Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Smart ABR System for Hearing Screening Smart ABR System can replace the traditional approach of hearing screening. This process shows significance improvement on the specificity percentage compared to traditional TEOAE method. Besides that, our hybrid kalman filtering method has greatly reduce the sweeps or stimulus time which make the application of ABR more feasible for baby hearing screening. The conventional 1000 sweeps stimulus ABR system consumes a lot of time for data collection. It is very irritating to baby thus baby might move around. Normally doctors will give sedation to the baby so that the data collection (1000 sweeps) is smooth without any baby movement. The reduction of 1000 sweeps to 400 sweeps manage to reduce the data collection time from 3-4 minutes to 1 minutes which makes the baby feel comfortable, hence no sedation needed. With this breakthrough, the proposed product will become a useful tool for baby hearing screening. Conventional 1000 sweeps averaging ABR method has been long introduced to government hospital. However, the time consuming method requires sedation and most parents do not want their baby to go through the process. This made the system not widely used for baby hearing screening and only used for those high risk group people. With our innovative way of enhancing the recording time and stimulus pattern, this system can now be widely used for baby hearing screening and do not need any sedation for it. This product uses 3 active electrodes to capture EEG signal from brain. The chirp stimulus is given to the ear needed to be tested. The signal captured will be analysed using hybrid Wavelet-Kalman Filtering method. The hybrid Wavelet-Kalman Filtering has never been implemented for Auditory Brainstem Response (ABR) application. It has better performance compared to conventional techniques which use averaging method. Besides that, we propose a better stimulus type by replacing conventional clicks with chirps. Chirps stimulus enhances the wave V amplitude which is the indicator for hearing ability. We propose a new idea of hybrid wavelet-kalman filter technique which is novel technique. Kalman filter is an optimal recursive data processing algorithm, where it considers the criteria of state, system dynamical, controls, system error function, observation measurement, etc. The technique optimizes digital signal processing and dramatically reduces artifacts coming from muscular and ocular electrical activity. Its implementation in the ABR signal processing would help better noise filtering with the minimal number of sweeps. Advantages High sensitivity and specificity (by using new chirp stimulus to enhance the amplitude of wave V) Durable and less time recording (by reducing of 1000 sweeps to 400 sweeps) The system is now feasible for all babies (fast detection and do not need sedation) Application: Newborn hearing screening Adult hearing screening