Comunicació

This thesis addresses the extraction of medical knowledge from clinical text
using deep learning techniques. In particular, the proposed methods focus
on cancer clinical trial protocols and chest x-rays reports. As a result of
this work, two large-scale labeled medical datasets were generated to help
advance both the research area of patient's eligibility in clinical trials,
and also the field of medical image recognition by making it possible to
apply supervised deep learning techniques on x-rays to obtain radiological
findings, anatomical localizations and pathologies.