TI Scout Research Articles

Automated inversion time selection for late gadolinium–enhanced cardiac magnetic resonance imaging

ObjectivesTo develop and share a deep learning method that can accurately identify optimal inversion time (TI) from multi-vendor, multi-institutional and multi-field strength inversion scout (TI scout) sequences for late gadolinium enhancement cardiac MRI.Materials and methodsRetrospective multicentre study conducted on 1136 1.5-T and 3-T cardiac MRI examinations from four centres and three scanner vendors. Deep learning models, comprising a convolutional neural network (CNN) that provides input to a long short-term memory (LSTM) network, were trained on TI scout pixel data from centres 1 to 3 to identify optimal TI, using ground truth annotations by two readers. Accuracy within 50 ms, mean absolute error (MAE), Lin’s concordance coefficient (LCCC) and reduced major axis regression (RMAR) were used to select the best model from validation results, and applied to holdout test data. Robustness of the best-performing model was also tested on imaging data from centre 4.ResultsThe best model (SE-ResNet18-LSTM) produced accuracy of 96.1%, MAE 22.9 ms and LCCC 0.47 compared to ground truth on the holdout test set and accuracy of 97.3%, MAE 15.2 ms and LCCC 0.64 when tested on unseen external (centre 4) data. Differences in vendor performance were observed, with greatest accuracy for the most commonly represented vendor in the training data.ConclusionA deep learning model was developed that can identify optimal inversion time from TI scout images on multi-vendor data with high accuracy, including on previously unseen external data. We make this model available to the scientific community for further assessment or development.Clinical relevance statementA robust automated inversion time selection tool for late gadolinium–enhanced imaging allows for reproducible and efficient cross-vendor inversion time selection.Key Points• A model comprising convolutional and recurrent neural networks was developed to extract optimal TI from TI scout images.• Model accuracy within 50 ms of ground truth on multi-vendor holdout and external data of 96.1% and 97.3% respectively was achieved.• This model could improve workflow efficiency and standardise optimal TI selection for consistent LGE imaging.

Automated inversion time selection for black-blood late gadolinium enhancement cardiac imaging in clinical practice

ObjectiveTo simplify black-blood late gadolinium enhancement (BL-LGE) cardiac imaging in clinical practice using an image-based algorithm for automated inversion time (TI) selection.Materials and methodsThe algorithm selects from BL-LGE TI scout images, the TI corresponding to the image with the highest number of sub-threshold pixels within a region of interest (ROI) encompassing the blood-pool and myocardium. The threshold value corresponds to the most recurrent pixel intensity of all scout images within the ROI. ROI dimensions were optimized in 40 patients’ scans. The algorithm was validated retrospectively (80 patients) versus two experts and tested prospectively (5 patients) on a 1.5 T clinical scanner.ResultsAutomated TI selection took ~ 40 ms per dataset (manual: ~ 17 s). Fleiss’ kappa coefficient for automated-manual, intra-observer and inter-observer agreements were overline{kappa }= 0.73, overline{kappa } = 0.70 and overline{kappa } = 0.63, respectively. The agreement between the algorithm and any expert was better than the agreement between the two experts or between two selections of one expert.DiscussionThanks to its good performance and simplicity of implementation, the proposed algorithm is a good candidate for automated BL-LGE imaging in clinical practice.

Cardiac magnetic resonance imaging in patients with left bundle branch block: Patterns of dyssynchrony and implications for late gadolinium enhancement imaging.

