| Due to the accelerated pace of life in modern society and increasing work pressure,the incidence of cerebrovascular diseases has remained high,seriously threatening human life and health.Cerebrovascular disease has a high mortality rate and disability rate.It is extremely important to make early warning and diagnosis of brain vascular disease better.Transcranial Doppler ultrasound diagnostic equipment has the characteristics of non-invasive,non-radiative,low cost and good real-time performance compared with conventional detection methods such as CT and MRI.It is very suitable for the initial diagnosis and real-time monitoring of cerebrovascular diseases.It is widely used clinically.With the gradual deepening of cerebrovascular disease research,the demand for transcranial Doppler ultrasound diagnostic equipment in blood flow information display and effective data storage is increasing.Due to the current variety of transcranial Doppler in the design of hardware modules,Software systems with high development costs and slower update speeds,short development cycles,and high development efficiency have become the focus of development.This study aimed at the analysis of the needs of transcranial Doppler ultrasound diagnostic equipment in the actual detection scenario,and designed a software system with complete control functions,real-time blood flow information feedback and complete data management.The system has stable and accurate control functions and a friendly human-computer interaction interface to ensure that medical personnel can easily and quickly input and retrieve information such as patients and blood vessels during use,and provide parameter adjustment and image freezing during the detection process.Operational functions such as view switching;information output shows blood flow information collected by transcranial Doppler system in a way that blood flow power spectrum display,spectral parameter calculation and sound output are displayed,so that medical personnel can monitor the brain of the examiner The blood vessel condition is accurately judged;at the same time,the system data management module completes the preservation of the detector information and the detection result,provides historical data query and parameter preset,and provides support for extending the TCD application.Thrombosis in the cerebral arteries can block the cerebral blood vessels and cause ischemic stroke.An important function of the transcranial Doppler ultrasound diagnostic equipment is to achieve thrombus detection.Traditional thrombus detection relies mainly on the manual observation of blood flow power spectrum by professionals.Subjective factors have large influence and long time.In order to overcome these shortcomings,based on the analysis of Doppler ultrasound signals,this study proposes an automatic thrombus detection system based on dual-tree complex wavelet transform.In order to solve the problem that the Doppler ultrasound signal can not distinguish the interference between the thrombus and the artificial artifact in the time domain,the system selects the double-tree complex wavelet transform with approximate translation invariance and anti-frequency aliasing from the frequency domain analysis algorithm,combined with the principal component.The analysis algorithm completes feature extraction and dimensionality reduction,and finally feeds the feature vector into the support vector machine to achieve thrombus recognition.According to the results of lateral contrast experiments with different frequency domain analysis algorithms,it is proved that the dual-tree complex wavelet transform improves the robustness of the coefficients extracted from Doppler ultrasound signals.The automated thrombus detection scheme designed in this study is fast,accurate and the strong. |
PDF Full Text Request