The Design and Development of Narrow-band Noise Generator for Tinnitus

碩士 === 國立成功大學 === 醫學工程研究所碩博士班 === 94 ===  There are 17% of the total population suffering from Tinnitus in UK and USA; the prevalence of tinnitus patients has been increasing with the aging. There are approximately 30% of the elderly population suffering from Tinnitus. The noises from tinnitus have...

Full description

Bibliographic Details
Main Authors: Chien-Wei Chen, 陳建維
Other Authors: Kao-Chi Chung
Format: Others
Language:zh-TW
Published: 2005
Online Access:http://ndltd.ncl.edu.tw/handle/40737693835577896102
Description
Summary:碩士 === 國立成功大學 === 醫學工程研究所碩博士班 === 94 ===  There are 17% of the total population suffering from Tinnitus in UK and USA; the prevalence of tinnitus patients has been increasing with the aging. There are approximately 30% of the elderly population suffering from Tinnitus. The noises from tinnitus have influenced seriously on the daily life for tinnitus patients sometimes. Currently, most of tinnitus devices are to provide wide-band noises for therapeutic purpose, and also expensive. The disadvantages of using wide-band therapeutic noises are: (1) tinnitus patients are suffering from noises with varied frequency bands and intensities; (2) more than 70% of patients also suffering from hearing loss usually at high frequency regions. When tinnitus patients have wear the current wide-band noise device for a period of time, they are usually at high risk of hearing loss. This research, based on the Tinnitus Masking and TRT for patient’s specific frequency band and volume, is to design and develop a microprocessor-based platform to provide customized band-pass noise for the tinnitus. This DSP technique would like to improve the shortcoming of current wide-band noise generator and ultimately be beneficial to the tinnitus.  This system consists of a PC, Dlr169 emulation board with MSP430F169 chip, earphone, FLASH Emulation Tool for software programming and JTAG transmission line. A wide-band white noise, which is generated through random numbers in PC, is input to the Dlr169 for FIR filter processing. Filter parameters and processing programs, which are designed and programmed through FDATooL of MATLAB, are then down-loaded to the Dlir169. The prescribed narrow-band noise with volume are then output from the speaker.  The prototype of narrow-band noise platform has been completed. The system calibration has been investigated for different stop bandwidths and rank and volume intensities on optimal parameter design in order to provide high quality narrow-band noises. A comparative study is conducted and the results indicate that the microprocessor-based system provides better narrow-band noise quality than the PC-based platform system.  A clinical investigation on using the system is suggested in the future.