The simple method exploits the Cerenkov effect, and uses just a water tank, a standard commercial CMOS camera and a fluorescent dye dissolved in tap water. As the X-ray beam is incident on the water, Cerenkov light is generated, which excites the dissolved fluorophore. The fluorophore emits a characteristic fluorescent signal that is acquired by the camera. These optical images of Cerenkov-stimulated fluorescence are then equated back to the deposited dose in water.

"Although other methods such as scintillation and gel dosimetry have been proposed with the same goal in mind, this is the first demonstration of using only the fluorescence induced by the inherent Cerenkov radiation to measure the deposited dose," explained Adam Glaser from Dartmouth College's Thayer School of Engineering (Hanover, NH). "We hope that the method will become an important tool for research and potentially clinical applications."

Beam profiling

The technique developed by Glaser and his Dartmouth colleagues has been verified through a series of experiments using a clinical linac from Varian Medical Systems. The first step in the process is to fill a water tank with tap water and dissolve the fluorophore (quinine sulphate) to a concentration of 1.0 g/l. Then, a standard commercial CMOS camera is positioned at a given distance from the water tank, perpendicular to the incident beam, and focused to the beam's isocentre.

When the beam is turned on, a 2D projection image is captured using a 10 s exposure time, and an equivalent image with the beam off is recorded and subtracted to isolate the Cerenkov-excited fluorescence for direct correlation to the deposited dose.

"Each image is immediately downloaded from the camera to a computer and can be viewed in real time," explained Glaser. "Our experiments in this proof-of-concept study show that the strength of the fluorescence signal equates near-linearly to the dose imparted in the water. We believe this is the first demonstration of using Cerenkov light to indirectly determine the spatial distribution of a charged particle's energy deposition within a medium."

To challenge convention

While the results from this study are encouraging, it is important to note that the proposed technique cannot as yet challenge today's ionization chambers without further refinement.

"Although our method is orders of magnitude faster, a similar accuracy must be achieved in order to compete with ionisation chamber measurements," explained Glaser. "An acceptable tolerance would be ±1%, which we currently cannot achieve without further work. Also, the 2D results we present are not immediately useful, i.e., the 2D projection data must be resolved to 3D data by tomography to yield meaning results."

One of the next steps for the Dartmouth team is to tomographically acquire many projection images of the induced light volume from various angles and perform a 3D reconstruction. They also plan to investigate different back projection algorithms to see which gives the optimal result.

"With refinement, our technique could be used to replace ionization chamber measurements for linac installation and quality assurance," said Glaser. "Or, if multiple cameras were simultaneously used to provide near real-time tomographic data, more complex radiotherapy procedures such as intensity modulated and arc therapy treatment plans could be investigated using full 3D dosimetry."

Related articles in PMB
Projection imaging of photon beams using Čerenkov-excited fluorescence
Adam K Glaser et al Phys. Med. Biol. 58 601
The potential for Cerenkov luminescence imaging of alpha-emitting radionuclides
N L Ackerman and E E Graves Phys. Med. Biol. 57 771
Plastic scintillation dosimetry: comparison of three solutions for the Cerenkov challenge
P Z Y Liu et al Phys. Med. Biol. 56 5805