Left bundle branch block (LBBB) is a ventricular conduction delay with high prevalence. Aim of our study is to identify possible recurring patterns of artefacts in late gadolinium enhancement (LGE) imaging in patients with LBBB who undergo cardiac magnetic resonance imaging (MRI) and to define parameters of mechanical dyssynchrony associated with artefacts in LGE images. Fifty-five patients with LBBB and 62 controls were retrospectively included. Inversion time (TI) scout and LGE images were reviewed for artefacts. Dyssynchrony was identified using cardiac MRI by determining left ventricular systolic dyssynchrony indices (global, septal segments, and free wall segments) derived from strain analysis and features of mechanical dyssynchrony (apical rocking and septal flash). Thirty-seven patients (67%) with LBBB exhibited inhomogeneous myocardial nulling in TI scout images. Among them 25 (68%) patients also showed recurring artefact patterns in the septum or free wall on LGE images and artefacts also persisted in 18 (72%) of those cases when utilising phase sensitive inversion recovery. Only the systolic dyssynchrony index of septal segments allowed differentiation of patient subgroups (artefact/no artefact) and healthy controls (given as median, median ± interquartile range); LBBB with artefact: 10.44% (0.44-20.44%); LBBB without artefact: 6.82% (-2.18-15.83%); controls: 4.38% (1.38-7.38%); p < 0.05 with an area under the curve of 0.863 (81% sensitivity, 89% specificity). Septal flash and apical rocking were more frequent in the LBBB with artefact group than in the LBBB without artefact group (70 and 62% versus 33 and 17%; p < 0.05). Patients with LBBB show recurring artefact patterns in LGE imaging. Use of strain analysis and evaluation of mechanical dyssynchrony may predict the occurrence of such artefacts already during the examination and counteract misinterpretation.

DIAGNOSIS OF CARDIAC AMYLOIDOSIS BASED ON TEMPORAL SEQUENCE OF NULLING ON TI SCOUT: A POOR MAN'S T1 MAP

Non-invasive diagnosis of cardiac amyloidosis may obviate the need for biopsy. The aim of our study was to determine the utility of identifying the type of nulling pattern on TI Scout sequence in diagnosis of cardiac amyloid infiltration. Materials and Methods-This retrospective observational study

Automated selection of myocardial inversion time with a convolutional neural network: Spatial temporal ensemble myocardium inversion network (STEMI-NET).

Delayed enhancement imaging is an essential component of cardiac MRI, which is used widely for the evaluation of myocardial scar and viability. The selection of an optimal inversion time (TI) or null point (TINP ) to suppress the background myocardial signal is required. The purpose of this study was to assess the feasibility of automated selection of TINP using a convolutional neural network (CNN). We hypothesized that a CNN may use spatial and temporal imaging characteristics from an inversion-recovery scout to select TINP , without the aid of a human observer. We retrospectively collected 425 clinically acquired cardiac MRI exams performed at 1.5 T that included inversion-recovery scout acquisitions. We developed a VGG19 classifier ensembled with long short-term memory to identify the TINP . We compared the performance of the ensemble CNN in predicting TINP against ground truth, using linear regression analysis. Ground truth was defined as the expert physician annotation of the optimal TI. In a backtrack approach, saliency maps were generated to interpret the classification outcome and to increase the model's transparency. Prediction of TINP from our ensemble VGG19 long short-term memory closely matched with expert annotation (ρ = 0.88). Ninety-four percent of the predicted TINP were within ±36 ms, and 83% were at or after expert TI selection. In this study, we show that a CNN is capable of automated prediction of myocardial TI from an inversion-recovery experiment. Merging the spatial and temporal characteristics of the VGG-19 and long short-term-memory CNN structures appears to be sufficient to predict myocardial TI from TI scout.

Comparison of quantitative imaging parameters using cardiovascular magnetic resonance between cardiac amyloidosis and hypertrophic cardiomyopathy: inversion time scout versus T1 mapping

To compare inversion time (TI) parameters, native T1, and extracellular volume (ECV) on cardiac magnetic resonance (CMR) imaging between patients with cardiac amyloidosis (CA) or hypertrophic cardiomyopathy (HCMP). Forty six patients with biopsy-confirmed CA and 30 patients with HCMP who underwent CMR were included. T1 and TI values were measured in the septum and cavity of the left ventricle on T1 mapping and TI scout images. TI values were selected at nulling point for each myocardium and blood pool. Native T1, ECV, and TI interval values were significantly different between the CA (1170.5 ± 86.4ms, 56.7 ± 12.2, -11.5 ± 28.4ms) and HCMP (1059.5 ± 63.4ms, 28.5 ± 5.8, 66.2 ± 25.4ms) (all p < 0.001). The diagnostic performance of the TI interval (area under the ROC curve, 0.975) was not inferior to that of the ECV (0.980, p = 0.776), and it was superior to that of the native T1 (0.845, p = 0.004). The diagnostic performance of TI interval was comparable to that of ECV for differential diagnosis between CA and HCMP. TI interval showed the feasibility as quantitative CMR parameter when T1 mapping images are not available.

