Using SNOMED-CT to Help the Transition from Microbiological Data to ICD-10 Sepsis Codes.

Assigning ICD-10 code of sepsis in regard of a pathogenic bacterium found in an haemoculture requires knowledge of microbiology because of the difference of granularity. The aim of this paper is to automate this coding thanks to the use of SNOMED-CT. A dichotomous classification of bacteria causing sepsis has been generated in respect of ICD-10. Our algorithm follows this and explores SNOMED-CT to assign the right ICD-10 code of the sepsis. Applied to a list of 164 bacteria, the system has an error rate of 1.22 %.

Development of an Automatic Coding System for Digestive Endoscopies.

Digestive endoscopies, along with all medical procedures in France are coded with the CCAM. This task is done by the physicians, is time-consuming and requires a good knowledge of the terminology besides a medical knowledge. This method offers an automatic coding of endoscopic procedures from free-text reports. Thanks to a supervised learning method, the reports are coded with an average precision and recall of 0.92 on a 1639 texts corpus.