The 9th revision of the International Classificationof Diseases (as modified in the United States, also known as ICD-9-CM) is one of the most widely used classification system forclinical problems and procedures. Electronic health recordsystems usually require physicians to encode diagnoses ormedical procedures through appropriate ICD-9-CM codes. SinceICD-9-CM is constituted by more than 13,000 diagnosis and3,500 procedure codes, clearly the coding procedure is a nontrivialtask. In this paper we present a coding support system forassisting physicians in the assignment of ICD-9-CM codes. Theproposed system is based on tools for natural languageprocessing and text-searches on knowledge bases. It analyzes ashort medical text and then returns the list of ICD-9-CM codesthat are most likely to be associated with the given text. A set ofnumerical experiments were carried out in order to assess thereliability and quality of the proposed support system. Theresults of the experiments show that the support system has anaccuracy of 92%, largely improving physicians' average performances.
A Coding Support System for the ICD-9-CM standard
Roberto Guarasci;Erika Pasceri;Maria Teresa Chiaravalloti;
2014
Abstract
The 9th revision of the International Classificationof Diseases (as modified in the United States, also known as ICD-9-CM) is one of the most widely used classification system forclinical problems and procedures. Electronic health recordsystems usually require physicians to encode diagnoses ormedical procedures through appropriate ICD-9-CM codes. SinceICD-9-CM is constituted by more than 13,000 diagnosis and3,500 procedure codes, clearly the coding procedure is a nontrivialtask. In this paper we present a coding support system forassisting physicians in the assignment of ICD-9-CM codes. Theproposed system is based on tools for natural languageprocessing and text-searches on knowledge bases. It analyzes ashort medical text and then returns the list of ICD-9-CM codesthat are most likely to be associated with the given text. A set ofnumerical experiments were carried out in order to assess thereliability and quality of the proposed support system. Theresults of the experiments show that the support system has anaccuracy of 92%, largely improving physicians' average performances.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.