Improved compressed sensing reconstruction for [Formula: see text]F magnetic resonance imaging.

OBJECTIVE: In magnetic resonance imaging (MRI), compressed sensing (CS) enables the reconstruction of undersampled sparse data sets. Thus, partial acquisition of the underlying k-space data is sufficient, which significantly reduces measurement time. While 19F MRI data sets are spatially sparse, they often suffer from low SNR. This can lead to artifacts in CS reconstructions that reduce the image quality. We present a method to improve the image quality of undersampled, reconstructed CS data sets. MATERIALS AND METHODS: Two resampling strategies in combination with CS reconstructions are presented. Numerical simulations are performed for low-SNR spatially sparse data obtained from 19F chemical-shift imaging measurements. Different parameter settings for undersampling factors and SNR values are tested and the error is quantified in terms of the root-mean-square error. RESULTS: An improvement in overall image quality compared to conventional CS reconstructions was observed for both strategies. Specifically spike artifacts in the background were suppressed, while the changes in signal pixels remained small. DISCUSSION: The proposed methods improve the quality of CS reconstructions. Furthermore, because resampling is applied during post-processing, no additional measurement time is required. This allows easy incorporation into existing protocols and application to already measured data.

Monitoring of monocyte recruitment in reperfused myocardial infarction with intramyocardial hemorrhage and microvascular obstruction by combined fluorine 19 and proton cardiac magnetic resonance imaging.

BACKGROUND: Monocytes and macrophages are indispensable in the healing process after myocardial infarction (MI); however, the spatiotemporal distribution of monocyte infiltration and its correlation to prognostic indicators of reperfused MI have not been well described. METHODS AND RESULTS: With combined fluorine 19/proton ((1)H) magnetic resonance imaging, we noninvasively visualized the spatiotemporal recruitment of monocytes in vivo in a rat model of reperfused MI. Blood monocytes were labeled by intravenous injection of (19)F-perfluorocarbon emulsion 1 day after MI. The distribution patterns of monocyte infiltration were correlated to the presence of microvascular obstruction (MVO) and intramyocardial hemorrhage. In vivo, (19)F/(1)H magnetic resonance imaging performed in series revealed that monocyte infiltration was spatially inhomogeneous in reperfused MI areas. In the absence of MVO, monocyte infiltration was more intense in MI regions with serious ischemia-reperfusion injuries, indicated by severe intramyocardial hemorrhage; however, monocyte recruitment was significantly impaired in MVO areas accompanied by severe intramyocardial hemorrhage. Compared with MI with isolated intramyocardial hemorrhage, MI with MVO resulted in significantly worse pump function of the left ventricle 28 days after MI. CONCLUSIONS: Monocyte recruitment was inhomogeneous in reperfused MI tissue. It was highly reduced in MVO areas defined by magnetic resonance imaging. The impaired monocyte infiltration in MVO regions could be related to delayed healing and worse functional outcomes in the long term. Therefore, monocyte recruitment in MI with MVO could be a potential diagnostic and therapeutic target that could be monitored noninvasively and longitudinally by (19)F/(1)H magnetic resonance imaging in vivo.

Compressed sensing for reduction of noise and artefacts in direct PET image reconstruction.

