| Purpose: The object of our study is to determine quantitative parameters in DYNFS (dynamic fat saturation, DYNFS) and LAVA (liver acquisition with volume acceleration, LAVA) sequences, and to discuss the role in displaying the pulmonary vessels by four following contrast parameters as SI (signal intensity), SNR (signal noise ratio), CNR (contrast to noise ratio) and SIR (signal intensity ratio), and to compare which is the best enhanced sequences in displaying pulmonary vessels. It can provide reliable parameters and objective basis for clinical in comparison sequences of pulmonary vascular scanning. Investigate the enhancement degree of the lung parenchyma. It can provide a standard of quantitative perfusion range for clinical. And this standard may be help to diagnosis pulmonary embolism and analyze the perfusion defects. And may be provide the informations of pulmonary function for clinical.Method: From 2008 to 2010, 25 cases met the inclusion criteria were enrolled into our study. All patients underwent conventional chest MRI sequences. 15 cases did the DYNFS enhanced scan firstly, subsequently did the LAVA3D enhanced scan in the second check. While the other 10 cases did the LAVA3D enhanced scan firstly, then did the DYNFS enhanced scan. By using post-processing software workstations (AW4.1), the MIP (maximum intensity projection, MIP) and the partial volume reconstruction were performed from original coronal enhanced images. The original images and reconstruct images of MIP were used for quantitative evaluation the research of objective method and subjective method.Our study is divided into three parts:The first part is the measurement of the objective method. Using four parameters quantitative measure the images respectively and choosing the best compare parameter. The Reference of choosing parameters: It is account of the better parameter which can enlarge signals'difference between pulmonary artery and lung tissue, and it suffer the lest affecting by other factors. Taking into account the complexity of the study, we used a single image method in the objective research and drew the ROI (regions of interest, ROI) in the coronary images of the aortic center plane. In the corresponding region of interest area, the right pulmonary artery SI, lung field SI, bilateral background SI were measured, then calculated the SNR of the right pulmonary artery, the right lung's SNR, the left lung's SNR and the average SNR of both lungs with the following formula. The CNR and the SIR between pulmonary artery and lungs were calculated also. Every parameter's effect was compared and the advantages of showing blood vessels were evaluated between the DYNFS and LAVA3D enhanced sequences. SIR=SI pulmonary artery / SI lungs SNR lung = SI lung / SD background SNR pulmonary artery = SI pulmonary artery / SD background CNR = (SI pulmonary artery─SI lung) / SD backgroundThe second part is evaluation which sequence is better display the pulmonary vessels using subjective method. The grade of pulmonary artery and vein were statistics with Branch method step by step on the corresponding coronal plane images, and subsequently give a mark to the image with the scoring standard. Statistical analysis the differences of displaying pulmonary vessels on the DYNFS and LAVA3D enhanced sequences. In accordance with the situation of pulmonary vessels, the enhancement of the lung parenchyma was divided into 0 degree, 1 degree, 2 degree and 3 degree. The relationship between image quality and the enhancement degree of lung parenchyma was Investigated. And the quantitative range of the lung parenchyma in general was determined.The third part is compare the results of subjective method with the objective method. SPSS16.0 Statistical software was used to make Paired t test statistical analysis of SI, SNR, CNR and SIR. Both parameters were used in identifying DYNFS and LAVA3D enhanced sequence, Choose the sequence which can display the Pulmonary vessels clearer and select optimal parameters for quantitative evaluate both enhanced sequence.Results:1 Quantitative and objective method results:(1) Left lung's SI compare to right, DYNFS and LAVA3D sequences all show the result of left lung's SI is higher than the right lung (P=0).