TY - GEN
T1 - Data mining of audiology patient records
T2 - ACM 5th International Workshop on Data and Text Mining in Biomedical Informatics, DTMBIO'11, in Conjunction with the 20th ACM International Conference on Information and Knowledge Management, CIKM'11
AU - Anwar, Muhammad Naveed
AU - Oakes, Michael Philip
PY - 2011/12/15
Y1 - 2011/12/15
N2 - In this paper we describe our analysis of a database of over 180,000 patient records, collected from over 23,000 patients, by the hearing aid clinic at James Cook University Hospital in Middlesbrough, UK. These records consist of audiograms (graphs of the faintest sounds audible to the patient at six different pitches), categorical data (such as age, gender, diagnosis and hearing aid type) and brief free text notes made by the technicians. We mine this data to determine which factors contribute to the decision to fit a BTE (worn behind the ear) hearing aid as opposed to an ITE (worn in the ear) hearing aid. From PCA (principal component analysis) we determined four main audiogram types, and we relate these to the type of hearing aid chosen. We combine the effects of age, gender, diagnosis, masker, mould and individual audiogram frequencies into a single model by means of logistic regression. We also discovered some significant keywords in the free text fields by using the chi-squared (χ2) test, which can also be used in the model. The final model can act a decision support tool to help decide whether an individual patient should be offered a BTE or an ITE hearing aid.
AB - In this paper we describe our analysis of a database of over 180,000 patient records, collected from over 23,000 patients, by the hearing aid clinic at James Cook University Hospital in Middlesbrough, UK. These records consist of audiograms (graphs of the faintest sounds audible to the patient at six different pitches), categorical data (such as age, gender, diagnosis and hearing aid type) and brief free text notes made by the technicians. We mine this data to determine which factors contribute to the decision to fit a BTE (worn behind the ear) hearing aid as opposed to an ITE (worn in the ear) hearing aid. From PCA (principal component analysis) we determined four main audiogram types, and we relate these to the type of hearing aid chosen. We combine the effects of age, gender, diagnosis, masker, mould and individual audiogram frequencies into a single model by means of logistic regression. We also discovered some significant keywords in the free text fields by using the chi-squared (χ2) test, which can also be used in the model. The final model can act a decision support tool to help decide whether an individual patient should be offered a BTE or an ITE hearing aid.
KW - chi-squared
KW - hearing aids
KW - logistic regression
KW - principal component analysis
UR - http://www.scopus.com/inward/record.url?scp=83255193404&partnerID=8YFLogxK
U2 - 10.1145/2064696.2064701
DO - 10.1145/2064696.2064701
M3 - Conference contribution
AN - SCOPUS:83255193404
SN - 9781450309608
T3 - International Conference on Information and Knowledge Management, Proceedings
SP - 11
EP - 18
BT - Proceedings of the ACM Fifth International Workshop on Data and Text Mining in Biomedical Informatics (DTMBio 2011)
A2 - Association for Computing Machinery,
PB - ACM
CY - New York
Y2 - 24 October 2011 through 24 October 2011
ER -