Research On The Fat Quantification And Its Application Based On Magnetic Resonance Imaging

Adipose tissue is widely distributed in the subcutaneous,abdominal,bone marrow and organs.It is not only an important tissue for human energy storage,but also an important endocrine and immune organ.Abnormal fat metabolism is closely related to the occurrence and development of many diseases,especially metabolic syndrome（obesity,hyperglycemia,hypertension,dyslipidemia）,nonalcoholic steatohepatitis and other major diseases that seriously endanger the health of the human.Studies have shown that the characteristics of body fat distribution are closely related to the risk of obesity and related metabolic diseases.However,the existing anthropological diagnostic methods such as waist circumference and Body Mass Index（BMI）cannot accurately reflect the information of human body fat distribution,also cannot describe the metabolic activity of different adipose tissues.Therefore,it has become an urgent need for clinical diagnosis and treatment to use a new type of fat quantitative technology to evaluate the problems related to human fat metabolism.Magnetic resonance imaging（MRI）can not only visualize and accurately quantify the distribution of whole body fat,but also reflect the metabolic activity of special adipose tissue through a variety of functional imaging methods.Therefore,starting from the MRI fat quantification technology developed in the project team,this thesis carries out the following three research work,and explores the important value of MRI technology in fat metabolism-related research.Firstly,in order to more accurately reflect the fat distribution characteristics of various parts of the human body.Based on MRI Proton Density Fat Fraction（PDFF or FF）,combined with an automatic analysis tool for whole body fat distribution based on deep learning network,this thesis accurately measures the adipose tissue volume of total adipose tissue,subcutaneous adipose tissue,internal adipose tissue,abdominal subcutaneous adipose tissue,visceral adipose tissue and fat distribution characteristic parameters.Correlation analysis with traditional anthropometric measures showed that BMI was only highly correlated with total adipose tissue volume and subcutaneous adipose tissue volume,and could not reflect the degree of visceral adipose tissue deposition,so it had insufficient specificity for predicting adipose tissue distribution.The analysis shows that there are significant differences in the adipose tissue distribution characteristics of men and women,indicating that the adipose tissue distribution characteristics of men and women are different,which cannot be reflected by traditional anthropometric parameters such as BMI.Secondly,to assess the the metabolic activity of brown adipose tissue（BAT）,this thesis used the interscapular BAT（i BAT）of rats and activated by（Norepinephrine,NE）,the dual-step iterative temperature estimation（DITE）fat-referenced proton resonance frequency shift（PRFS）method by MRI was used to measure changes in temperature and fat fraction in i BAT.When i BAT was activated with different concentrations of NE,the results showed that the temperature change of i BAT increased with the dose of NE,while the fat fraction of i BAT was lower,suggesting that high concentrations of NE promoted i BAT to consume more fatty acids and produced a more calories.Biochemical assay results showed that the expression of thermogenic gene uncoupling protein-1（UCP-1）increased with increasing NE concentration,consistent with MRI measurements of temperature and fat fraction.The thesis demonstrates that the DITE fat-referenced temperature imaging technique by MRI can non-invasively and dynamically assess the temperature changes and fat content changes of BAT,thereby reflecting the metabolic activity of BAT.Thirdly,abnormal metabolism of fats can trigger ectopic fat deposition in organs other than adipose tissue.Non-alcoholic steatohepatitis（NASH）is one of the diseases caused by ectopic fat deposition in liver.Therefore,in view of the lack of early non-invasive diagnostic techniques for detecting of NASH progression,NASH rat model was constructed by feeding Methionine Choline Deficient（MCD）diet in this study,and used a MRI multi-parameter quantitative method to produce the T1 value of the water component（water specific T1,or w T1）in liver,the fat fraction of liver could be obtained simultaneously from the same sequence for exs examining steatosis.Meanwhile,we combined with a liver-specific contrast agent of Primovist（Gd-EOB-DTPA）to explore the relationship between w T1 and early progression characteristics of NASH.The results showed that a moderate but significant negative correlation between w T1 and histopathological inflammation grades was observed（r_s=-0.41）,and the w T1 could differentiate none to mild inflammation from moderate to severe inflammation in the early stage of the NASH rat model（AUC=0.78,p=0.0003）.But Gd-EOB-DTPA does not provide better diagnostic performance for identifying inflammation progression in NASH.