One possibility is to use Cerenkov emission (CE) to perform portal imaging on the CyberKnife. Cerenkov radiation (emitted when charged particles travel faster than the speed of light in a given dielectric medium) is produced with an intensity proportional to the deposited radiation dose. The approach has already been used to measure entrance dose in radiotherapy studies. Now, a UK–US collaboration has demonstrated that CE can also be used for portal imaging from the exit face of an irradiated medium (Phys. Med. Biol. 60 N419).

"Optical imaging of the Cerenkov signal produced by the exiting treatment beam potentially allows imaging of exit dose in scenarios where existing solutions are not applicable, without further use of ionizing radiation," explained Hamid Dehghani from the University of Birmingham. "This method potentially allows visual verification of a patient's internal anatomy within the treatment field during delivery, which may permit confident reduction in treatment volumes and reduced toxicity."

Phantom studies

Dehghani and colleagues, from Dartmouth College and the Queen Elizabeth Hospital, used a CyberKnife system to irradiate a water tank with a 6 MV, 60 mm diameter beam. To assess the resolution of the measured CE they placed a highly attenuating phantom, comprising lead sheets spaced with plywood, between the beam source and the tank. The tank was partially filled with water and set up such that half of the radiation beam travelled through the phantom and water, and half through the phantom and air. An opaque solid water slab at the beam exit face ensured that only CE due to radiation at the exit face was imaged.

The researchers measured the CE induced at the exit face using a gated electron-multiplying-intensified-charged-coupled device. To maximize signal-to-background ratio, they synchronized image acquisition with the beam pulses and background-subtracted radiation-free images. CE images were acquired at 30 frames/s, with each frame encompassing five pulses.

Intensity profiles plots through the acquired CE images showed that objects 3.4 mm apart could easily be detected, with the separation measured as 3.58 or 3.92 mm for radiation travelling through air or water, respectively. Each 2 mm lead sheet was visible, highlighting the potential for higher resolution imaging.

To assess the contrast of the CE at the exit face, the team imaged a number of 28 mm diameter tissue-equivalent rods and calculated the mean signal decrease due to the presence of the rods. A cortical bone rod with an electron density of 1.69 relative to water demonstrated a CE contrast of 15% or 14%, for beams passing through air or water, respectively. Tissues with relative electron density of greater than 4% (with respect to water) demonstrated CE contrast of 8% or greater, demonstrating that CE detected at the exit face is sensitive to the small contrasts seen in biological tissues.

To compare CE-based portal imaging with current technology, the researchers repeated the experiment on an Elekta Synergy linac, acquiring portal images using its iViewGT EPID. They found that EPID-based portal images had higher contrast than CE images for beams travelling through air (19% for the cortical bone rod), but lower contrast (12%) for beams travelling through water.

This higher measured contrast of CE images from radiation passing through water (or tissue) is due to the fact that, unlike EPID-based images, CE images only utilize visible optical photons produced at the exit surface via CE, minimizing the contribution of scattered radiation. This finding highlights an additional benefit of CE portal imaging over EPID-based techniques.

CE in motion

The researchers also demonstrated the ability of CE to track motion, by placing the phantom on a moving platform. Real-time CE video (30 frames/s) showed that the periodic movement (0.25 Hz) could be observed.

Real-time CE of a moving graticule and a tissue equivalent rod.

The researchers concluded that CE based portal imaging could potentially be used for systems such as the CyberKnife that lack alternative options. As well as real-time verification of treatment delivery accuracy, the method could detect small deviations in patient positioning and intra-fraction anatomical movements. The team is also studying the use of CE for tomographic reconstruction of delivered dose distribution within the patient.

"We will continue to investigate these possibilities in phantom studies to allow quantitative verification, followed by internal mapping of the delivered dose distribution using tomographic methods," said Dehghani. "Hopefully, this will lead to exploration in patients."

Related articles in PMB
Real-time Cherenkov emission portal imaging during CyberKnife® radiotherapy
Yiannis Roussakis et al Phys. Med. Biol. 60 N419
The physics of Cerenkov light production during proton therapy
Y Helo et al Phys. Med. Biol. 59 7107
Optical dosimetry of radiotherapy beams using Cherenkov radiation: the relationship between light emission and dose
Adam K Glaser et al Phys. Med. Biol. 59 3789
Imaging Cerenkov emission as a quality assurance tool in electron radiotherapy
Yusuf Helo et al Phys. Med. Biol. 59 1963

Related stories

• Cerenkoscopy monitors breast treatments
• Can Cerenkov light track proton therapy?
• Imaging scheme tracks tumour oxygenation
• Cerenkov light tracks radiation dose