Improved inversion time (TI) scout sequence for late gadolinium enhancement MRI of patients with implantable cardiac devices

Background Late gadolinium enhancement (LGE) cardiac MRI is the clinical gold standard for non-invasive characterization of myocardial scar [1]. The LGE technique is a contrast enhanced inversion recovery (IR) FLASH sequence, and requires a priori knowledge of the initial inversion time (TI) to produce optimal contrast between healthy myocardium and scar substrate. The TI is typically assessed using a Look-Locker based TI scout sequence. Recent advances [2,3] have enabled successful application of LGE MRI to patients with cardiac implantable devices at our institution. Presence of cardiac devices, such as implantable cardioverter defibrillators (ICD), gives rise to severe offresonance in the myocardium. This produces banding artifacts in standard TI scout images which cannot be used to estimate the initial TI for LGE imaging. This can result in LGE images with poor contrast and delays in patient scanning. In this abstract, we present a modification to the standard TI scout sequence that prominently reduces artifacts in TI scout images. Methods The TI scout sequence is an inversion prepared cine sequence with TrueFISP readout. Severe off-resonance from an ICD produces severe banding artifacts in TrueFISP images. In addition, in IR sequences, severe off-resonance can prevent inversion of spins in the myocardium, giving rise to hyper-intensity artifacts [2,3]. Recently, we demonstrated that a wideband (3.8 kHz) inversion pulse can overcome this off-resonance effect and invert off-resonant spins sufficiently, thereby eliminating hyper-intensity artifacts [2,3]. We modified the TI scout sequence by replacing the conventional inversion pulse (bandwidth: 1.1 kHz) with a wideband inversion pulse (spectral bandwidth: 3.8 kHz). We also replaced the TrueFISP readout with a FLASH readout. The modified sequence was tested on a group of T1 phantoms and one ICD patient. Results Figure 1 shows phantom images of the modified TI scout sequence. The phantom results show that when a flip angle of 5° is used, the modified TI scout sequence can be used to determine the optimal TI quite accurately. At higher flip angles, larger TIs may be underestimated due to signal burn-off. Figure 2 demonstrates the banding artifacts produced when the standard TI scout sequence is used in an ICD patient. The modified TI scout sequence is able to remove these artifacts completely and allow effective determination of the optimal TI. Conclusions We have modified the TI scout sequence by implementing a wideband inversion pulse and FLASH readout. Phantom results and patient studies demonstrate that the modified sequence has none of the image artifacts that occur with the standard sequence, and produces a reliable estimation of inversion time. We expect that the new TI scout sequence will lead to improved application of LGE MRI in patients with implantable cardiac devices. Funding

Intraindividual comparison of T1 relaxation times after gadobutrol and Gd-DTPA administration for cardiac late enhancement imaging

