Michael Ben Ezra

• Præsentation
• Publikationer

Big datasets describing pathologies are an invaluable and largely untapped resource in the effort to characterize human aging and discover new mechanisms of aging and aging biomarkers. Our research takes advantage of pathology reports written by skilled pathologists describing the microanatomy of millions of tissue samples from pathoanatomical surveys dating back to the 1970s.

The key challenge in leveraging the rich information in clinical narrative text is the unstructured nature of natural language. Using information extraction methods features describing the tissue architecture can be extracted from clinical text into structured form making it amenable to computational analysis. We seek to construct a set of metadata for the millions of biological samples stored in Denmark’s biobanks. Through the use of advanced hypothesis generation and knowledge discovery methods, this novel dataset will be used to identify opportunities for epidemiological studies leading to better understanding of the human aging process.