Practical Aspects in Optimisation of Radiological Protection in Digital Radiography, Fluoroscopy, and CT


Draft document: Practical Aspects in Optimisation of Radiological Protection in Digital Radiography, Fluoroscopy, and CT
Submitted by Peter Thomas, Australian Radiation Protection and Nuclear Safety Agency
Commenting on behalf of the organisation

Comments from Australian Radiation Protection and Nuclear Safety Agency (ARPANSA) Medical Imaging section on ICRP draft document - Practical Aspects in Optimisation of Radiological Protection in Digital Radiography, Fluoroscopy, and CT

General comments

Is there a way to highlight in the document some of the key issues that will make a difference and are commonly areas where typical practices may be sub-optimal? I.e. things that are commonly not done, even by otherwise good facilities, but should be done, such as evaluation of reject images as part of QA and process improvement. It is covered but in a way that is no different to any other line items. Another example is clearly defining responsibility for tasks and processes.

Should the format of the QA checklists be as uniform as possible across modalities? E.g. X-rays and fluoro – many of the requirements are similar and could be laid out the same way.

A clearer or better process to ensure greater consistency of diagnostic image quality when optimising. i.e. greater focus on radiologists’ needs to enable the best and also the least talented radiologists to achieve optimal diagnoses.  (Paragraph 166 makes some acknowledgement of this: “… observer perceptions, and training and experience of the interpreter…”) Diagnostic effectiveness is almost always more important than minor dose savings.

Lines 163-166

Suggest replacing “anatomical regions” with “clinical tasks” (see also line 437 and paragraph 36) Exposure factors should be established not only for anatomical regions and patient characteristics but also for various clinical tasks (which have different requirements on image quality)

Suggest replacing “possible use of copper filtration” with “possible use of different filtrations”

This will cover use of Sn filters in CT as well as Cu filters in planar imaging.

Fig 1.1 caption, line 235

Swap order of B and C in “moving upwards from D, through B, and C, towards A”.

Paragraph 9, lines 275-287

“Reasonably achievable” can be understood to include achieving the clinical goal. Avoiding unnecessarily high dose is an important aim of optimisation. E.g., a higher mAs image will give a better image but is it necessary to use high mAs (dose creeping)?

Paragraph 16 Line 368

“This will require not only regulatory approval of algorithms and procedures…”

Regulators may lack the capabilities and resources required. May require assistance of relevant professional bodies.

Line 445

Change to “... constant Exposure Index for a particular clinical task.”

Paragraph 25 Line 468

“facilities”

Paragraph 29 Line 501

suggest inclusion of medical physicists (where available) as well

Box 2.1

“The image recorded is stored directly in a diode array within the imaging detector. “ Don’t think diodes can be used as charge storage elements.

Line 517

Change to “... measured by a meter attached to the output port of the x-ray unit or derived from a look-up table.” Many systems do not have a built-in KAP meter and instead rely on a look-up table.

Lines 657-8

The quoted air kerma incident on the image sensor seems excessively high.

For example, Trixell Pixium CsI/aSi detectors normally only require 2.5µGy incident air kerma at the image sensor.

Line 681

Change to “... and more recently AAPM Task Group 232 (Dave et al., 2018) and IEC 62494 (IEC, 2008)”. Include reference to IEC 62494

Lines 703-4

Change to “... a DI above +1 indicates a higher exposure than expected and a DI less than −1 indicates a lower exposure than expected.” Worth keeping the full wording for clarity.

Paragraph 55, line 718

The term “technologist” is used here but “radiographer” appears much more widely in the document. Suggest using “radiographer” for consistency with the rest of the document.

Line 740

Change to “... 20% if the SID is shortened by 10 cm.” Again, worth keeping the full wording for clarity.

Paragraph 59, lines 750-753

Is there any reference or non-anecdotal evidence to back up this claim?

