The current criteria for glioma classification in central nervous system (CNS) was introduced by World Health Organization (WHO) in 2016. This criteria for glioma classification requires the integration of histology with genomics. In 2017, the Consortium to Inform Molecular and Practical Approaches to CNS Tumor Taxonomy (cIMPACT-NOW) was established to provide up-to-date recommendations for CNS tumor classification, which in turn the WHO is expected to adopt in its upcoming edition. In this work, we propose a novel glioma analytical method that, for the first time in the literature, integrates a cellularity feature derived from the digital analysis of brain histopathology images integrated with molecular features following the latest WHO criteria. We first propose a novel over-segmentation strategy for region-of-interest (ROI) selection in large histopathology whole slide images (WSIs). A Deep Neural Network (DNN)-based classification method then fuses molecular features with cellularity features to improve tumor classification performance. We evaluate the proposed method with 549 patient cases from The Cancer Genome Atlas (TCGA) dataset for evaluation. The cross validated classification accuracies are 93.81% for lower-grade glioma (LGG) and high-grade glioma (HGG) using a regular DNN, and 73.95% for LGG II and LGG III using a residual neural network (ResNet) DNN, respectively. Our experiments suggest that the type of deep learning has a significant impact on tumor subtype discrimination between LGG II vs. LGG III. These results outperform state-of-the-art methods in classifying LGG II vs. LGG III and offer competitive performance in distinguishing LGG vs. HGG in the literature. In addition, we also investigate molecular subtype classification using pathology images and cellularity information. Finally, for the first time in literature this work shows promise for cellularity quantification to predict brain tumor grading for LGGs with IDH mutations.
RESUMO
Objective To classify and grade the environments with Oncomelania hupensis snails inside embankment in marsh-land and lake areas,so as to improve the work efficiency and realize the systematic management. Methods The schistosomiasis endemic area Liaodi and Xiongdi villages in Jiangling County,Hubei Province were selected as the experiment villages,and the environments with snails in the two villages were divided into sections with a length of 500m,then the snail situation were sur-veyed by the systematic sampling method with 10m and 50m a frame respectively. The environments were classified according to the discovery of infected snails and wild fecal contamination,and the numbers of sites with snails and their areas in different class-es were recorded. Meanwhile,the data of sites with infected snails in Jingzhou City during the recent 5 years were collected and graded according to the average density of snails,so as to discuss the correlativity between the grades of environments with snails and the numbers of sites with infected snails and the occurrence rate of frames with snails. Results There were 1 967 sites with in-fected snails in Jingzhou City during the recent 5 years,and there was a positive correlation between the grades of environments with snails and the occurrence rate of frames with snails(rs=0.77,P0.05),but the time-consuming of the former was 3 times of the latter. Conclusions The classification and grading of the environments with snails inside embankment in marshland and lake areas can master the key point of the snail survey. The subsection and setting frames at a suitable distance can save time and manpower,improve work efficiency,as well as understand the distribution of snail status of the environment with snails,which can realize the sort management of the snail envi-ronments inside embankment.
