COVID-19 Information

PhD Thesis Proposal: Ramish Ashraf

Monday, November 23, 2020, 9:45am


For info on how to attend this videoconference, please email

"Radioluminescence Based Dosimetry for Small Fields and FLASH Radiotherapy"


For accurate dosimetric chraterization of modern dose-delivery techniques, a dosimeter should ideally be energy independent, exhibit high spatio-temporal resolution and dose-rate independence. Radioluminescence, the phenomena of production of optical photons in response to radiation, can be an effective tool for performing accurate dosimetry because it exhibits the aforementioned ideal characteristics. The first goal of this work was to characterize a time-gated intensified CMOS camera which was to be used for imaging radioluminescence. The camera was successfully employed to perform end-to-end patient specific quality assurance by imaging 2D projected dose distributions in real-time for complex, highly modulated radiotherapy plans. In particular, the technique was shown to be more sensitive to multi-leaf collimator positioning errors (~ 1 mm) as compared to a commercial cylindrical diode array.

The next aim of this work was to use optical imaging for measuring 3D dose distributions. Depending on the application (static vs complex dynamic plans), different 3D reconstruction algorithms were used. For static beams, a single optical camera and tomography based methods were used to reconstruct beams as small as 4 mm. The central axis profiles from the optically reconstructed 3D dose distributions matched well with a commercial stereotactic diode. In the case of FLASH (10 MeV electron beam), this approach provided central axis data for individual linac pulses delivered at a mean dose-rate of 60 Gy/s. This technique enabled rapid quality assurance of FLASH beams on an individual pulse basis without suffering from dose-rate effects. For imaging complex dynamic plans, two optical cameras were employed providing two orthogonal views. The two orthogonal views were successfully combined to image a static multitarget beam, similar to a single control point of a stereotactic plan.

Future work in this thesis will primarily focus on two aspects; 1) accurate determination of output factors for small fields which have been typically difficult to quantify using most conventional detectors, and 2) beam monitoring and real time feedback for a FLASH linear accelerator since conventional transmission ionization chambers saturate and show significant dose per pulse dependence at dose rates relevant in FLASH.

Thesis Committee

For more information, contact Daryl Laware at