Feasibility of a 3D cubic electronic detector array for radiotherapy quality assurance

The need for compelling 3D dosimetry QA in radiotherapy with an electronic device has eluded us for too long. This study could be the first step to fulfill this necessity.

To demonstrate that a cubic electronic detector array consisting of active matrices can provide reliable and reproducible 3D dosimetric data in radiotherapy.

A 30 × 30 × 15 cm³ PMMA phantom of 2645 pixels distributed in five active matrices (AMs) of 18 × 18 cm² sensitive area was constructed. A pixel consists of a diode, capacitor, and MOSFET, and are distributed in a 23 × 23 matrix within the AM. Twenty-three shared data channels per AM extract data from 529 pixels in an AM. Raw data is read via a DDC264EVM board and fed to a Python-based program for data analysis. Linearity (10, 20, 40, 50, 80, 100, 200, and 400 MUs), dose-rate (100, 200, 300, 400, 500, and 600 MU/min) effects for 50 MU delivery, PDD (5 × 5, 10 × 10, and 20 × 20 cm² field sizes), and profile (5 × 5, 10 × 10, and 20 × 20 cm²) measurements for a 6 MV photon beam were collected. PDD and profile data for a specific field size were acquired simultaneously with few MUs and no beam scan. Measured data were compared to 400 MU linearity baseline, 600 MU/min dose rate baseline, and Monte Carlo (MC) detector model data, and MC data of a homogeneous PMMA medium for PDD and profiles.

Mean linearity results deviated from baseline by 0.82%, 0.84%, 0.88%, 0.90%, and 0.96% for each (top-to-bottom) AM after multiplying it by its appropriate factor. Mean dose rate deviations were 0.5%, 0.5%, 0.5%, 0.5%, and 0.6% for each (top-to-bottom) AM. Maximum PDD data differences for measurement vs MC model/measurement vs MC PMMA were 1.5%/3.3%, 0.5%/2.4%, and 0.8%/1.7% for 5 × 5, 10 × 10, and 20 × 20 cm², respectively. Profile data gamma analysis revealed 100.0% for a 5 × 5 cm², 100.0% for 10 × 10 cm², 99.6%-100.0% for 20 × 20 cm².

This study demonstrates that a 3D cubic electronic detector array can provide reliable dosimetric data not only in 2D, but 3D also. If realized, this technology could be a paradigm shift in radiotherapy linac commissioning, monthly quality assurance (QA), and patient-specific plan QA.