Medical Physicist's Notes · Delta4 Phantom+
Every radiotherapy department runs two parallel QA programmes — routine machine QA against AAPM TG-142 / TG-198, and patient-specific pre-treatment QA against AAPM TG-218. In most departments, those programmes run on separate equipment chains. The Delta4 Phantom+, with the optional Delta4 Machine QA software module, runs both workflows from the same 1,069-detector array. This page lays out what each programme requires, what the Machine QA module is documented to cover, and what changes when the two share one calibration and one software base.
Neither programme is optional. The AAPM task-group reports define what each one is for and what tolerances it must hold to.
Verifies the planned dose distribution can be delivered as intended on this machine, this day.
Universal tolerance limit 95 % at 3 %/2 mm gamma; universal action limit 90 %.
Miften M et al., AAPM TG-218, Med Phys 45(4):e53–e83 (2018).
Demonstrates the linac itself is operating within tolerance: beam output, profile constancy, MLC, mechanical and imaging tests.
TG-142 specifies tolerance bands by frequency; TG-198 walks through implementation. Both updated AAPM TG-40.
Klein EE et al., AAPM TG-142, Med Phys 36(9):4197–4212 (2009); Hanley J et al., AAPM TG-198, Med Phys 48(10):e830–e885 (2021).
In practice, the two programmes run on separate equipment chains. Patient-specific QA runs on an array detector — diode array, ion chamber array, or EPID-based system. Routine machine QA runs on different hardware again: a daily quick-check device for output and flatness, a separate phantom for MLC and picket-fence, a third arrangement for mechanical isocentre and Winston-Lutz. Each chain has its own absolute calibration, its own SOP, its own training cycle, and its own storage footprint in the bunker.
Both task-group reports define what the programmes have to deliver. Neither prescribes that they have to use separate hardware. That separation is a clinical-practice convention — and one of the things a multi-purpose array detector with switchable software can revisit.
Framing: observed clinical practice, not citation. AAPM TG-142 / TG-198 / TG-218 do not mandate distinct equipment for routine and patient-specific QA.
What does the QA programme look like when one detector, one calibration and one software family covers both pre-treatment patient QA and routine monthly machine QA?
The optional Delta4 Machine QA software module turns the patient-QA phantom into a routine machine-QA tool. Each row below is taken from the Manufacturer product page; I do not list tests the manufacturer does not document.
Beam constancy at multiple gantry angles
Output, flatness, symmetry and energy measured simultaneously across multiple projections per session — using the same 1,069 p-Si diode array.
Source: the manufacturer Delta4 Machine QA — feature list.
MLC performance — automated picket-fence
Absolute and relative leaf-position deviations measured at multiple gantry angles. Halcyon dual-layer MLC supported.
Source: the manufacturer Delta4 Machine QA — feature list.
Accelerator start-up behaviour, sub-MU resolution
The same per-pulse architecture used for patient QA captures linac warm-up at single-pulse resolution (~1 mGy / 50 nGy). Early-pulse deviations that an integrating chamber would average over are resolved.
Source: the manufacturer Delta4 Phantom+ and Machine QA technical spec.
Long-term trending and database
Machine QA results store alongside patient QA records in the same Delta4 database — one statistical view of the entire QA programme.
Source: the manufacturer Delta4 Machine QA — feature list.
Tests not currently in the published Delta4 Machine QA spec — Winston-Lutz mechanical isocentre, couch rotation isocentricity, light/radiation coincidence — remain on dedicated tools in your existing QA chain. Frame Delta4 Machine QA as the array-detector-dependent portion of TG-142, not as a single-tool replacement for the entire TG-142 test set.
Four practical shifts — each is a workflow observation, not a numerical claim.
The Delta4 Machine QA module is a software add-on to the same phantom hardware already used for patient-specific QA. The avoided line item is the second array detector — not the daily quick-check device, which sits in a different role.
The same absolute calibration of the same detector array is the reference for both workflows. Cross-checking patient QA results against machine QA trends does not require reconciling two independent calibration chains.
Both workflows are driven from the Delta4 software family — the patient-QA core software (included) and the Machine QA module (optional). A new physicist or registrar learns one interface, not two.
The wireless Li-ion phantom that wheels in for patient QA is the same unit that wheels in for monthly machine QA. Two QA SOPs share one footprint and one cable-free setup.
Scope of this page
The coverage table reflects the documented feature list of the Delta4 Machine QA software module on the Manufacturer product page. We are not aware of a peer-reviewed paper that independently validates Delta4 against the full AAPM TG-142 test set, and we do not claim one. For TG-142 tests outside the Machine QA module's published scope (mechanical isocentre, couch rotation, light/radiation coincidence), continue with your existing QA chain.
Sources cited on this page
Delta4 Phantom+
Sibling posts in the Delta4 Phantom+ family.
For Medical Physicist
At what magnitude of MLC, gantry, collimator or MU error does the Delta4 gamma passing rate cross the 95 % action threshold? Per Udee 2024, J Med Phys.
Read this Medical PhysicistIon chambers under-respond and many diodes drift at high dose-per-pulse. What changes when the detector integrates one pulse at a time at 50 nGy resolution?
Read this