Lung MRI suffers from low tissue density and severe field inhomogeneity due to the air-tissue interface, resulting in prohibitively short T₂^* relaxation times [1]. However, the reported T₂ relaxation time constants in the lungs are considerably longer [2]. In this study, we demonstrate the utility of a modified under-sampled single-shot turbo spin-echo (TSE) for carbon-13 hyperpolarized MRI of the lungs. The sequence exploits long T₂ relaxation times to obtain high-resolution carbon-13 images. Although a similar approach has been used for HP-MRI of the lungs [3], we aim to investigate the utility of single-shot TSE sequence for chemical shift imaging. First, we measured the T₂ relaxation time of carbon-13 in the lungs using a dual-echo single-shot TSE sequence. Then, we used a higher-resolution single-echo sequence to obtain dynamic high-resolution images. Finally, we demonstrated the applicability of this sequence to obtaining chemical shift images using selective excitation in a phantom.Eight BLAB/c mice (25 ± 2 g) were imaged for the in vivo studies. A venous tail vein catheter was placed and mice were placed in a 9.4T vertical-bore (Bruker Inc.) micro-imaging MRI system. All images were acquired using a 25mm ¹H/¹³C saddle-coil (Bruker Inc.). Respiration was monitored using a respiration cushion. Proton T₂-weighted images were acquired over the lungs using a respiratory-gated multi-slice fast spin-echo (FSE) sequence (TR/TE=570/2.3ms, ETL=4, NA=9, 192x192 voxels). 45µL of [1-¹³C]-pyruvate (Cambridge Isotopes) was polarized using a HyperSense DNP polarizer (Oxford Instruments) to over 15%. The sample was melted using a dissolution buffer (40mM Trizma, 160mM NaOH, 50mM NaCl, 0.1g/L EDTA) at 180°C to yield an isotonic solution of 160mM [1-¹³C]-pyruvate with neutral pH at 37°C. Ten seconds after the dissolution a 350µL aliquot was injected through the tail-vein over 12 seconds. Carbon-13 imaging was performed using a single or dual-echo under-sampled single-shot TSE sequence (partial fourier=1.68) with an outwards centric phase encoding table. For the in-vivo studies, a 1ms SLR excitation/flip-back pulses (α=90°) and 0.5ms adiabatic refocusing pulses (α=180°) were used. Four mice were imaged using the dual-echo version of the sequence (TE₁/TE₂=2.98/50.66ms, 32x32 voxels, BW=40KHz, in-plane FOV=25x25mm² and 10mm slice thickness) for T₂ mapping and optimizing the timing of the high resolution sequence (Figure 1). The other four mice were imaged using a higher resolution version of the sequence (TE=2.3ms, 48x48 voxels, BW=40KHz, in-plane FOV=25x25mm² 3mm slice thickness). Six images were acquired every 5 seconds (Figures 2 and 3). To demonstrate the feasibility of chemical shift imaging a phantom consisting of three 5 mm tubes containing (1) 1.5M [1-¹³C]-glycine, (2) [1-¹³C]-sodium propionate and (3) a mixture of both was used (Figure 4). A modified single-echo single-shot TSE sequence (TE=9.42ms, echo-spacing = 4.11ms, 48x48 voxels, BW = 40KHz, in-plane FOV=25x25mm² and 10mm slice thickness, NA=8) was used. Alternating back-to-back selective excitation was performed using 9ms excitation/flip-back RF pulses (α=90°) and 2ms adiabatic refocusing pulses (α=180°). All images were exported to DICOM format for OsiriX 7.0. T₂ maps were created from the images acquired using the dual-echo sequence using the T₂ fit map pluginFigure 1 shows a T₂ map in the lungs (T₂ = 71 ± 23ms) which suggests the possibility of imaging the lungs with minimal blurring for scan times shorter than 100ms. Figure 2 shows a 48x48 carbon-13 scan with 0.5x0.5mm² in-plane resolution (total scan time = 68ms). The mean SNR in the lungs was 68. Figure 3 shows dynamic imaging in another mouse with 5 seconds temporal resolution. The blurring at (t = 5s) may be due to a heart-beat coinciding with the scan, which can be resolved by using a cardiac-gated scan. Figure 4 shows the chemical shift images of the carbon phantom acquired with the frequency-selective version of sequence. For in vivo applications, the 9ms minimum phase RF pulse provides sufficiently sharp spectral selectivity (4ppm bandwidth at 9.4T) to image pyruvate and lactate in vivo as it can exclude the pyruvate-hydrate and alanine peaks. Each single-shot scan was 121ms, which is short enough to yield high resolution in vivo metabolic images of the lungs. It is notable that longer T₂ relaxation times in tumors and other organs allows for forming much longer echo trains after the excitation, which can be used for high-resolution 3D chemical shift imaging with minimal blurring. We showed the utility of a modified under-sampled single-shot TSE sequence for high resolution imaging of hyperpolarized carbon-13 agents the lungs with excellent signal quality. We further demonstrated the feasibility of extending this method to obtain high resolution chemical shift imaging via selective excitation.