AIM: Image reconstruction in positron emission tomography (PET) can be performed using either direct or iterative methods. Direct reconstruction methods need a short reconstruction time. However, for data containing few counts, they often result in poor visual images with high noise and reconstruction artefacts. Iterative reconstruction methods such as ordered subset expectation maximization (OSEM) can lead to overestimation of activity in cold regions distorting quantitative analysis. The present work investigates the possibilities to reduce noise and reconstruction artefacts of direct reconstruction methods using compressed sensing (CS). MATERIALS AND METHODS: Raw data are generated either using Monte Carlo simulations using GATE or are taken from PET measurements with a Siemens Inveon small-animal PET scanner. The fully sampled dataset was reconstructed using filtered backprojection (FBP) and reduced in Fourier space by multiplication with an incoherently undersampled sampling pattern, followed by an additional reconstruction with CS. Different sampling patterns are used and an average of the reconstructions is taken. The images are compared to the results of an OSEM reconstruction and quantified using signal-to-noise ratio (SNR). RESULTS: The application of the proposed CS post-processing technique clearly improves the image contrast. Dependent on the undersampling factor, noise and artefacts are reduced resulting in an SNR that is increased up to 3.4-fold. For short acquisition times with low count statistics the SNR of the CS reconstructed image exceeds the SNR of the OSEM reconstruction. CONCLUSION: Especially for low count data, the proposed CS-based post-processing method applied to FBP reconstructed PET images enhances the image quality significantly.

In vivo imaging of stepwise vessel occlusion in cerebral photothrombosis of mice by 19F MRI.

BACKGROUND: (19)F magnetic resonance imaging (MRI) was recently introduced as a promising technique for in vivo cell tracking. In the present study we compared (19)F MRI with iron-enhanced MRI in mice with photothrombosis (PT) at 7 Tesla. PT represents a model of focal cerebral ischemia exhibiting acute vessel occlusion and delayed neuroinflammation. METHODS/PRINCIPAL FINDINGS: Perfluorocarbons (PFC) or superparamagnetic iron oxide particles (SPIO) were injected intravenously at different time points after photothrombotic infarction. While administration of PFC directly after PT induction led to a strong (19)F signal throughout the entire lesion, two hours delayed application resulted in a rim-like (19)F signal at the outer edge of the lesion. These findings closely resembled the distribution of signal loss on T2-weighted MRI seen after SPIO injection reflecting intravascular accumulation of iron particles trapped in vessel thrombi as confirmed histologically. By sequential administration of two chemically shifted PFC compounds 0 and 2 hours after illumination the different spatial distribution of the (19)F markers (infarct core/rim) could be visualized in the same animal. When PFC were applied at day 6 the fluorine marker was only detected after long acquisition times ex vivo. SPIO-enhanced MRI showed slight signal loss in vivo which was much more prominent ex vivo indicative for neuroinflammation at this late lesion stage. CONCLUSION: Our study shows that vessel occlusion can be followed in vivo by (19)F and SPIO-enhanced high-field MRI while in vivo imaging of neuroinflammation remains challenging. The timing of contrast agent application was the major determinant of the underlying processes depicted by both imaging techniques. Importantly, sequential application of different PFC compounds allowed depiction of ongoing vessel occlusion from the core to the margin of the ischemic lesions in a single MRI measurement.

Magnetic resonance imaging of tumors colonized with bacterial ferritin-expressing Escherichia coli.

BACKGROUND: Recent studies have shown that human ferritin can be used as a reporter of gene expression for magnetic resonance imaging (MRI). Bacteria also encode three classes of ferritin-type molecules with iron accumulation properties. METHODS AND FINDINGS: Here, we investigated whether these bacterial ferritins can also be used as MRI reporter genes and which of the bacterial ferritins is the most suitable reporter. Bacterial ferritins were overexpressed in probiotic E. coli Nissle 1917. Cultures of these bacteria were analyzed and those generating highest MRI contrast were further investigated in tumor bearing mice. Among members of three classes of bacterial ferritin tested, bacterioferritin showed the most promise as a reporter gene. Although all three proteins accumulated similar amounts of iron when overexpressed individually, bacterioferritin showed the highest contrast change. By site-directed mutagenesis we also show that the heme iron, a unique part of the bacterioferritin molecule, is not critical for MRI contrast change. Tumor-specific induction of bacterioferritin-expression in colonized tumors resulted in contrast changes within the bacteria-colonized tumors. CONCLUSIONS: Our data suggest that colonization and gene expression by live vectors expressing bacterioferritin can be monitored by MRI due to contrast changes.