PurposeTo evaluate T1-relaxation times of chronic myocardial infarction (CMI) using gadobutrol and gadopentetate dimeglumine (Gd-DTPA) over time and to determine the optimal imaging window for late enhancement imaging with both contrast agents. Material and methodsTwelve patients with CMI were prospectively included and examined on a 1.5T magnetic resonance (MR) system using relaxivity-adjusted doses of gadobutrol (0.15mmol/kg) and Gd-DTPA (0.2mmol/kg) in random order. T1-relaxation times of remote myocardium (RM), infarcted myocardium (IM), and left ventricular cavity (LVC) were assessed from short-axis TI scout imaging using the Look–Locker approach and compared intraindividually using a Wilcoxon paired signed-rank test (α<0.05). ResultsWithin 3min of contrast agent administration (CA), IM showed significantly lower T1-relaxation times than RM with both contrast agents, indicating beginning cardiac late enhancement. Differences between gadobutrol and Gd-DTPA in T1-relaxation times of IM and RM were statistically not significant through all time points. However, gadobutrol led to significantly higher T1-relaxation times of LVC than Gd-DTPA from 6 to 9min (220±15ms vs. 195±30ms p<0.01) onwards, resulting in a significantly greater ΔT1 of IM to LVC at 9–12min (−20±35ms vs. 0±35ms, p<0.05) and 12–15min (−25±45ms vs. −10±60ms, p<0.05). Using Gd-DTPA, comparable ΔT1 values were reached only after 25–35min. ConclusionThis study indicates good delineation of IM to RM with both contrast agents as early as 3min after administration. However, we found significant differences in T1 relaxation times with greater ΔT1 IM–LVC using 0.15mmol/kg gadobutrol compared to 0.20mmol/kg Gd-DTPA after 9–15min post-CA suggesting earlier differentiability of IM and LVC using gadobutrol.

Diagnostic value of cardiac magnetic resonance for the differential diagnosis of thrombus vs tumor

Background Correct differentiation of a cardiac tumor from a thrombus remains important in clinical practice as the prognosis and therapeutic approach vary substantially. Techniques such as echocardiography may have limitations due to reduced image quality and poor tissue characterization. Cardiac magnetic resonance (CMR) has emerged as a promising tool in this regard; however, studies assessing its value are scarce. Methods We retrospectively analyzed the CMR of patients with a definite cardiac mass. Thrombus was defined as a noninfiltrating mass that fulfilled any of the following criteria: a) adjacent to an akinetic or dyskinetic myocardial segment (typically infarcted) or central catheter, b) located in the atrial appendage in patients with atrial fibrillation, c) experienced a significant reduction in size under anticoagulation therapy, or d) had surgical and/or pathological confirmation. Tumor was defined as a mass that did not fulfill any of the above criteria plus a) was infiltrative, or b) located in the left aspect of the fossa ovalis with typical features of myxoma, or c) had surgical and/or pathological confirmation. Mass features on common CMR sequences (cine image, T1-weighted [T1w] and T2 weighed [T2w] spin-echo, contrast first pass perfusion, post-contrast TI scout, and late gadolinium enhancement (LGE)) were analyzed. Categorical data were summarized as frequencies and percentages, and continuous variables were expressed as mean ± SD. Difference in CMR characteristics between thrombi and tumors were compared by Chi-square, Fisher’ so r t-student tests as appropriate. A two sided P-value of <0.05 was considered significant. Results

Regional variations in T1 in the healthy left ventricle

Methods We assessed regional variation in T1 in the septum of patients with no left ventricular pathology. An adjunct phantom study was performed to assess whether location closer to a surface coil affected T1. Main study: Retrospective. 40 patients whose cardiac MRI showed no left ventricular disease were identified. Patients were imaged on a Siemens Avanto 1.5T scanner with a 32-channel cardiac coil. All patients had a 4-chamber TI scout approximately 12 minutes post injection of 0.14 mmol/kg gad-BOPTA. QMASS (Medis, NL) was used to generate T1 maps and measurements obtained in small regions of interest throughout the septum. Basal, middle third, and apical regions were evaluated in all 40; a subset of 15 patients

Comparison between magnitude reconstruction and phase-sensitive mR imaging in the detection of myocardial infarction

