The project was to build and release the first publicly available code evidence dataset called MDACE on a subset of the MIMIC-III clinical records. We believe that the release of MDACE will greatly improve the understanding and application of deep learning technologies for medical coding, and benefit machine learning models for classification.
Hi again! I've noticed that some spans are outside the text in the discharge summary. For instance, there is a discharge summary with 13 367 characters, but the evidence spans are between 24 740 and 24 787. Have you concatenated the addendums to the discharge summaries or something?
Hi again! I've noticed that some spans are outside the text in the discharge summary. For instance, there is a discharge summary with 13 367 characters, but the evidence spans are between 24 740 and 24 787. Have you concatenated the addendums to the discharge summaries or something?