Radiotherapy treatment planning stations and systems (TPS) · regulatory class, image import and CT number calibration, anatomical modelling and DVH, dose-calculation accuracy against measurement, MU checks, commissioning and periodic QA to IAEA TRS-430, plan output and transfer
A treatment planning station is the computer workstation and planning software on which radiation oncologists and medical physicists outline targets and organs on CT images and design and optimise radiotherapy treatment plans before delivery on a linear accelerator. Ask three things first: which treatment machines, beam types and techniques the planning software must model; how images and plans move between scanners, the planning system and the treatment machines; and how the system will be commissioned and quality-assured before clinical use.
Radiotherapy treatment planning systems and workstations — planning software, dose-calculation engines and the computers and displays that run them — for external-beam photon, electron and particle therapy. Linear accelerators, oncology information and record-and-verify systems, contouring-only software, dosimetry equipment (see ionization chamber array) and general medical-imaging workstations are outside this page except where compared.
Start with your task
01
US rule 21 CFR 892.5050 identifies a medical charged-particle radiation therapy system as a device that accelerates high-energy charged particles such as electrons and protons for radiation therapy, and states that this generic device type may include signal analysis and display equipment, patient and equipment supports, treatment planning computer programs, component parts and accessories. It is classified Class II, with a film dosimetry (film scanning) system used as a quality-control accessory exempt from premarket notification subject to the limitations in 892.9. [1]
Editorial note: ask the supplier for the regulatory clearance or certification that covers the specific planning software version in each market where it will be used, and whether the hardware workstation is part of that clearance.
Give the supplier the treatment machines (make, model, energies, MLC), beam types (photon, electron, proton) and techniques (3D-CRT, IMRT, VMAT, stereotactic), imaging inputs (CT, MR, PET) and DICOM connectivity, number of planning and contouring seats, dose-calculation algorithms required, optimisation and automation features, integration with the oncology information system, hardware specification, regulatory clearance per market, commissioning support and training, and service terms.
02
An AAPM meeting abstract reports commissioning a commercial 3D TPS for 6 MV photons using IAEA Technical Reports Series 430. CT images of a phantom with known inhomogeneities were transferred to the system and tested for orientation, CT number–density correlation, image geometry, anatomical modelling, dose-volume histograms and plan output and transfer; dose calculation was compared with measured depth doses, profiles and monitor units for square, rectangular, asymmetric, MLC-shaped and wedged fields, and clinical test cases were checked for MU and point dose. Results showed CT number variation up to ±20 HU, a 3.7% volume reduction for a 350 cc cylinder from contour expansion at sharp corners, 2 mm depth-dose variation in the build-up region, up to 1 mm in the penumbra (2.5 mm with a blocked central axis), MU agreement within ±2% except 3.8% with the central axis blocked, and clinical point doses within ±2%. A subset of TRS-430 commissioning tests was then adopted as a periodic QA programme. [2]
Editorial note: budget physicist time and phantom measurements for commissioning before first clinical use, and ask the supplier which commissioning data, test plans and validation tools it provides. [2]
IAEA-TECDOC-1583 (Commissioning of Radiotherapy Treatment Planning Systems, January 2008) turns TRS-430 into a step-by-step clinical commissioning protocol built on the CIRS 002LFC anthropomorphic phantom with certified electron-density plugs; its companion IAEA-TECDOC-1540 covers specification and acceptance testing against IEC 62083. Its sample acceptability criteria for external-beam calculations are 2% for output factors and central-axis data in simple geometry, 3% for off-axis data and for wedged, inhomogeneous, irregular or asymmetric fields, 3–4% outside beam edges, and 2 mm for radiological field width and penumbra. [3]
The report shows CT number to relative electron density curves from six institutions diverging above water density, and estimates the resulting dose error at about 2% for a 6 MV beam crossing 5 cm of material with relative electron density 1.5 (800 CT numbers), which is why each site's own CT calibration must be entered and checked. If clinical commissioning results fall outside the acceptance tolerances, it advises beam-specific calculation checks first and then the manufacturer. [3]
03
Dose calculation in a TPS depends on the table that converts CT numbers to density, which the AAPM commissioning abstract above tested as "CT number–density correlation". [2] A 2025 national survey in the Journal of Radiation Research covered 34 Japanese institutions using the Radixact helical tomotherapy system and compared their registered CT-to-mass-density (CT–MD) tables. Tables from planning CT had a median of 10 data points, and tables for the on-board ClearRT and CTrue images a median of 8. ClearRT tables varied by up to 300 HU in cortical bone between institutions, and air-plug CT numbers varied by about ±30 HU. All CTrue plug values stayed within the AAPM Task Group 148 tolerances the authors quote: ±50 HU for bone, ±30 HU for water and ±50 HU for lung. [4]
The authors conclude that standardising phantom geometry, air CT number handling and QA protocols, in particular half-phantom calibration, may improve CT–MD table consistency and dose accuracy in adaptive radiotherapy. [4] Editorial note: ask the TPS supplier how many CT–MD tables the system can store, whether a table can be tied to each scanner, protocol and reconstruction method, and how table changes are logged, then set a routine table QA interval in the acceptance plan.