ObjectiveTo determine the deference between phase sensitive magnetic resonance (MR) imaging and magnitude reconstruction to detect infracted myocardium. MethodsTwenty patients (16 men; 4 women; mean age, 56 years). experienced Q-wave myocardial infarction 2 weeks earlier were examined with a 3.0-T MR system 10 minutes after administration of 0.1 mmol/kg body weight gadobenate dimeglumine. To determine the optimal TI, a TI scout sequence was used. A segmented 2D IR true fast imaging with steady-state precession (trueFISP) sequence that produces both phase-sensitive and magnitude-reconstructed images were used at TI values of 200-600 msec (TI values were varied in 100-msec steps) and at optimal TI (mean value, 330 msec). Contrast-noise ratios (CNRs) of normal and infarcted myocardium and the area of infarcted myocardium were determined. Two-tailed unpaired sample Student t test was used to compare CNRs, and area of infarction. ResultsMMean CNR phase-sensitive and magnitude-reconstructed images at optimal TI (mean value, 330 msec) were 6.2, and 6.1, respectively. For a TI of 200 msec, CNR values were 5.5, and 4.2, respectively; for TI of 600 msec, CNR values were 5.8 and 4.3, respectively. Area of infarcted myocardium was underestimated on magnitude-reconstruction images (P = 0.002-0.03) for short TI values (ie., 200 msec) but not on phase sensitive reconstructed when compared with IR tureFISP images obtained at optimal TI. ConclusionsLPhase-sensitive image reconstruction results in reduced need for precise choice of TI and more consistent image quality.

Utility of the inversion scout sequence (TI scout) in diagnosing myocardial amyloid infiltration

To evaluate the utility of inversion scout (TI-scout) obtained during cardiac magnetic resonance imaging (CMR) in diagnosing myocardial amyloid infiltration. A retrospective analysis of CMR exams in 39 patients (24 males, age range 29-77 years) was performed. Imaging was performed on a 1.5T system, and included steady state cine, post contrast TI-scout and delayed enhancement images. Evaluations included studies in 13 patients with myocardial amyloidosis and 26 patients without myocardial amyloidosis. To characterize abnormal nulling, the time to myocardial nulling on the TI scout was compared to the null times of the blood pool and spleen for each scan. The sensitivity and specificity of different tissue nulling abnormalities for myocardial amyloidosis were computed. The null times of tissues in 18/26 (69%) patients in the non-amyloid group followed a consistent order with the blood pool null time preceding the myocardial nulling which was equal to that of splenic nulling (Type 1 pattern). This order differed in all 13 patients with myocardial amyloidosis described as three non-mutually exclusive nulling categories: 10 patients had myocardial null time preceding or coincident with blood pool (Type 2 pattern); in 11 patients myocardial null time was non-coincident with splenic nulling (Type 3 pattern); and in 8 patients myocardial null time was non-coincident with both blood pool AND splenic nulling (Type 4 pattern). While no patient exhibited Type 4 nulling pattern in the non-amyloid group, 1/26 patient had a Type 2 and 7/26 patients had a Type 3 nulling pattern. A sensitivity of 100% was obtained when either Type 2 OR Type 3 nulling was present while a specificity of 100% was obtained when both Type 2 AND Type 3 nulling were present together (Type 4 pattern). Our study demonstrates that the pattern of nulling on the TI scout sequence CMR has potential diagnostic utility for the presence of myocardial amyloidosis. The temporal pattern of myocardial, blood pool and splenic nulling needs to be carefully evaluated on the TI scout sequence and could prove useful in other infiltrative cardiomyopathies.

Updates in Cardiac Amyloidosis: A Review

Systemic amyloidosis is a relatively rare multisystem disease caused by the deposition of misfolded protein in various tissues and organs. It may present to almost any specialty, and diagnosis is frequently delayed.[1][1] Cardiac involvement is a leading cause of morbidity and mortality, especially

Alström Syndrome: Cardiac Magnetic Resonance findings

BackgroundAlström Syndrome (ALMS) is an extremely rare multiorgan disease caused by mutations in ALMS1. Dilated cardiomyopathy (DCM) is a common finding but only one series has been investigated by Cardiac Magnetic Resonance (CMR). MethodsEight genetically proven ALMS patients (ages 11–41) underwent CMR performed by standard cine steady state, T1, T2 and late gadolinium enhancement (LGE) sequences.Ejection fraction (EF), Diastolic Volume (EDV) and Systolic Volume normalized for body surface area (ESV), and mass indices were determined, as well as EDV/Mass ratio, an index expressing the adequacy of cardiac mass to heart volume. Regional fibrosis was assessed by LGE; diffuse fibrosis was measured by a TI scout sequence acquired at 5, 10 and 15min after gadolinium by comparing inversion time values (TI) at null time in ALMS and control group. ResultsIn one patient severe DCM was present with diffuse LGE. There were seven cases without clinical DCM. In these patients, EF was at lower normal limits or slightly reduced and ESV index increased; six patients had decreased mass index and EDV/Mass ratio. Mild regional non ischemic fibrosis was detected by LGE in three cases; diffuse fibrosis was observed in all cases, as demonstrated by shorter TI values in ALMS in comparison with controls (5min: 152±12 vs 186±16, p 0.0002; 10min: 175±8 vs 204±18, p 0.0012; 15min: 193±9 vs 224±16, p 0.0002). ConclusionsCardiac involvement in ALMS is characterized by progressive DCM, associated with systolic dysfunction, myocardial fibrosis and reduced myocardial mass.

