Using Statistical Models for Ultrasonic Backscattering to Clinically Assess Osteoporosis

碩士 === 中原大學 === 醫學工程研究所 === 91 === Osteoporosis is commonly detected using DEXA, QCT, and QUS. However, DEXA or QCT have disadvantages of radiation risk and high cost. On the other hand, QUS is an irradiational modality in which its results of measurement are affected by fat and other soft tissue. F...

Full description

Bibliographic Details
Main Authors: Chih-Hong Huang, 黃智宏
Other Authors: Shyh-Hau Wang
Format: Others
Language:zh-TW
Published: 2003
Online Access:http://ndltd.ncl.edu.tw/handle/8jvnar
Description
Summary:碩士 === 中原大學 === 醫學工程研究所 === 91 === Osteoporosis is commonly detected using DEXA, QCT, and QUS. However, DEXA or QCT have disadvantages of radiation risk and high cost. On the other hand, QUS is an irradiational modality in which its results of measurement are affected by fat and other soft tissue. For these reasons, this study attempted to assess the degree of osteoporosis by detecting microstructures of the bone using Nakagami statistical parameters for ultrasonic backscattering signals. The influence of fat and other soft tissues could be eliminated using this method. Moreover, it has advantage of low cost and non-radiation. This investigation included software simulations, experiments on phantoms, and the clinical experiments. Results obtained from software simulations and phantoms experiments showed that Nakagami parameter approaches to 1 associated with the increase of scatterers concentration. The parameter is unstable when the length of gated scattering is less than 10mm. In clinical experiments, A transducer of 1 MHz was used to measure human tibia and calcaneus. To better quantitatively assess the degree of osteoporosis, four normalized statistical parameters were defined. Result demonstrated that the correlation coefficients between these four parameters and BMD of hip are ranged from 0.66 to 0.70. Although these relations are less than that obtained using DEXA, these are larger than relation between DEXA and BUA. These results showed that the proposed four statistical parameters have a great potential in assessing degree of osteoporosis. To further make a convenience in clinical measurements, the study also developed a portable ultrasound measurement system that has functions of excitation and reception for ultrasonic signals, filter, amplification, digitization, and data transfer. The maximum amplification of signals is 35.6dB. Digital signals have 12bits resolution and 65MHz sampling frequency. The phantom test by this system showed a correlation coefficient between results obtained from this system and these from previous system is 0.99. This just developed system will be further applied to clinical investigation.