The Application Of Cardiac Compressed Sensing Cine Magnetic Resonance Imaging In Heart Failure And And Fully Quantitative Rest Myocardial Perfusion Magnetic Resonance Imaging In Myocardial Ischemia

Part 1 The Application of Compressed Sensing Cine of Cardiac Magnetic Resonance Imaging in Heart Failure PatientsBackground and Purposes:A standard multiple-breath-hold cine(MBH-cine)cardiac magnetic resonance(CMR)imaging can classify heart failure(HF)to guide the treatment of HF.However,some HF patients who undergo a standard MBH-cine examination often have difficulty performing multiple breath-holds.Compressed sensing(CS)is a new signal acquisition method that can improve the scan speed of CMR cine imaging.The CS cine can complete the left ventricular scan in a single-breath-hold(SBH)period.At present,overseas studies have shown that the CS cine during a single breath-hold and/or during free breathing can obtain good image quality and reliable cardiac function parameters,but the study sample sizes are relatively small.In addition,there is no research report on the classification of HF by the CS cine.The purposes of this study were to evaluate the image quality differences,the cardiac function differences,and the agreement on HF classifications between the standard MBH-cine and the CS SBH-cine.Materials and Methods:154 patients with known or suspected HF underwent 3 T CMR using a standard MBH-cine and a CS SBH-cine and were divided into the non-atrial-fibrillation(NAF)group(n=121)and the atrial fibrillation(AF)group(n=33).The image quality was independently analyzed by the two observers according to the image features on a 5 point scale(1=nondiagnostic,5=excellent).The software was used to aquire the left ventricular(LV)ejection fraction(LVEF),LV mass index(LVMI),LV end diastolic volume(LVEDV),and LV end systolic volume(LVESV).The HF were divided into HF with a preserved ejection fraction,HF with a mid-range ejection fraction,and HF with a reduced ejection fraction based on LVEF.The interobserver agreement on image quality of standard MBH-cine and the interobserver agreement on image quality of CS SBH-cine were analyzed using the kappa test.The method to compare the image quality of two sequences was T-test.The cardiac parameters was compared using the Wilcoxon signed-rank test.The Bland-Altman analyses was to analyze the agreement and the linear regression was used to analyze the correlation of the cardiac parameters acquired by both methods.The same statistical methods were used to analyze the interobserver and intraobserver agreement and the correlation of the cardiac function of the CS SBH-cine.The method to analyze the agreement on HF classifications of two sequences was the kappa test.Results:1.The scan time for the CS SBH-cine(18.80 s±1.30 s)was reduced by 88%compared with that of the standard MBH-cine(157.60 s±8.10 s).2.The interobserver agreement(kappa score=0.78,P<0.05)of image quality for the CS SBH-cine and the interobserver agreement(kappa score=0.86,P<0.05)of image quality for the standard MBH-cine were good.The image quality scores acquired by the first observer were selected to analyze the image quality.In AF group,the image quality acquired by standard MBH-cine was slightly decreased compared with the CS SBH-cine(3.3 ± 0.5 vs.3.8±0.7,T=-4.50,P<0.05).In the NAF group,the image quality acquired by standard MBH-cine was slightly higher(4.7± 0.5 vs.4.5±0.6,T=5.04,P<0.05).3.LVMI,LVEF,LVEDV,and LVESV acquired by the CS SBH-cine were not significantly different(all P>0.05)from those acquired by the standard MBH-cine.The correlation and agreement of the LV cardiac parameters acquired by two methods were excellent.The interobserver and intraobserver correlation and agreement of the quantitative LV function calculated from the CS SBH-cine were excellent.4.The result displayed excellent agreement(kappa=0.93,P<0.05)on HF classifications of two sequences.Conclusion:1.The CS SBH-cine could greatly shorten the scan time and improve the comfort of examination.2.The image quality of the CS SBH-cine could meet the diagnostic needs.The standard MBH-cine image quality score was slightly decreased compared with the CS SBH-cine in patients with AF.3.The CS SBH-cine could accurately evaluate cardiac function.It was feasible and reliable to classify HF using the CS SBH-cine.Part 2The Application of Fully Quantitative Rest Perfusion of Cardiac Magnetic