Searching the Resource Information Network
Are you sure you want to leave this community? Leaving the community will revoke any permissions you have been granted in this community.
Hepatocellular carcinoma (HCC) is the most common subtype of liver cancer, and assessing its histopathological grade requires visual inspection by an experienced pathologist. In this study, the histopathological H&E images from the Genomic Data Commons Databases were used to train a neural network (inception V3) for automatic classification. According to the evaluation of our model by the Matthews correlation coefficient, the performance level was close to the ability of a 5-year experience pathologist, with 96.0% accuracy for benign and malignant classification, and 89.6% accuracy for well, moderate, and poor tumor differentiation. Furthermore, the model was trained to predict the ten most common and prognostic mutated genes in HCC. We found that four of them, including CTNNB1, FMN2, TP53, and ZFX4, could be predicted from histopathology images, with external AUCs from 0.71 to 0.89. The findings demonstrated that convolutional neural networks could be used to assist pathologists in the classification and detection of gene mutation in liver cancer.
Pubmed ID: 32550270
Publication data is provided by the National Library of Medicine ® and PubMed ®. Data is retrieved from PubMed ® on a weekly schedule. For terms and conditions see the National Library of Medicine Terms and Conditions.
Software environment and programming language for statistical computing and graphics. R is integrated suite of software facilities for data manipulation, calculation and graphical display. Can be extended via packages. Some packages are supplied with the R distribution and more are available through CRAN family.It compiles and runs on wide variety of UNIX platforms, Windows and MacOS.
View all literature mentionsProject exploring the spectrum of genomic changes involved in more than 20 types of human cancer that provides a platform for researchers to search, download, and analyze data sets generated. As a pilot project it confirmed that an atlas of changes could be created for specific cancer types. It also showed that a national network of research and technology teams working on distinct but related projects could pool the results of their efforts, create an economy of scale and develop an infrastructure for making the data publicly accessible. Its success committed resources to collect and characterize more than 20 additional tumor types. Components of the TCGA Research Network: * Biospecimen Core Resource (BCR); Tissue samples are carefully cataloged, processed, checked for quality and stored, complete with important medical information about the patient. * Genome Characterization Centers (GCCs); Several technologies will be used to analyze genomic changes involved in cancer. The genomic changes that are identified will be further studied by the Genome Sequencing Centers. * Genome Sequencing Centers (GSCs); High-throughput Genome Sequencing Centers will identify the changes in DNA sequences that are associated with specific types of cancer. * Proteome Characterization Centers (PCCs); The centers, a component of NCI's Clinical Proteomic Tumor Analysis Consortium, will ascertain and analyze the total proteomic content of a subset of TCGA samples. * Data Coordinating Center (DCC); The information that is generated by TCGA will be centrally managed at the DCC and entered into the TCGA Data Portal and Cancer Genomics Hub as it becomes available. Centralization of data facilitates data transfer between the network and the research community, and makes data analysis more efficient. The DCC manages the TCGA Data Portal. * Cancer Genomics Hub (CGHub); Lower level sequence data will be deposited into a secure repository. This database stores cancer genome sequences and alignments. * Genome Data Analysis Centers (GDACs) - Immense amounts of data from array and second-generation sequencing technologies must be integrated across thousands of samples. These centers will provide novel informatics tools to the entire research community to facilitate broader use of TCGA data. TCGA is actively developing a network of collaborators who are able to provide samples that are collected retrospectively (tissues that had already been collected and stored) or prospectively (tissues that will be collected in the future).
View all literature mentionsA unified data repository of the National Cancer Institute (NCI)'s Genomic Data Commons (GDC) that enables data sharing across cancer genomic studies in support of precision medicine. The GDC supports several cancer genome programs at the NCI Center for Cancer Genomics (CCG), including The Cancer Genome Atlas (TCGA), Therapeutically Applicable Research to Generate Effective Treatments (TARGET), and the Cancer Genome Characterization Initiative (CGCI). The GDC Data Portal provides a platform for efficiently querying and downloading high quality and complete data. The GDC also provides a GDC Data Transfer Tool and a GDC API for programmatic access.
View all literature mentionsSoftware as an open source machine learning framework for everyone. Library for high performance numerical computation. Allows deployment of computation across a variety of platforms (CPUs, GPUs, TPUs), and from desktops to clusters of servers to mobile and edge devices.
View all literature mentions