Development of image quality related reference doses called acceptable quality doses (AQD) in paediatric CT exams in Qatar

Abstract

Objectives To describe first experience of integrating assessment of image quality in paediatric X-ray computed tomography (CT) with analysis of the radiation dose indices to develop reference doses called acceptable quality dose (AQD). Methods Image quality was scored by the radiologists at a tertiary care hospital in Qatar on a scale of 0 to 4 using the recently published scoring criteria. The patients undergoing head, chest and abdomen CT were divided in different weight groups as follows: < 5 kg, 5–< 15 kg, 15–< 30 kg, 30–< 50 kg, 50–< 80 kg and > 80 kg. The images that were clinically acceptable (score of 3) were included for assessment of median values of CTDIvol and DLP to obtain AQDs in different weight groups. Results After initial training in image quality scoring of CT images of 49 patients by three radiologists, the study on 715 patients indicated 665 studies (93%) were clinically acceptable as per scoring criteria. The median CTDIvol values for the above weight groups were 16, 20, 22, 22, 27 and 27 mGy and the median DLP values for these weight groups were 271, 377, 463, 486, 568 and 570 mGy cm, respectively, for head CT. Similar values are presented for chest and abdomen CTs. Conclusions The first ever experience of starting with image quality assessment and integrating it with analysis of dose indices to obtain AQD values shall provide a workable model for others and values for comparison within the facility and in other facilities leading to optimisation. Key Points • The first study to integrate image quality assessment with analysis of patient dose indices shows feasibility for routine practice in other centres. • The values of acceptable quality dose (AQD) were provided for head, chest and abdomen CT of children divided into weight groups rather than age. They shall act as reference values for future studies. • Verification of our findings on proportional increase in exposure parameters (CTDIvol and DLP) with weight by other investigators shall be helpful.

Paediatric diagnostic reference levels for common radiological examinations using the European guidelines

“…Is has been suggested that image quality should be part of the DRL system and a so-called acceptable quality doses should be derived. 15 This system was later clinically used and evaluated 16 using image quality scoring criteria developed for paediatric CT. 17 With this approach, doses from patient examinations with suboptimal image quality are excluded before the typical values are derived.…”

Revising and exploring the variations in methodologies for establishing the diagnostic reference levels for paediatric PET/CT imaging

“…For standard procedures, it is not anticipated that these levels will be exceeded when good and conventional practice in terms of diagnostic and technical performance are used [7]. DRL methods are advisory practices and remain opposed to the dose-limit concept and it aims to minimise radiation doses to patients to acquire optimum image quality [22,23]. DRL can be performed at three levels: local DRL (LDRL), national DRL (NDRL), or regional (RDRL) [7,24,25].…”

Five-star rating system for acceptable quality and dose in CT

“…The 5-star rating for image quality (Tables 1 and 2) differs from previous linear, 3-point scoring for noise and acceptable spatial resolution (optimum; too much; too little) and 4-point scoring for diagnostic acceptability (1: fully acceptable; 2: probably acceptable; 3: acceptable only for limited conditions; 4: unacceptable) such as the one proposed as European Quality Criteria [10]. Conversion of a point scale to a 5-star rating goes with the intent of recent publications [6,7]. The non-linear 5-star rating of diagnostic quality rewards top rating to the clinical indication-based personalized protocols with "just adequate" diagnostic quality and, by extension, radiation doses such as low-dose CT for nodule follow-up, kidney stone evaluation, or paranasal sinus CT before an endoscopic surgery.…”

“…The situation calls for a black box, artificial intelligence (AI) models to take over the task of selecting the right dose for the right quality [5]. While that may be thought for future, recent studies have used subjective assessment on an institutional level so as to have an acceptable quality dose (AQD) [6,7]. AQD can be personalized to body habitus, body region, scanner type, clinical indication, and quality and dose requirements.…”