Resonance Imaging in Myocardial Ischemia Background and Purposes:All kinds of heart diseases may be complicated with myocardial ischemia.There are many imaging methods to evaluate myocardial ischemia.The nuclide myocardial perfusion imaging is recognized as the most reliable noninvasive imaging method to diagnose.myocardial ischemia.With the development of computed tomography(CT)technology,CT myocardial perfusion has also been used to evaluate myocardial ischemia.But these methods have radiation,which limits their clinical application to a certain extent.Invasive fractional flow reserve and non-invasive CT-based fractional flow reserve can indirectly diagnose myocardial ischemia by evaluating the functional stenosis of coronary artery.Cardiac magnetic resonance(CMR)perfusion imaging is non-invasive and non-radioactive method and has good diagnostic value in evaluating myocardial ischemia.The qualitative evaluation of CMR perfusion imaging is the main way in clinical work.This study sought to explore the feasibility of fully quantitative rest perfusion of CMR for evaluating myocardial ischemia against a reference standard of the rest single-photon emission computed tomography myocardial perfusion imaging(SPECT-MPI)and compare the fully quantitative rest perfusion of CMR with qualitative and semi-quantitative for evaluating myocardial ischemia.Materials and Methods:Thirty patients with known or suspected myocardial ischemia(twenty males,ten females;mean age 45.10 ±15.55 years)performed rest SPECT-MPI and fully quantitative rest perfusion of CMR,and the interval time between the two examinations was less than three days.Ten healthy volunteers(four males,six females;mean age 45.50 ± 10.07 years)performed fully quantitative rest perfusion of CMR.Semiquantitative(Slope Max,Peak,and Time to Peak)and fully quantitative parameters(myocardial blood flow,MBF;relative myocardial blood flow,rMBF)of left ventricular myocardial perfusion were obtained using fully automated cardiac specific software.The visual evaluation was used to qualitatively analyze rest SPECT-MPI and rest perfusion of CMR.The patients' myocardial segments were divided into the group of ischemic segments and the group of non-ischemic segments according to the rest SPECT-MPI.The differences of MBF and the differences of rMBF among the groups of ischemic segments,non-ischemic segments and healthy volunteers' myocardial segments were compared using one-factor analysis of variance.Receiver operating characteristic(ROC)curves were used to analyze the qualitative evaluation,the semiquantitative evaluation and the fully quantitative evaluation of the rest perfusion of CMR versus a reference standard of rest SPECT-MPI.Results:1.There were 160 myocardial segments in ten healthy volunteers.Thirty patients'myocardial segments were divided into 165 ischemic segments and 315 non-ischemic segments according to the rest SPECT-MPI.2.There were statistically significant differences(both P<0.05)in MBF and rMBF between the group of ischemic segments(MBF 1.33±0.47 ml/min/g;rMBF 0.70 ± 0.18)and the group of non-ischemic segments(MBF 1.62 ± 0.55 ml/min/g;rMBF 0.81 ± 0.17).There were statistically significant differences(both P<0.05)in MBF and rMBF between the group of ischemic segments and the group of healthy volunteers'myocardial segments(MBF 1.46±0.42 ml/min/g;rMBF 0.81 ± 0.16).There was statistically significant difference(P<0.05)in MBF between the group of non-ischemic segments and the group of healthy volunteers' myocardial segments.However,there was no statistically significant difference(P=0.98)in rMBF between the group of non-ischemic segments and the group of healthy volunteers' myocardial segments.3.The results of ROC curves showed that the area under the curve(AUC)of rMBF and the AUC of MBF were both larger than the AUC of the qualitative assessment(0.67 vs.0.57,0.65 vs.0.57,both P<0.05).The AUC of Slope Max was larger than the AUC of the qualitative assessment(0.66 vs.0.57,P<0.05).The AUC of rMBF and the AUC of MBF were both larger than the AUC of the Peak(0.67 vs.0.58,0.65 vs.0.58,both P<0.05).Conclusions:1.The fully quantitative rest perfusion of CMR could objectively evaluate myocardial ischemia and has potential clinical application value.2.The diagnostic value of the fully quantitative rest perfusion of CMR was superior to that of the semiquantitative assessment and the qualitative assessment.