With this aim, researchers at the University of North Carolina at Chapel Hill (UNC) are investigating the use of magnetomotive ultrasound (MMUS) to image platelets labelled with superparamagnetic iron oxide (SPIO) nanoparticles. The team has developed a non-invasive, portable MMUS imaging system and used it to identify clots formed from SPIO-labelled platelets (Phys. Med. Biol. 58 7277).

In the body, platelets respond to endothelial disruption by forming a platelet-rich thrombus that inhibits blood flow and can, in some instances, lead to stroke, heart attack or pulmonary embolism. "The current clinical standard for detecting platelet-rich thrombi is X-ray angiography, and emerging strategies are being developed using MRI, CT and PET," explained researcher Ava Pope. "Though these imaging modalities are helpful in diagnosing blood related diseases, ultrasonic imaging offers lower cost, portability, and avoids the radiation hazard associated with X-rays."

System design

Magnetomotive imaging works by detecting the magnetically-induced motion of echogenic materials coupled to (non-echogenic) magnetically labelled agents such as SPIOs. The amplitude of motion depends upon the particle distribution and magnetization, the strength and gradient of the applied magnetic field, and the mechanical properties of the surrounding medium.

Platelets have been shown to readily uptake SPIO contrast agents, and the UNC team previously demonstrated optical coherence tomography (OCT)-based magnetomotive imaging of such platelets. OCT, however, offers a limited penetration depth of 1–2 mm. By translating the technique to an ultrasound platform, the researchers have extended the imaging penetration depth, enabling in vivo imaging of peripheral vasculature.

"The clinical ultrasound machine used for these studies is capable of imaging at depths of up to 9 cm," said Pope. "However, the penetration depth of MMUS also depends on the depth penetration of the magnetic field. Our current electromagnets can induce detectable platelet displacements at depths up to approximately 3 cm, but a different magnet design could potentially provide displacements up to the maximum penetration depth of the ultrasound."

The MMUS system incorporates an ultrasound transducer mounted between a pair of custom electromagnets, which provide a temporally modulated magnetic field gradient during imaging. This design allows for open air imaging of arbitrarily thick samples, making it readily translatable to large animal and human studies. The apparatus is mounted on a vibration isolation table to reduce ambient mechanical noise.

MMUS data are collected with the magnets in anti-parallel configuration (which produced the maximum force) with an average field of 0.05 T and gradient of 0.09 T/m within the imaging area. Each MMUS data set consisted of two 8 s sets of ultrasound data, taken with the magnetic field on and off, enabling subtraction of ambient noise during processing.

The researchers employed a frequency- and phase-locked motion detection algorithm that is robust against ambient noise and may avoid interference from patient motion. This method rejects noise outside a narrow frequency passband, and also rejects background motion that is out-of-phase with the driving magnetic field. "As such, the system would reject any patient motion that is not locked in frequency and phase with the driving magnetic field," added Pope.

Phantom tests

To test the MMUS approach, Pope and colleagues created homogeneous phantoms with tissue-equivalent mechanical and acoustic properties by combining agar with varying concentrations of SPIO (Fe3O4 nanoparticles). They also made control phantoms without nanoparticles.

Imaging a 0.3 mg Fe/ml phantom and a control phantom showed that the MMUS contrast was only apparent in the SPIO-containing phantom, and only in the presence of a modulating magnetic field. Inhomogeneity seen in the MMUS image was attributed to corresponding inhomogeneity of the magnetic field gradient.

Results showed that the system could detect magnetically induced motion at SPIO concentrations as low as 0.091 mg Fe/ml, and that the signal-to-noise ratio (SNR) monotonically increased as the concentration ranged from 0.091 to 0.91 mg Fe/ml.

To illustrate the potential of MMUS as a clinical tool for thrombus detection, the researchers also created tissue phantoms containing inclusions of either SPIO-laden agar or SPIO-labelled platelet-rich clots.

MMUS imaging of the phantoms revealed contrast only at the positions of the SPIO-laden inclusions. No contrast was seen at other positions or from a control phantom without SPIO. Comparing 3D scans of the phantoms showed a larger SNR for the SPIO-labelled clot inclusion (6.35 dB) than the SPIO agar inclusion (1.27 dB). The authors attribute this mainly to differing iron concentration in the phantoms: 0.91 mg Fe/ml in the agar inclusion and 4.7 mg Fe/ml (140 fg of Fe per platelet) in the clot.

The authors concluded that their MMUS system is able to locate SPIO-labelled platelet-rich clots within 3D tissue phantom volumes. "Our next step is to validate this technique in vivo and learn about the challenges associated with physiological motion," Pope told medicalphysicsweb. "We will be studying in vivo animal models of atherosclerosis and of vascular injury to understand the efficacy of the SPIO platelets for targeting these conditions."

Related stories

• Particle-loaded blood cells enhance MPI
• MRI with USPIO particles finds more lesions
• Super-thin probe finds fluorescing plaques
• Photoacoustic imaging reveals DVT age