A shared task involving multi-label classification of clinical free text

摘要: This paper reports on a shared task involving the assignment of ICD-9-CM codes to radiology reports. Two features distinguished this from previous tasks in biomedical domain. One is that it resulted first freely distributable corpus fully anonymized clinical text. resource permanently available and will (we hope) facilitate future research. The other key feature required categorization with respect large commercially significant set labels. number participants was larger than any challenge task. We describe data production process evaluation measures, give preliminary analysis results. Many systems performed at levels approaching inter-coder agreement, suggesting human-like performance within reach currently technologies.