دانلود کتاب متن کاوی زیست پزشکی

Text mining, a branch of data mining, is applied to automatically extract information from the unstructured text and to present the extracted information in a structured form. Text mining can be applied to any text ranging from general English novels to specific documents such as published biomedical articles and patients’ electronic health records (EHR). Depending on the type of input text, text mining has emerged with more specific names (e.g., biomedical text mining and clinical text mining) and applications in recent years. The published biomedical articles are the best source of knowledge on biological and biomedical discoveries. The number of published articles keeps growing exponentially because of the advances in bioinformatics and high-throughput experiments. PubMed contains ~32 million articles. A recent study says that thousands of articles are added to PubMed every day. Extracting the information from this overwhelming database is possible only through automated approaches. Biomedical text mining or biomedical literature mining is a specialized branch of text mining applied to automatically extract the information from published biomedical articles. In 1999, biomedical literature mining was first applied to PubMed articles for extracting protein-protein interaction information and gene expression profiling. Biomedical text mining includes several subtasks such as named entity recognition, information retrieval, information extraction, hypothesis generation, knowledge discovery, and literature-based discovery. In recent years, several hybrid approaches have emerged to improve the performance of biomedical text mining approaches. These hybrid approaches combine the biomedical text mining with other domains such as machine learning and genomics. In this book, step-by-step instructions on selected biomedical text mining protocols on various emerging topics such as disease comorbidity, drug interactions, and adverse drug events are provided. The protocols on combining biomedical text mining with machine learning or genomics for predicting novel biomedical discoveries are presented. Interesting topics such as literature-based discovery for uncovering unknown public knowledge by combining two pieces of information from two different sets of PubMed articles and recent trends in biomedical text mining such as BioBERT are included. The book also covers the role of data science, especially biomedical text mining in understanding the outbreaks such as COVID-19 and the underlying genetic link between complex diseases such as multiple sclerosis and cognitive disorder. A brief introduction to biomedical text mining and its three major components, information retrieval, named entity recognition, and information extraction, are presented in Chapter 1. A text mining protocol on retrieving significant drug-gene interaction information from PubMed abstracts is discussed in Chapter 2. In Chapter 3, a hybrid approach based on microarray expression data analysis and text mining for finding novel gene targets for any disease of interest is presented. Chapter 4 presents an interesting protocol on finding gene associations using text mining and annotating the associations using gene ontology. A unique biomedical text mining protocol for repurposing laboratory tests is presented in Chapter 5. While several computational approaches are available for repurposing drugs for diseases, repurposing laboratory tests through text mining is a novel approach that may gain much interest in near future.