Radiation Therapy Deficiencies Identified During On-Site Dosimetry Visits by the Imaging and Radiation Oncology Core Houston Quality Assurance Center

Abstract

Purpose To review the dosimetric, mechanical, and programmatic deficiencies most frequently observed during on-site visits of radiotherapy facilities by the Imaging and Radiation Oncology Core Houston office (IROC Houston). Methods IROC Houston’s findings between 2000 and 2014, including 409 institutions and 1020 linacs, were compiled. IROC Houston’s on-site evaluation includes verification of absolute calibration (tolerance of ±3%), relative dosimetric review (tolerances of ±2% between TPS calculation and measurement), mechanical evaluation (including MLC and kV-MV evaluation against TG-142 tolerances), and general programmatic review (including institutional QA program versus TG-40 and TG-142). Results An average of 3.1 deficiencies was identified at each institution visited, a number that has decreased slightly with time. The most common errors are tabulated, and include TG-40/TG-142 compliance (82% of institutions were deficient), small field size output factors (59% of institutions had errors ≥3%), and wedge factors (33% of institutions had errors ≥3%). Dosimetric errors of ≥10%, including in beam calibration, were seen at many institutions. Conclusions There is substantial room for improvement of both dosimetric and programmatic issues in radiotherapy, which should be a high priority for the medical physics community. Particularly relevant was suboptimal beam modeling in the treatment planning system and a corresponding failure to detect these errors by not including TPS data in the linac QA process.

A comparison of IROC and ACDS on‐site audits of reference and non‐reference dosimetry

“…This fundamental part of dosimetry testing remains relevant to identify flaws in individual implementation of beam models and limitations in commercial algorithms . The IROC recommendations similarly continue to show fundamental issues, particularly with small field output factors and wedge factors . The broader scope of IROC on‐site visit also includes QA program review, with shared aspects with the QUATRO clinical audit approach .…”

“…Independent audits of dosimetry in radiotherapy clinics are an excellent quality improvement tool for detecting systemic errors in dosimetry and encouraging consistency in radiotherapy practice. Dosimetry audits are recognized as international best practice for departmental quality assurance and clinical trial accreditation and have uncovered systemic problems with radiotherapy dose determination, such as dosimetric inaccuracies in heterogeneous dose calculations and small field dose calculations, as well as identifying unique errors in calibration and dose determination at individual clinics …”

“…Dosimetry audits are recognized as international best practice for departmental quality assurance and clinical trial accreditation and have uncovered systemic problems with radiotherapy dose determination, such as dosimetric inaccuracies in heterogeneous dose calculations [1][2][3][4] and small field dose calculations, [5][6][7] as well as identifying unique errors in calibration and dose determination at individual clinics. 8,9 Dosimetry audits can test different elements of the radiotherapy chain, including a simple check of reference dosimetry (Level I audit), increasing in complexity to an end-to-end dose delivery evaluation (Level III). [10][11][12] An intermediate (Level II) audit is one that probes the commissioning data and development of the treatment planning system (TPS) beam model.…”

Deep learning for patient‐specific quality assurance: Identifying errors in radiotherapy delivery by radiomic analysis of gamma images with convolutional neural networks

“…The quantitative comparison of delivered vs expected dose is usually done with gamma analysis, which compares quantitative agreement and distance‐to‐agreement between planned and delivered doses . Clinical verification of radiotherapy plans is important, because errors in linear accelerator and treatment planning system commissioning are commonly detected on IROC site audits and phantom credentialing . However, despite its near universal prevalence in the clinical environment, IMRT QA is often criticized as being insensitive to errors or less effective than other common physics checks …”

“…1 Clinical verification of radiotherapy plans is important, because errors in linear accelerator and treatment planning system commissioning are commonly detected on IROC site audits and phantom credentialing. 2,3 However, despite its near universal prevalence in the clinical environment, IMRT QA is often criticized as being insensitive to errors or less effective than other common physics checks. [4][5][6][7][8] Recently in medical physics, there has been considerable interest in the application of radiomics, the quantitative data mining of characteristics such as intensity, frequency, shape, and texture from medical images.…”