Quality assurance of a radiotherapy registry

Sammanfattning: The radiotherapy clinics in Sweden have been without a functioning national platform consisting of dose data from patients undergoing radiotherapy. A national collaboration between clinics will improve the quality of radiotherapy since clinics will be able to compare dose data from treatment plans between clinics. It will also help and improve future researches in radiotherapy. A new national quality registry for radiotherapy in Sweden is under development and is located on the INCA platform. The aim of this study is to do a quality assurance of the INCA registry. The data stored in the registry are calculated from the treatment plans stored locally at the clinics. The quality assurance of the registry is done by creating a program run by Python code and by using Streamlit as the graphical user interface. The program takes dose and volume data from the dose volume histograms located in treatment plans from the INCA database and compares it with the dose and volume data from the local clinics' treatment planning system. The different treatment planning systems considered in the program are Oncentra(Elekta, Sweden), Eclipse(Varian, U.S.), RayStation(RaySearch Laboratories, Sweden) and Monaco(Electa, Sweden). The compared absorbed doses are the dose to 99% of the structure volume(D99%), D98%, D50%, D2% and D1%. The program generates how much the INCA data differs from the TPS data in percent and is named QARS(Quality Assurance of the Radiotherapy Database in Sweden). A verification of the created program and a preliminary evaluation is done on a limited dataset containing three patient groups(prostate patients, lung patients and head and neck patients) with five patients in each group. The dataset is run through the program with patient data from both Oncentra and Eclipse. The result indicates that all the near-maximum doses, D2% and D1% in INCA are very close to their corresponding TPS dose. There is a more noticeable difference in the near-minimum doses, D99% and D98% but also for some D50% where the difference seems to increase in larger structure volumes with very low doses and in very small structure volumes, smaller than 0.01 cm3. It is compared how well INCA agrees with Oncentra and Eclipse respectively and it is clear that Eclipse has a smaller difference to INCA than Oncentra for structures with very small volumes and larger structures with low doses. To summarise the study, it generates a program for quality assurance of the national quality registry for radiotherapy in Sweden which hopefully can help improve the quality of radiotherapy and help future researches in the field.