Which radiochromic film grade matches which measurement, how masks and indexed baseplates actually hold setup accuracy, and the order to build a QA programme in when a department is starting from nothing.
Radiotherapy is a treatment delivered blind. The beam is invisible, the dose is deposited inside the patient, and nobody in the room can see whether the plan was executed correctly. Quality assurance is how a department converts that uncertainty into evidence. As Bangladesh commissions more linear accelerators, the QA programme around them deserves the same attention as the machine purchase, and it usually gets a fraction of it.
These two terms are used loosely, which is why departments buy the wrong consumables.
The practical consequence of blurring the two is a single purchase order for one film grade that is expected to cover both, which then does one job adequately and the other badly. Order them as separate lines with separate annual quantities, estimated from the number of scheduled accelerator QA sessions in the year and the number of modulated plans the department expects to verify. Both numbers are knowable a year ahead, and a department that cannot state them has not planned its QA programme, only bought equipment for it.
Radiochromic film changes colour on exposure to ionising radiation. It is self-developing, so it needs no darkroom and no chemical processing, which removes an entire failure mode from the workflow. It also gives genuinely two-dimensional dose information at high spatial resolution, which point detectors and coarse arrays cannot. For a department in Bangladesh the practical advantage is that film requires no wet processing infrastructure at all and can be stored and handled in normal room light.
What it does need is a flatbed scanner in a known configuration and a physicist who scans consistently. Same scanner, same orientation, same warm-up, same region of the glass, every time. A film programme fails on scanning discipline far more often than it fails on the film.
Gafchromic film from Ashland is the reference standard, and the range is differentiated by dose. Choosing the wrong grade is the most common error in a new programme: a film saturated beyond its useful range gives an unusable result, and one exposed far below it gives a noisy one.
| Film | Stated dose range | Primary use |
|---|---|---|
| EBT3 | 0.2 Gy to 10 Gy, 100 keV into the MV range | The general-purpose workhorse. IMRT and VMAT plan verification, brachytherapy, Star Shot, Picket Fence, flatness and symmetry |
| EBT-XD | 0.4 Gy to 40 Gy, 100 keV into the MV range | Patient QA at higher dose: stereotactic radiosurgery, small field dosimetry and hypofractionated treatment |
| RTQA2 | 0.2 Gy to 10 Gy, 100 keV into the MV range | Routine machine QA: light field and radiation field alignment, Star Shot, Picket Fence, HDR brachytherapy position verification |
| MD-V3 | Up to 100 Gy | Gamma and X-ray dose mapping, research work, high dose dosimetry and sterilisation applications |
| HD-V2 | Up to 1000 Gy | Unlaminated film with the top layer omitted, for low energy photon and electron detection and very high dose work |
Two practical notes for a Bangladeshi department. Film is sensitive to heat and to ambient light over long exposure, so store it in the physics office rather than the treatment area, and do not leave a box in a room whose air conditioning goes off with the mains. And buy from one batch where you can: batch-to-batch response varies, so a calibration curve built on one batch and applied to another is a quiet source of error that nobody notices until an audit.
A department can verify a plan perfectly and still deliver it to the wrong tissue if the patient is not in the same position on day 20 as on the planning CT. Modern techniques have tightened margins, which means setup reproducibility now carries more of the accuracy budget than it used to. Immobilisation equipment is not an accessory to the accelerator, it is part of the treatment chain, and it is the part most often bought on price. It is also the only part of that chain a department can improve without buying a new accelerator, which makes it the highest-return line in most oncology equipment budgets and the one that gets the least attention at tender.
Thermoplastic sheets soften in warm water, are moulded to the patient, and set into a rigid personal shell. They are the standard for head, head-and-neck and head-neck-shoulder treatment. Masks come in families matched to the baseplate type: S-type head and head-neck-shoulder masks, U-type in slim and standard frames including reinforced, openface and open eyes and mouth versions, L-type triangular head, head-neck-shoulder and chest-pelvic masks for O-type baseplates, P-type in the same shapes with an openface series, and violet X-Knife masks for stereotactic frames with a non-stick coating so the material does not catch on hair or beard.
Standard thicknesses are 2.4 mm and 3.0 mm, with 2.0 mm and 3.2 mm available, and perforation options of 22 per cent and 38 per cent. Thicker, less perforated material is more rigid and more restrictive; thinner material is more comfortable and slightly less constraining. That choice is clinical, not commercial, and the openface variants exist because claustrophobia during a six-week course is a real cause of interrupted treatment rather than a comfort nicety.
The mask has to attach to something that itself attaches to the couch in a repeatable position. Carbon fibre baseplate systems cover head, neck, shoulder, thorax, pelvic and whole-body treatment: all-in-one plates that take a patient from head to full body, head and shoulder plates around 550 mm square, dedicated pelvic plates, U-frame head plates, and pin-lock head-shoulder plates around 1000 mm by 606 mm that are dual configured for adult and paediatric patients. They index to the couch so the whole assembly returns to the same coordinates each session.
Buying masks without a properly indexed baseplate wastes most of the benefit, and buying a baseplate without the indexing bar that matches your couch wastes all of it. Check the couch interface before the order, not on the delivery note.
For body and extremity treatment where a mask is not appropriate, a vacuum cushion moulds to the patient's contour and locks when air is evacuated. Look for a coated nylon shell that resists tearing and staining, a built-in indexing batten, and adapters that fit whichever pump the department already owns. MR safe versions exist and matter if simulation is moving to MRI.
Stereotactic body radiotherapy needs its own set: a baseplate with adjustable breathing bridges, belly and knee paddles that move left to right and up and down to control the breathing range, a wing board, a respiratory belt, knee and feet cushions, a headrest, an indexing bar and a shoulder retractor. Breast treatment in the prone position needs a dedicated board with variable aperture openings and a reversible breast section so either side can be treated on the same equipment.
Departments moving to MRI-based simulation or MR-guided radiotherapy need non-ferrous positioning equipment. Standard baseplates and frames cannot go into the magnet. If MRI simulation is anywhere in your five-year plan, specify MRI-compatible equipment now rather than buying the same items twice.
The failures that show up in Bangladeshi departments are consistent enough to list.