Bone matrix imaged in vivo by water- and fat-suppressed proton projection MRI (WASPI) of animal and human subjects

Abstract

Purpose: To demonstrate water- and fat-suppressed proton projection MRI (WASPI) in a clinical scanner to visualize the solid bone matrix in animal and human subjects.

Materials and methods: Pig bone specimens and polymer pellets were used to optimize the WASPI method in terms of soft-tissue suppression, image resolution, signal-to-noise ratio, and scan time on a 3T MRI scanner. The ankles of healthy 2-3-month-old live Yorkshire pigs were scanned with the optimized method. The method was also applied to the wrists of six healthy adult human volunteers to demonstrate the feasibility of the WASPI method in human subjects. A transmit/receive coil built with proton-free materials was utilized to produce a strong B(1) field. A fast transmit/receive switch was developed to reduce the long receiver dead time that would otherwise obscure the signals.

Results: Clear 3D WASPI images of pig ankles and human wrists, showing only the solid bone matrix and other tissues with high solid content (eg, tendons), with a spatial resolution of 2.0 mm in all three dimensions were obtained in as briefly as 12 minutes.

Conclusion: WASPI of the solid matrix of bone in humans and animals in vivo is feasible.

Figure 1. Proton spectrum of bone matrix and polyethylene, and SMRI of a solid polyethylene cylindrical phantom

a. Water and fat suppressed proton spectrum of a pig bone specimen acquired with a home made transmit/receive (T/R) coil at 3T on a Siemens Trio scanner. The excitation pulse was 10 μs (10°). The full width at half height of the spectrum was 1.5–2 kHz. The low-power, long-duration water- and fat-suppression pulses cause some loss in the solid bone matrix signal and result in incompletely suppressed fluid signal, which is manifested as dips in the matrix spectrum. b. Single pulse proton spectrum of a polyethylene phantom acquired with a 10 μs (10°) pulse. The full width at half height of the spectrum was 2.0–2.4 kHz. c, d. Transverse (c) and longitudinal (d) view of SMRI images of the polyethylene phantom, acquired with the body coil. The excitation pulse was 30 μs (11°). Using the body coil results in insufficient B₁ field for excitation, and low SNR.

Figure 2

A T/R RF birdcage coil built on an acrylic tube. a. SMRI image (FOV: 300 mm) of a 10 mm diameter tube of water at the center of the coil. The acrylic tube and the polyethylene dielectric and jacket of the cable are clearly visible because of their proton content. The image was acquired with a conventional actively driven PIN diode T/R switch. A smear of background signal is spread across the FOV due to the loss of data in the receiver dead time. b. Photograph of the coil.

Figure 3

A T/R RF birdcage coil built on a Teflon tube. a. Photograph of the coil with a solid polyethylene cylindrical phantom (diameter 90 mm; length 183 mm) inside it. b. Transverse view of the WASPI image of the polyethylene phantom. The acquisition parameters were: receiver dead time 10 μs, pulse length 10 μs, FOV 120 mm, number of projections 8148, TR 65 ms, total acquisition time 18 min. c. Coronal view of the WASPI image of the polyethylene phantom. The images of the polyethylene phantom serve as a presentation of the B₁ field distribution. The extent of the region of strong B₁ field is about 75 mm in the z direction and 90 mm in diameter. Within the desired cylindrical volume (40 mm along the z axis and 70 mm in diameter), the variation of B₁ is about 5%.

Figure 4. The recovery times of the PIN diode and crossed-diodes transmit/receive (T/R) switches and their effects on WASPI images

a. The magnitude response curves of the PIN diode (solid line) and crossed-diodes (dashed line) T/R switches. See the text for details of the experimental procedure. b. WASPI image of a pig leg specimen acquired with a PIN diode T/R switch and reconstructed by assuming the recovery time was 10 μs and two data points were lost. The actual receiver dead time was much longer. Severe background artifacts are present and the polymer pellet phantom was not visible. c. In vivo WASPI of a pig ankle and a polymer pellet acquired with a crossed-diode T/R switch. This image shows the solid-only character of the WASPI image: the solid component of the bone and the polymer pellet are bright while the soft tissues are not visible. d. In vivo spin echo image of the same pig ankle. This image shows the character of liquid state MRI: the solid bone is dark, the polymer pellet is not visible, and muscle and marrow are bright.

Figure 5

Non-suppressed SMRI and WASPI images of a bottle of saline and a bottle of corn oil. All images are displayed with the same window and center settings so that image brightness may be directly compared. In the water image, the water frequency was set at zero offset frequency; in the corn oil image, the fat frequency was set at zero offset and the water frequency at 3.5 ppm upfield. In both cases, the water and fat signals in the WASPI images were suppressed to be less than 5% of their intensities in the non-suppressed SMRI images. a–d: SMRI images (a,b: saline; c,d: oil). e–h: WASPI images (e,f: saline; g,h: oil).

Figure 6. WASPI images of polymer pellets separated by 3 mm thick Teflon disks to demonstrate the spatial resolution

a. A coronal view of a WASPI image acquired with FOV 120 mm, gradient 19.57 mT/m, number of projections 8148. b. A coronal view of a WASPI image acquired with the same parameters except that the FOV was 100 mm and gradient strength was 23.48 mT/m.

Figure 7. In vivo WASPI and non-suppressed SMRI images of a 57-year-old male volunteer’s left wrist

a, b, c. Transverse, coronal, and sagittal slices from the 3D WASPI image data set. The SNR of the cortical bone is above 40 and that of the trabecular bone is above 10. d, e, f. Corresponding slices from the non-suppressed SMRI image data set.

Figure 8. In vivo images of the right wrist of a 24-year-old male volunteer comparing different FOVs

a. Transverse slice from a WASPI 3D image, FOV 100 mm, number of projections 5216. b. Same slice from a WASPI 3D image, FOV 80 mm, number of projections 8148. c. Same slice from a conventional spin-echo image, FOV 80 mm, slice thickness 2.5 mm.