(2) The average SI comparison of the lung , DYNES enhanced sequence is larger than LAVA3D enhanced sequence (P=0).(3) The average SI comparison of the pulmonary artery , The result of DYNFS and LAVA3D enhanced sequences has no significant deviation (P=0.97).(4) The average SI comparison of the bilateral background, The result of DYNFS and LAVA3D enhanced sequences has no significant deviation (P=0.135).(5) The average SNR comparison of the lung, DYNFS enhanced sequence is larger than LAVA3D enhanced sequence (P=0.029).(6) Compare the pulmonary artery SNR, DYNES enhanced sequence is less than JAVA3D enhanced sequence (P=0.001).(7) The CNR comparison of DYNFS and LAVA3D enhanced sequence, DYNFS enhanced sequence is less than LAVA3D enhanced sequence (P=0).(8) The SIR comparison of DYNFS and LAVA3D enhanced sequence, DYNFS enhanced sequence is less than LAVA3D enhanced sequence (P=0).(9)Comparison of pulmonary artery SNR and CNR, DYNFS and LAVA3D enhanced sequences all show the pulmonary artery SNR is greater than CNR (P=0).(10) CNR compared with the SIR in the DYNFS enhanced sequence, the signal enlarge capacity of CNR between pulmonary vessels and lung tissue is greater than SIR(P=0). And in the LAVA3D enhanced sequence, the signal enlarge capacity of CNR is greater than SIR(P=0) and this capacity is better than DYNFS enhanced sequence.2 Subjective method results:(1) The comparison of pulmonary artery series between DYNFS and LAVA3D enhanced sequences, DYNFS enhanced sequence is less than LAVA3D enhanced sequence (P=0.001).(2) The comparison of pulmonary vein series between DYNFS and LAVA3D enhanced sequences, DYNFS enhanced sequence is less than LAVA3D enhanced sequence (P=0).(3) The comparison of displaying series between pulmonary arteries and veins in DYNFS enhanced sequence, the veins display series is superior than pulmonary artery (P=0).(4) The comparison of displaying series between pulmonary arteries and veins in LAVA3D enhanced sequence, the veins display series is superior than pulmonary artery (P=0).(5) The comparison of scores between DYNFS and LAVA3D enhanced sequences, DYNFS enhanced sequence's manifestation: 4 scores have 0 cases (0%), 3 scores have 6 cases (24%), 2 scores have 11 cases (44%), 1 scores have 8 cases (32% ); LAVA3D enhanced sequence's manifestation: 4 scores have 12 patients (48%), 3 scores have 6 cases (24%), 2 scores have 3 patients (12%), 1 scores have 4 patients (16%).Conclusion:1 In objective method, SI is not appropriate for evaluating DYNFS and LAVA3D enhanced sequence to show advantages and disadvantages of pulmonary vessels. The SNR of Pulmonary artery, CNR and SIR all can be used to compare MRI pulmonary vascular image quality. Although the SNR of Pulmonary artery is greater than CNR, the changes of signal intensity make a great influence of SNR, coupled with the characteristic influence of the fat suppression sequences, the SNR's values of Practical Applications is limited. while SIR has excluded the influence of noise, but it's capacity of enlarge signals'difference between pulmonary artery and lung is relatively small. CNR not only reduced the dependence of the SI but also amplified the signal difference between the two tissues. therefore CNR becomes the best parameter to display pulmonary vascular in evaluating DYNFS and LAVA3D enhanced sequences.2 Objective and subjective methods'results all show LAVA3D enhanced sequence is superior than DYNFS enhanced sequence in evaluation of pulmonary vascular. Two methods'results were consistent.3 The SI's degree of lung parenchyma as follows: For DYNFS enhanced sequences, 1 degree: 22.63 (19.13 ~ 25.09). 2 degree: 29.58 (20.89 ~ 41.09). 3 degrees: 47.14 (35.74 ~ 59.89). For LAVA3D enhanced sequence, 0 degree: 16.52 (13.11 ~ 20.97). 1 degree: 23.79 (14.31 ~ 33.42). 2 degree: 24.51 (19.65 ~ 30.64).
|
PDF Full Text Request