Automatic pure tone audiometry method based on Bayesian active learning and Gaussian process regression

• Main Author: LIU WENPENG
• Format: Patent
• Language: Chinese; English
• Published: 03.11.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DIAGNOSIS; ELECTRIC DIGITAL DATA PROCESSING; HUMAN NECESSITIES; HYGIENE; IDENTIFICATION; MEDICAL OR VETERINARY SCIENCE; PHYSICS; SURGERY

The invention provides an automatic pure tone audiometry method based on Bayesian active learning and Gaussian process regression, and the method comprises the steps: carrying out the pretest, building a regression model through the Gaussian process, and building an audiogram through the regression data; screening a maximum uncertainty iteration point by using a Bayesian active learning principle to obtain audiometric signal frequency and hearing grade; performing audiometry on a subject to obtain data and adding the data into a data set; and performing Gaussian process regression and active learning iteration until a target audiogram is obtained. According to the method, fixed steps of a traditional audiometry method are replaced with the prediction capacity of machine learning, the test time can be shortened, the test inaccuracy possibly caused by long-time test can be reduced, a smooth and accurate audiogram can be provided, and the audiogram precision is far higher than that of the traditional audiometry