Experiences of otoneurological expert system for vertigo.

We have developed an OtoNeurological Expert system (ONE) to aid the diagnostics of vertigo, to assist teaching and to implement the database for research. The database contains detailed information on the patient history, signs and test results necessary for the diagnostic work with vertiginous patients. The pattern recognition method was used in the reasoning process. Questions regarding symptoms, signs and test results are weighted and scored for each disease, and the most likely disease is recognized from the defined disease profiles. Uncertainties in reasoning, caused by missing information, were solved with a method resembling fuzzy logic. We have also applied adaptive computer applications, such as genetic algorithms and decision trees, in the reasoning process. In the validation the expert system ONE proved to be a sound decision maker, by solving 65% of the cases correctly, while the physicians' mean was 69%. To improve the expert system ONE further, a follow-up should be implemented for the patients, to ease the diagnostic work of some difficult diseases. The six diseases were detected with high accuracy also with adaptive learning methods and discriminant analysis. An expert system is a practical tool in otoneurology. We aim to construct a hybrid program for the reasoning, where the best reasoning method for each disease is used.

Database for a hearing conservation program.

We have developed a database and an analysis program (NoiseScan) for noise-induced hearing loss (NIHL). The exposure data are based on the evaluation of the noise immission level, which includes duration, frequency content, and the use of, and the attenuation performance of, hearing protectors. The input data can handle an unlimited number of exposure periods. If the noise exposure level is not known, the program lists noise levels of comparable work places, and thus provides an estimate of exposure. Confounding medical factors that may contribute to NIHL, such as elevated serum cholesterol level, hypertension, and extensive use of pain killers, are collected. Combined exposure to agents that clearly contribute to NIHL, such as hand-arm vibration, tobacco smoking, use of aminoglycosides and exposure to solvents are also assessed. An unlimited number of audiograms can be stored, and all the data can be completed and edited following collection. The program gives the predicted hearing loss according to the ISO 1999 model based on total exposure. At present, our NoiseScan program (under continuous development in an EU research program) is suitable for the data collection of various risk factors. It can be used to determine whether the hearing loss is occupational in origin and to estimate the efficiency of hearing conservation measures. NoiseScan also predicts the development of hearing loss in individuals in 5-year periods. The goal is to improve and validate the rules by which single and combined risk factors contribute to HIHL, thus leading to more precise prediction of individual hearing loss, and for the evaluation of success of the hearing conservation programs.