Myocardial T1 measurement: comparison of modified Look-Locker inversion recovery (MOLLI) and TI scout in the Multi-ethnic Study of Atherosclerosis (MESA)

Summary Different cardiac MR acquisition sequences have been used to obtain myocardial T1 values. Among which available techniques, MOdified Look-Locker Inversionrecovery (MOLLI) and inversion-recovery TrueFisp (TI scout) sequences are widely used in research and clinical settings. We compared myocardial T1 values derived from both MOLLI and TI scout techniques in the post gadolinium delayed enhancement experiments in a large-scale multicenter study (Multi-Ethnic Study of Atherosclerosis, MESA). Our results indicated that myocardial T1 measurement using MOLLI is reproducible with TI scout. Background Different cardiac MR sequences have been used to obtain myocardial T1 values. The multipoint approach, as first described by Look and Locker (TI scout), has been widely used for estimation of the optimal TI in the assessment of myocardial delayed enhancement (DE). The other technique, MOdified Look-Locker Inversionrecovery (MOLLI) allows a rapid and highly reproducible T1 mapping of the heart with high levels of intra and inter-observer agreement. Even though each technique claims its own merit and reliability, the post-contrast myocardial T1 value is confounded by various factors that inhibit comparison of myocardial T1 between patients. In this work, we compared myocardial T1 values derived from both MOLLI and TI scout

Estimation of Gadolinium-induced T1-shortening with Measurement of Simple Signal Intensity Ratio between the Cochlea and Brain Parenchyma on 3D-FLAIR: Correlation with T1 Measurement by TI Scout Sequence

T(1)-shortening of labyrinthine fluid on 3-dimensional fluid-attenuated inversion recovery (3D-FLAIR) has been reported in many inner ear disorders. Although semi-quantitative assessment by simple signal intensity ratio between cochlear fluid and brain tissue has been tried, its feasibility using a multi-channel phased-array head coil with an inherently inhomogenous sensitivity distribution has not been fully evaluated. We evaluated the feasibility of measuring simple signal intensity ratio by correlating rapid T(1) measurements using an inversion time (TI) scout sequence. We evaluated 10 patients with Meniere's disease and 4 patients with sudden deafness. Nine of the patients with Meniere's disease received a unilateral intratympanic injection of Gd-DTPA; the tenth patient received bilateral injections. The 4 patients with sudden deafness received a double-dose intravenous injection. Magnetic resonance (MR) images were obtained 24 hours after intratympanic injections and 4 hours after intravenous injections at 3 tesla using a 32-channel head coil. We measured the ratio (CM ratio) between the signal intensity of the perilymph in the cochlea (C) and that of the medulla oblongata (M) and correlated it with the null-point inversion time (TI(null)) obtained with the TI scout sequence. The TI scout consisted of 85 images obtained with TI values between 132.5 and 3087.5 ms at increments of 37.5 ms. The correlation coefficient between TI(null) and the natural logarithm of the CM ratio was -0.88 (P<0.01). There was significant negative linear correlation. Measurement of the simple signal intensity ratio between the cochlea and the medulla can be used for semi-quantitative analysis of 3D-FLAIR. The results of this study may facilitate clinical research of inner-ear disease using 3D-FLAIR.

