| High intensity focused ultrasound(HIFU)therapy is a new non-invasive local therapy technique which has been used in the treatment of cancer tumors.Its main principle is to focus the ultrasonic wave through the skin of the human body to be treated in the body,and the energy carried by the ultrasonic wave will cause the treatment area to produce high temperature instantly,causing the degeneration of the tumor tissue.Ultrasonic irradiation will make the cells in the treatment area inactivated,will not affect the surrounding tissues,so as to achieve the purpose of non-invasive treatment.Due to the non-visibility and irreversibility of HIFU treatment,improper control of ultrasound treatment can seriously affect the treatment outcome.Therefore,in order to ensure the safety and efficacy of HIFU treatment,real-time monitoring of the treatment area and identification of biological tissue degeneration are of great significance.In this paper,by building a high-intensity focused ultrasound experimental system and taking in vitro pork tissue as the experimental sample,the ultrasonic scattering echo signal collected during HIFU irradiation of in vitro tissue was processed and analyzed,so as to complete the related research on the efficacy of biological tissue treated with HIFU.The main work is as follows:1.During HIFU treatment,when ultrasonic scattering echo signal crosses the boundary between normal tissue and denatured tissue,the signal energy will change,and the time-frequency spectrum can well reflect the time-frequency distribution of the signal energy.In this paper,an modified S-transform is proposed by improving the Gaussian window function in S-transform,and it is proved that the modified S-transform has better time-frequency aggregation than the standard S transform.By using the time-frequency entropy algorithm to calculate the time-frequency distribution obtained by the modified S-transform,the change of the time-frequency distribution of the signal is analyzed,and the HIFU damage region is located.2.When HIFU irradiates biological tissue,many factors jointly affect the therapeutic effect,which makes the signal information obtained from biological tissue fluctuate.In this paper,the concept of multi-scale fuzzy entropy(MFE)is introduced,and the modified multi-scale fuzzy entropy(MMFE)is obtained according to the information change of ultrasonic echo signal.In this paper,the comparison effect between MMFE and MFE is analyzed through the simulation experiments of Gaussian white noise and 1/f noise.It is proved that MMFE has higher stability than MFE,and is less limited by the signal data length.Then,through the analysis of the actual collected ultrasonic echo signals,it was found that the MMFE value was also more stable than MFE value,which verified the effectiveness of the improvement.At the same time,the variation of the MMFE value of normal and denatured tissues with different embedding dimensions and scale factors was also studied.3.When the biological tissue is irradiated by HIFU,the tissue changes from normal state to denatured state,which is a gradual process.In the process of gradual change,it is a fuzzy state when the biological tissue is damaged but not completely deactivated.Therefore,the concept of fuzzy clustering is introduced in this paper to identify the state of tissue degeneration and specifically solve the ambiguity in the process of gradual change.In this paper,FCM clustering and GK fuzzy clustering method were used to conduct clustering analysis with MMFE as the signal characteristics.The experimental results proved that fuzzy clustering could effectively complete the division of normal and denatured tissue states,and GK fuzzy clustering had better recognition effect. |
PDF Full Text Request