From Sparse to Precise: Wavelet Implicit Neural Network for Radiotherapy Commissioning

The two cornerstones of radiation cancer therapy are medical linear accelerators (LINACs) for producing high energy X-rays and treatment planning systems (TPS) for computing the optimal therapy plans. The goal is to deliver targeted lethal doses to destroy tumors while sparing surrounding normal tissues. For LINAC and TPS to work seamlessly together, TPS must have an accurate physics model of the LINAC. To do this, commissioning is performed where extensive measurements of the LINAC are made and uploaded into the TPS. Commissioning is a tedious and expensive process, taking months to complete. To ensure continuous and accurate delivery of the treatment dose after the initial commissioning, routine quality assurance must also be performed where measurements are made and compared. The goal of this research is to develop a model to generate all necessary LINAC physics data needed from a very few physical measurements, therefore reducing the cost (e.g., labor intensiveness, machine down time, less measurement errors, etc.). Specifically, we employed Wavelet-based Implicit Neural Network (WINN) and investigated its performance with different wavelet functions-Symmetric Gaussian, Asymmetric Gaussian, Mexican Hat, and Morlet. The results demonstrated that our model activated by Morlet wavelet, with its unique design combining a cosine term filtered by a Gaussian envelope, consistently meets clinical tolerance with gamma passing rates over <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$95 \%(1 \% / 1 ~\text{mm}$</tex> criteria) and mean absolute errors <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\leq 0.5 \%$</tex>. This superiority arises from its ability to effectively capture the overall behavior of signals through the Gaussian component and the spatially varying details through the cosine component. Furthermore, it achieves a 99.46 % reduction in network parameters compared to previous work which is based on Sinusoidal Implicit REpresentation Networks (SIREN). Our model has a total size of less than 1 MB, making it ideal for deployment in resource-constrained clinical settings.