Phase-Sensitive Inversion Recovery Single-Shot Balanced Steady-State Free Precession for Detection of Myocardial Infarction During a Single Breathhold

We sought to show that phase-sensitive detection and a single-shot technique allow imaging of the heart for detection of myocardial infarction during a single breathhold without adaptation of the inversion time. Thirty-five patients at 2 weeks to 3 months after Q-wave myocardial infarction were examined on a 1.5-T MR system 10 minutes after the administration of a double-dose extravascular contrast agent. In order to determine the optimal inversion recovery time (TI), a TI scout sequence was performed. An IR-turboFlash sequence with optimized TI was used as standard of reference. A phase-sensitive inversion recovery (PSIR) single-shot TrueFISP sequence, which allows imaging of nine slices during one breathhold (TR/TE/FA/BW: 2.2 ms/1.1 ms/60 degrees , 8 degrees /1220 Hz/Px) was used with a nominal TI of 200 ms. Spatial resolution was identical for both techniques: 1.3 mm x 1.8 mm x 8 mm. Infarct volumes, area of infarction on a selected slice, and scan time for imaging delayed contrast enhancement (DCE) were compared. The mean values for the time of imaging DCE were 10 minutes 43 seconds for the IR turboFLASH and 17 seconds (P<.001) for the PSIR single-shot TrueFISP sequence. No significant difference was found for the mean values of the infarct volumes with 18.7 ml (IR turboFLASH) and 17.3 ml (PSIR single-shot TrueFISP). The values for the correlation coefficients of the infarct volumes and infarct areas of the two different techniques were r=0.95 (P<.004) and r=0.97 (P<.002). The regression equations were y=0.76+0.92*x and y=0.07+0.93*x, respectively. PSIR single-shot TrueFISP allows for accurate identification of myocardial infarction during a single breathhold with reduction of scan time by a factor of 38.

Phase-Sensitive Inversion-Recovery MR Imaging in the Detection of Myocardial Infarction

To prospectively determine if phase-sensitive inversion-recovery (IR) magnetic resonance (MR) imaging eliminates the need to find the precise inversion time (TI) to null the signal of normal myocardium to achieve high contrast between infarcted and normal myocardium. Informed consent was obtained from each patient for this prospective MR imaging research study, which was approved by the institutional review board. Twenty patients (16 men; four women; mean age, 56 years +/- 12.3) who experienced Q-wave myocardial infarction 2 weeks earlier were examined with a 1.5-T MR system 10 minutes after administration of 0.1 mmol per kilogram of body weight gadobenate dimeglumine. To determine the optimal TI, a TI scout sequence was used. A segmented two-dimensional IR turbo fast low-angle shot (FLASH) sequence and a segmented two-dimensional IR true fast imaging with steady-state precession (FISP) sequence that produces both phase-sensitive and magnitude-reconstructed images were used at TI values of 200-600 msec (TI values were varied in 100-msec steps) and at optimal TI (mean value, 330 msec). Contrast-to-noise ratios (CNRs) of normal and infarcted myocardium and the area of infarcted myocardium were determined. Magnitude-reconstructed IR turbo FLASH images were compared with magnitude-reconstructed and phase-sensitive IR true FISP images. Two-tailed unpaired sample Student t test was used to compare CNRs, and two-tailed paired-sample Student t test was used to compare area of infarction. Mean CNR of images acquired with IR turbo FLASH and IR true FISP (phase-sensitive and magnitude-reconstructed images) at optimal TI (mean value, 330 msec) were 6.6, 6.2, and 6.1, respectively. For a TI of 200 msec, CNR values were -4.3, -4.0, and 7.2, respectively; for TI of 600 msec, CNR values were 3.1, 3.3, and 4.3, respectively. Area of infarcted myocardium was underestimated on magnitude-reconstruction images (P = .002-.03) for short TI values (ie, 200 msec) for both sequences and for a TI of 300 msec for IR true FISP but not on phase-sensitive reconstructed IR true FISP images when compared with IR turbo FLASH images obtained at optimal TI. Phase-sensitive image reconstruction results in reduced need for precise choice of TI and more consistent image quality.