Figure 2.3 c and d

Is this actually an example of poor collimation? It looks like it has in fact been correctly collimated and the extreme windowing and leveling in (d) is revealing the signal from scattered and extra focal radiation. Also adjust formatting in the caption to c) and d) for consistency.

Paragraph 62, line 797

The image for the knee should not really require a grid anyway (as it’s an extremity). Very little scatter to begin with. Virtual grid image is done with a higher kV and a lower exposure index, actual patient exposure (DAP) essentially equivalent

Table 2.4, Line 871

Action: reduction in number of images per procedure. Influence on diagnostic information: None. Not sure that follows (unless the removed views were duplicated)

Fig 2.7 caption, line 895

NICU acronym introduced before being defined.

Box 2.4, line 937

Is two-point verification sufficient? Australian standards require use of at least three patient identifiers.

Image Capture During the Exam

Change to “1. Beam, body part, and image receptor aligned, SID checked, use of grid determined”

Paragraph 90, lines 1059-1061

Noting the years of the UNSCEAR reports in the text would be useful for those of us who don’t know the dates of publication.

Paragraph 90, line 1066

Suggest “cataracts”

Paragraph 90, line 1071

Only instance of “... medical radiation physicist...”. Elsewhere “medical physicist” is used. Suggest changing to “medical physicist”.

Line 1097, Box 3.1, Conventional R/F Systems

Change to “... which can be moved by the operator closer to or further from …"

Paragraph 107, lines 1196-1197

Why are the eyes specifically mentioned? Viewing from larger distances reduces the dose in general, not just specifically for the eyes.

Box 3.3 (last sentence)

"… which normally leads to degradation of image quality for high attenuation objects such as obese patients, lateral or oblique projections, or thicker body parts....” This sentence implies the regulatory limit is normally not enough to cope with obese patients, lateral or oblique projections, or thicker body parts. Is there any evidence to support this? If no, then suggest changing “normally” to “may”.

Paragraph 126, line 1314

“…measure dose quantities.” Should be “measured”

Paragraph 143, line 1457

Replace “ADRS” with ADRC (as used throughout the rest of the publication).

Box 3.7, line 1492, Fluoroscopy dose

Instructions for removing anti-scatter grid are just given in relation to abdominal thickness. Could this advice be expanded to include limbs and extremities?

Box 3.7, line 1492, Monitoring dose

Missing closing parenthesis for last dot point.

Paragraph 154, line 1512

“…Section 3.6 is exceeded.” “are exceeded”

Paragraph 154, line 1513

In Australia, the “operator” is the technologist/radiographer. We would regard it as the responsibility of the radiologist/physician to make a note in the patient record rather than the operator (radiographer). Same comment for lines 1521 and 1523.

Paragraph 157, line 1546

“Clinical follow-up for high patient radiation doses.” Suggest further clarifying that this is in relation to skin injury or tissue reactions and not for concern over stochastic effects

Paragraph 159, line 1589

“…with measurements being made of the impact of changes.”

Suggest noting that measurements should cover both dose and image quality

Line 1607

“Epko” should be “Ekpo”

Line 1612

Grammar: “the lack of an adequate dose management (?) as part of the QA programme for CT system resources”

“the lack of adequate dose management as part of the QA programme”

Or “the lack of an adequate dose management programme as part of the QA programme”

Line 1629

Grammar: “The next step is protocol optimisation, potential dose reduction, with CT depends on appropriate selection of scanning parameters (both acquisition and reconstruction).”

Perhaps

“The next step is protocol optimisation and potential dose reduction, which in CT depends on appropriate selection of scanning parameters (both acquisition and reconstruction).”

Line 1645

Grammar: “The impact that exposure parameters have on patient dose or potential issues on the system performance hindering diagnostic image quality, will go undetected, unless scanner performance is characterised and dose levels and image quality are monitored.”

