DocumentCode :
3472192
Title :
Challenges in predicting community periodontal index from hospital dental care records
Author :
Vieira, Dario ; Hollmen, Jaakko ; Linden, Jari ; Suni, Jorma
Author_Institution :
X-akseli Oy, Espoo, Finland
fYear :
2013
fDate :
20-22 June 2013
Firstpage :
107
Lastpage :
112
Abstract :
Many studies have been performed in predicting periodontal diseases based on genetic information, dental images or patients habits but few have yet used dental visits records. This paper proposes a methodology based on Random Forest to classify the periodontal disease condition of patients and a way to assess the most important features that lead to a successful classification. We investigate three problematic issues found in dental care records: noise, class imbalance and concept drift and propose solutions to overcome them by respectively detecting and removing noise, under-sampling and only considering recent data. Experiments performed on records from Finnish public hospitals of two cities had good classification results and feature importance was able to detect dentists with poor performance with respect to diagnosis and treatment application.
Keywords :
dentistry; medical diagnostic computing; medical information systems; patient diagnosis; patient treatment; pattern classification; random processes; Finnish public hospitals; class imbalance; community periodontal index prediction; concept drift; dental images; dental visits records; dentists; diagnosis; genetic information; hospital dental care records; noise removal; periodontal disease classification; random forest; treatment application; undersampling; Accuracy; Dentistry; Diseases; Kernel; Noise; Support vector machines; Training;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer-Based Medical Systems (CBMS), 2013 IEEE 26th International Symposium on
Conference_Location :
Porto
Type :
conf
DOI :
10.1109/CBMS.2013.6627773
Filename :
6627773
Link To Document :
بازگشت