Research And Design Of Microwave Imaging Algorithms For Stroke Diagnosis

Stroke is a sudden disease of blood circulation disorder in the brain.It is caused by a variety of predisposing factors,such as hypertension,coronary heart disease,aging,etc.,causing clogging or rupture of blood vessels in the brain,leading to various aspects of the human brain.The function is declining to varying degrees,even life in crisis.Early rapid diagnosis plays a very important role due to its significant high morbidity and mortality.However,current mainstream detection techniques such as CT and MRI have obvious defects and deficiencies in the detection and diagnosis of stroke.In recent years,with the development of microwave imaging technology,it has played a huge role in more and more fields.Microwave imaging technology is an emerging non-destructive testing method.It records low-power microwave signals into the target imaging area,records and processes the received scattered signals,and reconstructs images according to certain methods.It has great application potential in the medical field.Considering the unique advantages of microwave imaging technology,this paper applies it to the diagnosis of stroke,and studies and designs the applicable microwave imaging method.At present,the mainstream imaging methods are mainly divided into microwave tomography methods and microwave radar imaging methods according to their principles.From the perspective of signal processing,they can be divided into two types of microwave imaging methods: time domain and frequency domain.Indepth study of focal imaging,time-reversal confocal imaging,UWB microwave imaging based on Huygens principle and optimized imaging based on subspace.Based on the MATLAB platform,the modeling and simulation of the head model and imaging algorithm were completed.At the same time,the imaging algorithm was optimized and improved for the particularity of stroke diagnosis.The computer simulation results show that the performance of the imaging algorithm designed in this paper basically meets the design requirements.In order to further verify the feasibility of the designed microwave imaging method in real environment,this paper establishes a simple experimental system for simulating stroke detection.The receiving and transmitting antennas use a UWB antenna and a vector network analyzer as a signal source.The target of the test selected the head model of the PVC material,filling its cavity with different substances to simulate the real intracranial condition.The scattering field data reception is completed by uniformly arranging 16 measurement positions in the circumference of the target to be tested.The received scattering field data is recorded and processed,and the image reconstruction is completed by the method designed in this paper.The results show that the time-delayed confocal imaging method and the subspace-based optimization method have better imaging performance.