04
When requesting quotations, ask each supplier for supported machines and techniques, dose algorithms, regulatory status, commissioning support, licensing model and integration references.
05
| Step | What is checked | Fix in the order |
|---|---|---|
| Image import | Orientation, CT number–density, geometry. [2] | DICOM sources. |
| Anatomical modelling | Contours, volumes, DVH. [2] | Tools. |
| Dose calculation | Depth dose, profiles vs measurement; 2–3% criteria. [2] [3] | Algorithms. |
| MU calculation | Within ±2% in the reported tests. [2] | Verification. |
| Plan output and transfer | To treatment machines. [2] | Interfaces. |
06
The accuracy rows mix one clinic's measured results (±20 HU, MU within ±2%) with IAEA's sample acceptance criteria; workstation hardware and licence counts appear in none of the reviewed sources.
| Field | What it means and published values | Ask the supplier for |
|---|---|---|
| Regulatory class | Class II under 21 CFR 892.5050 (US). [1] | Clearance per market. |
| Acceptance standard | IEC 62083 basis for acceptance tests. [3] | Test protocol and data. |
| Beam types | Electrons and protons named in rule. [1] | Supported beams. |
| CT number accuracy | ±20 HU in reported tests; site CT calibration needed. [2] [3] | Calibration method. |
| Dose accuracy | MU within ±2% reported; sample criteria 2–4% and 2 mm. [2] [3] | Validation data. |
| Workstation hardware | Not stated in reviewed sources. | CPU, GPU, displays. |
| Seats / licences | Not stated in reviewed sources. | Planning and contouring seats. |
| Connectivity | Plan output and transfer tested. [2] | DICOM-RT, OIS. |
07
Paste it into your enquiry below or into an email to a maker, then replace every bracket.
Machines: [models; energies; MLC] Techniques: [3D-CRT; IMRT; VMAT; SRS; protons] Imaging: [CT; MR; PET; DICOM] Seats: [planning; contouring] Integration: [OIS; record-and-verify] Regulatory: [markets] Support: [commissioning; training; service]
Manufacturer profiles associated with Treatment Planning Station.
Explore company information and manufacturing capabilities, then discuss the scope of your project with a relevant supplier.
21 CFR 892.5050 includes treatment planning computer programs within medical charged-particle radiation therapy systems, Class II. [1]
One clinic used IAEA TRS-430 tests covering image import, anatomical modelling, DVH, dose and MU calculation against measurement, and plan transfer. [2]
MU calculations within ±2% except 3.8% with a blocked central axis, and clinical point doses within ±2%. [2]
IAEA-TECDOC-1583 gives sample criteria of 2% for output factors and simple central-axis data, 3% for off-axis and complex geometry, 3–4% outside beam edges and 2 mm for field width and penumbra. [3]
A survey of 34 Japanese Radixact centres found ClearRT tables varying by up to 300 HU in cortical bone and air-plug values by about ±30 HU; 43% of centres did no routine QA of CTrue tables. [4]
Sources checked 2026-10-12. 21 CFR 892.5050 is a US regulation; the AAPM abstract reports one clinic's commissioning results and is not a general specification; IAEA-TECDOC-1583 gives commissioning tests and sample criteria; the Journal of Radiation Research survey reports CT-to-density tables at 34 Japanese tomotherapy centres.
Ask for use case, specification boundaries, supplier type, and RFQ preparation information for this product.
Compare manufacturer profiles with relevant product and process capability.