Research On Measuring Method Of Medical MRI Image Quality Parameters

In recent years,Magnetic Resonance Imaging(MRI)technology has been highly recognized by patients.It also has become an important auxiliary method for doctors'diagnosis due to its outstanding advantages such as non-invasive,non-ionizing radiation and strong resolving power of soft tissue.With the rapid increase of medical MRI equipment,magnetic resonance inspection has become very popular.Only by ensuring the stability of the magnetic resonance imaging equipment operation status and the accuracy of image information can provide sufficient and effective information for the judgment of illness.Therefore,how to ensure the quality of magnetic resonance images has become the focus of measurement work.Our roject started from the actual metrology and inspection work,explored the impact factors of the medical magnetic resonance image signal to noise ratio,image uniformity,spatial linearity,layer thickness,image artifacts,high contrast resolution and low contrast resolution.These parameters are core metrological parameters.We analyze the advantages and disadvantages of various evaluation methods from the aspects of theory and experimental verification,and propose new evaluation methods to ensure the accuracy of measurement parameters for future development.At the same time,the feasibility of automatic evaluation of image quality was explored to further improve the detection efficiency and accuracy.Based on this,this paper mainly studies the following contents:First,we conduct research and comparative analysis on the current status of medical magnetic resonance testing at home and abroad,current standards,test patterns,and testing procedures,as well as the evaluation methods of 7 core metrological parameters.We found that there are problems such as inconsistency in the presence of motifs,evaluation methods,and evaluation criteria.Based on actual testing experience and experimental analysis,the most suitable testing procedure in practical operation is summarized.At the same time,through analyzing the detection principle,combined with the existing detection conditions,we put forward our own innovative evaluation methods of individual metrological parameters,and carry out practical verification.Second,image quality is mainly determined by both the device's own performance and post-processing.What's more,image quality of current equipment in metrology and inspection is mainly determined by the used scanning phantom,scanning conditions and evaluation methods.We selected 8 sets of superconducting magnetic resonance equipment and combined the two most widely used detection phantoms(ACR phantom and SMR170 phantom)to perform gradient design on detection conditions.And then,we analyzed in detail the above-mentioned factors,and summarized the optimization verification procedures under various testing conditions,including our innovative methods combined with actual conditions.Third,automatic evaluation is the mainstream development trend of future metrology and inspection evaluation.It can not only improve the detection efficiency,but also reduce the human error and improve the detection accuracy.This paper first analyzes the feasibility of software and hardware for automated analysis of current experiments.At the same time,it presents the flow chart for automatic evaluation of those 7 core metrological parameters,and selects image signal to noise ratio and layer thickness for preliminary experimental research.Our experiment verified the feasibility of automatic evaluation of image quality and achieved some results.