MRI works by aligning the spins of hydrogen nuclei with a strong magnetic field. When these polarized protons are hit by an RF pulse they emit energy, which is then picked up by an RF coil in the MRI scanner. To generate a strong enough signal for good-quality imaging with the 1.5 T or 3 T magnets used in clinical scanners, there need to be lots of protons (in the form of hydrogen nuclei) present. "Standard proton MRI detects the signal from protons in molecules of water or fat within the body, giving us good pictures of soft tissue," explained Jim Wild, leader of the Sheffield hyperpolarized MRI group.

When it comes to looking at the lungs, however, standard MRI has limitations because the lung tissue density is low, so the magnetic field in the lungs is inhomogeneous. What's needed is a gas that can act as a contrast agent and which can be safely inhaled by patients. According to Wild, helium-3 fits the bill perfectly: "Helium-3 has a nuclear spin of one half, so magnetically it looks rather like a proton and we can do MRI with it," he told medicalphysicsweb.

In order to obtain high-resolution images using helium-3, however, the researchers need to hyperpolarize the gas before it is inhaled by the patient. By using a technique called laser optical pumping, they can align the spins of the helium nuclei to a much higher degree than can be achieved using just the MRI magnets. This results in a significantly stronger RF signal and hence much better spatial resolution.

While helium-3 MRI produces great images of lung structure, its key strength lies in the information it can provide about lung function. The helium only reaches the parts of the lungs that are ventilated, so comparing these images with anatomical images obtained by CT or standard MRI can show the extent of breathing impairment in patients with lung tumours or respiratory disease.

"Helium is insoluble, so once inhaled it stays in the lungs and gives us very high signal-to-noise images of ventilation and lung function," Wild noted. "There are big potential applications for assessing any obstructive airway disease, like asthma, chronic obstructive pulmonary disease or cystic fibrosis."

Another area in which helium-3 MRI could offer significant benefits is planning IMRT treatments for lung-cancer patients. "Adding extra functional information from our images to the CT planning images allows us to more sensibly plan the radiotherapy to avoid side-effects like radiation pneumonitis," said Wild. This inflammatory condition results from the irradiation of healthy lung tissue. Knowing which areas of the lungs are still working well would allow oncologists to avoid them and hence reduce the risk of collateral damage.

Hyperpolarized helium-3 MR images alone do not provide enough anatomical information for treatment planning, so they need to be registered to CT images. This has never been tried in vivo before, but in the International Journal of Radiation Oncology, Biology, Physics, group member Rob Ireland reports that it can be done using commercially available rigid registration tools.

What's more, Ireland (together with colleagues at Weston Park Hospital, Sheffield) notes that plans calculated based on helium-3 MRI information were better than those produced for the same patients without using this data. The dose to healthy lung tissue was reduced, while the dose to the tumour was maintained at a high level. This is good news for patients, because unlike SPECT - the most commonly used functional lung-imaging technique – helium-3 MRI doesn't subject them to additional ionizing radiation.

Wild thinks that it will take a while for the technique to make it into widespread clinical use though. "We've got a novel technique here, but what's limiting us is the availability of the technology," he said. "At the moment it's constrained to research laboratories like our own. It requires clinicians to take more interest in the technique to get it into the wider arena."