Application Of Magnetic Resonance Elastography And Magnetic Resonance Fingerprint Imaging In Pathological Classification And Stiffness Of Pituitary Tumors

Background and purposeThe pituitary gland is an important endocrine organ,which can produce some vasoactive substances and thus control the endocrine function of the whole body.Abnormalities in pituitary can lead to increased or decreased hormone secretion,causing changes in the body.Pituitary adenoma is a benign tumor originating from the adenohypophysis,and it accounts for 10-15%of intracranial brain tumors.Although pituitary tumors are benign tumors,some of them grow aggressively and extend beyond the sella,and then compress surrounding tissues,such as the optic chiasm,peripheral nerves,etc,resulting in visual field defects,ophthalmoplegia and other symptoms.Besides,pituitary microadenoma causes symptoms resulting from excessive or low hormone secretion.The preferred treatment for pituitary tumors is surgical resection except for prolactinoma,for which the preferred treatment is medication.Magnetic resonance imaging is the first choice for pituitary tumor imaging due to its inherent tissue contrast,direct multiplanar imaging and non-invasive imaging.The detection,postoperative recurrence and consistency of pituitary tumors with different parameters of conventional magnetic resonance have been explored.For example,use ADC value to assess the aggressiveness and hardness of pituitary tumors;perform imaging omics modeling on T1 and T2 values to predict tumor invasion;compare the detection rates of pituitary tumors with different field strengths;use magnetic resonance dynamic comparison Enhance or combine with PET functional imaging to identify postoperative changes and residual tumors or to identify small adenomas.This study will investigate the application of new techniques of magnetic resonance,namely magnetic resonance elastography and magnetic resonance fingerprint imaging,for pathological classification of pituitary tumors.It is possible to guide the preoperative medication of tumors and reduce the scope of tumor resection by the classification of pituitary tumors.Besides,doctors can achieve receptor targeted therapy for specific pathological types by combining pathological classification of pituitary tumors with the expression of receptors of different subtypes.Magnetic resonance elastography,which has recently emerged as a non-invasive examination for head imaging,can reflect the condition of the nervous system,aging and cognitive function.The external mechanical actuator generates mechanical waves which go through the"pillow"placed under the occipital bone,causing small vibration displacement.Post-processing procedures convert mechanical parameters into tissue characteristics.The research on the brain has gradually transformed from the function-structure relationship to the research on the hardness of brain tissue.Magnetic resonance elastography is an imaging index that is more sensitive to the structural integrity of the brain.With the increase of age,the hardness of the brain tissue decreases.Shear waves travel faster in hard tissues than in soft tissues,and the hardness of the brain tissue decreases when nerve cells are demyelinated and Alzheimer’s disease occurs.At the same time,magnetic resonance elastography can also predict the hardness of the tumor before surgery.Previous studies have used magnetic resonance elastography to image gliomas and the results show that gliomas are softer than normal brain tissue,and their hardness decreases with the increase of the tumor grade.Magnetic resonance fingerprint imaging uses a pseudo-random acquisition method to obtain different parameter information at the same time in the under-sampling technology,and then matches the data with the dictionary,saving scanning time.The Bloch equations are applied to simulate sequences of T1,T2,etc.to form a dictionary of corresponding organizations.The parameters of the scanned object are compared and searched with the dictionary to obtain the organization characteristic information.In this study,the applications of magnetic resonance elastography and magnetic resonance fingerprint imaging in the pathological grading and hardness of pituitary tumors will be investigated.Part 1 Application of Magnetic Resonance Elastography in Pathological Classification and Stiffness of Pituitary TumorsObject:To explore the application value of magnetic resonance elastography in the pathological classification and stiffness of pituitary tumors.Materials and methods:32 patients with pituitary tumors in our hospital were examined by preoperative MRE from January 1^th,2019 to October 27^th,2019,and patients who had undergone surgical treatment were excluded.Radiologists measured the stiffness values of MRE,and then the values were compared with the pathological classification of pituitaries.Secondly,the MRE values of pituitary tumor were compared with the intraoperative stiffness assessment.ROC and independent sample t test analysis were used for the comparison between gonadotropin pituitary tumors and other pituitary tumors,functional adenomas and non-functional adenomas by using SPSS25.Results:For the comparison of the stiffness value of gonadotropin pituitary tumors and other pituitary tumors,the area under the curve（AUC）is 0.651,while the cut-off value of these two groups is 1178.190,with the sensitivity of 85.7%,and the specificity of 60.0%.The independent sample t test results between the two groups showed that it was of no statistically significant（P>0.05）.The independent sample t test results between functional pituitary adenomas and non-functional pituitary adenomas showed that it was of no statistically significant（P>0.05）.Only 1 case of pituitary tumor is hard,which was not statistically analyzed.Conclusion:The MRE is of average value for the pathological classification of pituitary tumors.Part 2 Application of Magnetic Resonance Fingerprint Imaging in Pathological Classification and Stiffness of Pituitary TumorsObject: To explore the application value of magnetic resonance fingerprinting in the pathological classification and stiffness of pituitary tumors.Materials and methods: 58 patients with pituitary tumors for treatment in our hospital during the period from December 5th,2018 to November 29 th,2019 without surgery were enrolled and MRF examinations were performed.The MRF-T1 values and MRF-T2 values were obtained,and the correlations between these values and the pathological classification of pituitary tumors were investigated respectively.The ROC analysis and independent sample t test method were adopted for inspection.Results: The independent sample t test for the MRF-T1 values showed that there was no statistically significant between functional pituitary adenomas and non-functional pituitary adenomas（P>0.05）.The independent sample t test for the MRF-T1 values between gonadotropin pituitary tumors and other functional pituitary adenomas showed that it was of statistically significance between these two groups（P<0.05）.The ROC for the MRF-T1 values of gonadotropin pituitary tumors showed that AUC=0.832.The cut-off value of the MRF-T1 values of gonadotropin pituitary tumors and other pituitary tumors was 1566.65,with the sensitivity of 77.8% and the specificity of 76.9%.The independent sample t test showed that it was of statistically significance between these two groups（P<0.05）.However,there was no statistical significance for the comparison of the MRF-T2 values between functional pituitary adenomas and non-functional pituitary adenomas,neither for the gonadotropin pituitary tumors and other pituitary adenomas（P>0.05）.Only 1 case of pituitary tumor was hard,and the rest were soft,and so no statistical analysis was conducted.Conclusion: The MRF-T2 value is of average value for the classification of pituitary tumors,nevertheless,the MRF-T1 may be used as an imaging method to distinguish gonadotropin pituitary tumors from other pituitary tumors.The MRF-T1 value 1566.65 can be used as a cut-off value.