(CSI) Chemical shift imaging is an extension of MR spectroscopy, allowing metabolite information to be measured in an extended region and to add the chemical analysis of body tissues to the potential clinical utility of Magnetic Resonance. The spatial location is phase encoded and a spectrum is recorded at each phase encoding step to allow the spectra acquisition in a number of volumes covering the whole sample. CSI provides mapping of chemical shifts, analog to individual spectral lines or groups of lines.
Spatial resolution can be in one, two or three dimensions, but with long acquisition times od full 3D CSI. Commonly a slice-selected 2D acquisition is used. The chemical composition of each voxel is represented by spectra, or as an image in which the signal intensity depends on the concentration of an individual metabolite. Alternatively frequency-selective pulses excite only a single spectral component.
There are several methods of performing chemical shift imaging, e.g. the inversion recovery method, chemical shift selective imaging sequence, chemical shift insensitive slice selective RF pulse, the saturation method, spatial and chemical shift encoded excitation and quantitative chemical shift imaging.

See also Magnetic Resonance Spectroscopy.

(DIR or DIRT1) Double inversion recovery T1 measurement is a T1 weighted black blood MRA sequence in which the signal from blood is suppressed. The inversion time to suppress blood is described as the duration between the initial inversion pulse and time point that the longitudinal magnetization of blood reaches the zero point. The readout starts at the blood suppression inversion time (BSP TI) and blood in the imaging slice gives no signal. This inversion time is around 650 ms with a 60 beat per minute heart rate at 1.5 T.
The TI can be decreased by using a wider receive bandwidth, shorter echo train length and/or narrow trigger window. Wide bandwidth also decreases the blurring caused by long echo trains at the expense of signal to noise ratio. In case of in plane or slow flow the suppression of the signal from blood may be incomplete. With increased TE or change of the image plane the blood suppression can be improved.
Double inversion recovery is a breath hold technique with one image per acquisition used in cardiovascular imaging. The patient is instructed to hold the breath in expiration (if not possible also inspiration can be taken), so that the end diastolic volume in the cardiac chambers would be the same during entire scanning. DIR provides fine details of the boundary between the lumen and the wall of the cardiac chambers and main vascular and heart structures, pericardium, and mediastinal tissues.

(FAIR) In this sequence 2 inversion recovery images are acquired, one with a nonselective and the other with a slice selective inversion pulse. The z-magnetization in the first sequence is independent of flow. Inflowing spins give z-magnetization from second pulse. A major signal loss in FAIR is the T1 relaxation of tagged blood in transit to the imaging slice. Sharper edges of the inversion pulse give narrow spacing between the inversion edge and the 1st slice because reduced transit time gives lower T1 relaxation induced signal loss. The difference of the images in a consequence contains information proportional to flow (blood partition coefficient). Standard adiabatic inversion RF pulse does not have good slice-profile, because of power/SAR limitation. A c-shaped frequency offset corrected inversion (FOCI) RF pulse can help to increase the signal.
Perfusion imaging, e.g. myocardial, using tissue water as endogenous contrast is suggested.

(IR FGR) A gradient echo sequence with an inversion pulse.

See Gradient Recalled Echo Sequence, Inversion Recovery and Steady State Free Precession.