Perhaps

“The impact that exposure parameters have on patient dose, or potential issues with system performance hindering diagnostic image quality, will go undetected, unless scanner performance is characterised, and dose levels and image quality are monitored.”

Paragraph 175, line 1761

Typo: “…tube potentials between 70 and 150 kV can been applied” – “can be applied”

Paragraph 177, line 1778

How is the iodine used in this way? Methodology unclear. Or the grammar/wording

Paragraph 182, line 1839

Grammar: “… method used that has long been used …” – “method that has long been used”

Line 1979

Typo: “allowing the tube current to fall to a level whether the image quality may be compromised”, change “whether” to “where”

Paragraph 206, line 2080

IED acronym used before definition

Fig 4.6 caption, line 2088

Typo: “and the virtual-non-contrast mages used to visualise”, change “mages” to “images”

Paragraph 223, line 2246

“(including technologists, medical physicist, radiologist and vendor application specialist)”

Again, should ‘technologists’ be ‘radiographers’ here for consistency?

Box 4.3, line 2256 (2nd dot point)

“…which should be keep up to date regularly).” Better: ‘kept up to date’ or ‘updated regularly’

Box 4.6, line 2388

The equivalent boxes 2.5 and 3.9 don’t have a prelude paragraph, but box 4.6 does (as does box 5.3). Is this inconsistency deliberate and justified?

Fig 5.1, line 2432

The axes may need more explanation as at face value they suggest that the public has more education and awareness than medical professionals. Perhaps it is intended to indicate the level of additional education and awareness required, however in that case vendors may have better awareness of issues related to dose and image quality than some medical professionals.

Fig 5.2

The order of the four words in the message is different for the picture and the caption (“simple, direct, clear, resonant” or “simple, direct, resonant, clear”)

Line 2455

Unclear what DRLs/median reference values are being referred to in this sentence:

“For example, setting of a DRL should be followed by reaudit and continual reassessment to determine whether the facility can achieve the median reference value.”

Suggest:

For example, the setting of a local or national DRL should be followed by reaudit and continual reassessment to compare the facility median with the DRL.

 

Paragraph 249, lines 2540-2541

“A recent study in the emergency setting …”, It would be nice to have the study referenced here.

Paragraph 291, lines 2872-2874

Why not compare operators by KAP and air kerma as the first preference, instead of time, and only use time if the other metrics are not available?

Box 5.1, line 2968

Suggest using “scan projection radiograph” rather than the collection of ‘scout’, ‘scanogram’, or ‘topoview’ (note, the term ‘surview’ has been missed from the collection)

Paragraph 306, line 2976

“… number or projections …”. Should be “of”

Lines 3023-3025

Should all females be questioned, or should allowance be made for not questioning for procedures of low radiation dose (as given in the IAEA Basic Safety Standards, GSR3, clause 3.176) ?

Suggest “in which the conceptus could be directly exposed”

Paragaph 315, line 3069

“…if the conceptus is exposed primarily to radiation.” Suggest “if the conceptus could be directly exposed.”, or “if the conceptus is in the primary beam.”

Paragraph 328, line 3176

Grammar. Suggested change: “In a position statement …”

Paragraph 340, lines 3307-3308 and box 6.1

Rather than avoiding pitches less than 1 altogether, would it be better to advocate for their cautious use for scanners that don’t maintain dose but instead increase dose for smaller pitches? If pitch is decreased on a scanner like this, dose could be compensatively reduced (rather than avoiding pitch less than 1 altogether).

Table A.1, line 4202

“… essentially equal …” instead of “equal” since they are not exactly equal (there is a miniscule build up and attenuation effect).

Paragraph B8, line 4268

Grammar: “transferred”

Line 4363

Add “tube” to definition

Line 4371

“of” instead of “od”

Line 4464

Add “computed”

Line 4486

Grammar: “has there are”, suggest “has sub-domains:”

Line 4636

Italicise “Annals of the ICRP” as in the above lines


Back