KLASIFIKASI DATA REKAM MEDIS BERDASARKAN STATISTICAL CLASSIFICATION OF DISEASES (ICD-10) MENGGUNAKAN ALGORITMA K-NEAREST NEIGHBOR (K-NN)

Abstract

Hospital is a health service institution that provides complete individual health services that provide inpatient, outpatient and emergency services. In general, the benefits of medical records in hospitals are as a support for orderly administration in an effort to improve health services in hospitals. However, after the administrative goals are met medical records of patients going into the archive to be stored in accordance with Permenkes 269 / Menkes / Per / III / 2008 in Chapter IV, Article 8 provides that the medical records of inpatients in the hospital must be kept at the lack of for a period of 5 (five) years from the last date the patient was treated or discharged. In the end, the medical record data that grows every day becomes a pile of data. Therefore, the author wishes to create a system that can process hospital medical record data in the form of disease classification based on the International Statistical Classification of Disease and Related Health Problems (ICD-10). The study used pre-existing medical record data with a total of 1000 data. Where the experimental data is divided into 3 variations for the system experiment, namely 21 classes of ICD-10, 11 classes of ICD-10, and 3 classes of ICD-10. The results of this study get the highest accuracy of 73.13% on experimental data 80%: 20%. with 3 ICD-10 classes.