RPS 1113 - Improving quality to build safety

Purpose:

Despite the clinical benefits of computed tomography (CT), there are relevant concerns associated with radiation exposure to patients. Novel spectral shaping usually represented by a tin-filter (Sn) have already shown improved dose efficiency in chest and abdomen CTs. This study aims to investigate the feasibility of an Sn-filter in head CTs and to evaluate possible dose reduction.

Methods and materials:

This retrospective study involved 20 patients (mean age 55.9±18.3 years). 12 patients underwent temporal bone CTs using conventional 120kVp-protocol (220 reference mAs) while 8 were scanned using 130kVp-Sn-protocol (420 reference mAs). Objective image quality for each patient was evaluated by the signal-to-noise ratio (SNR), calculated as the mean signal divided by the noise measured in the internal auditory canal (IAC), the external auditory canal (EAC), and squama temporalis (ST). Subjective image quality was evaluated by two experienced radiologists using a 5-point Likert scale (1=excellent, 2=good, 3=moderate, 4=fair, and 5=non–diagnostic). The radiation dose was assessed by the mean of CTDIvol values collected from patient dose reports. Finally, the differences in SNR and radiation dose between investigated protocols were checked for statistical significance using unpaired t-tests.

Results:

The mean CTDIvol for the protocol with an Sn-filter was significantly lower than the one for the conventional protocol (26.8±2.5mGy vs 38.3±4.8mGy, p<0.001). SNR comparison has shown no significant difference between the two protocols (p=0.18, 0.11, and 0.35 for IAC, EAC, and ST, respectively). Subjective image quality assessment has shown substantial inter-reader agreement (79%) with moderate reliability (k=0.45). For both protocols, image quality for all the cases was rated as either “good” or “excellent”.

Conclusion:

The study indicated that Sn-filtration in temporal bone CT allows for substantial dose reduction without compromising diagnostic image quality.

Limitations:

The small cohort.

Ethics committee approval

Approved by the local ethics committee.

Funding:

No funding was received for this work.

Purpose:

The directive 2013/59/EURATOM establishes the need to establish quality assurance programmes and the assessment of patient dose, in particular, in practices such as interventional radiology (IR) and cardiology (IC). In the framework of the European project MEDIRAD, we have developed a method for organ dose and peak skin dose monitoring of patients undergoing IR/IC procedures based on the accelerated Monte Carlo code, MC-GPU.

Methods and materials:

MC-GPU has been validated by comparing the simulations with experimental measurements using TLD using regular shaped phantoms made of tissue-equivalent materials under realistic conditions. In addition, several tools have been developed to accelerate the patient organ dose calculation for a complete IR/IC procedure. On the one hand, we have prepared an interface between radiation dose structured reports and the MC-GPU code. The tool automatically generates the input files needed for the simulation. The patient is represented by the voxelised anthropomorphic phantom “REX” (Schlattl H et. al. 2007). Moreover, a graphical interface has been implemented to have a 3D colour-map of the patient’s dose in the different tissues.

Results:

The results show good agreement between the simulations and the experimental measurements. Differences between measured and calculated doses for the different experimental configurations were, in general, lower than 10%. The computational time needed for the simulation of a single irradiation event is a few seconds using 2 GPU cards (GeForce GTX 1080Ti).

Conclusion:

It has been verified that MC-GPU is a good tool for the assessment of the absorbed dose in the different organs and for the maximum skin dose determination in patients undergoing an IR procedure.

Limitations:

n/a

Ethics committee approval

n/a

Funding:

The Euratom research and training programme 2014-2018 under grant agreement No 755523.

Purpose:

To determine the probability of receiving a high radiation dose from single or multiple CT examinations and to assess the potential impact by evaluating clinical indications for multiple CT examinations and associated prognosis of high-dose patients.

Methods and materials:

All CT examinations performed at our institution between 01/07/2013-01/07/2018 for which radiation dose data was available were included. The effective dose received by single or multiple CT examinations was assessed for each patient using dose monitoring software. The probability of receiving a high dose (i.e. ≥50mSv (single) or ≥100mSv (multiple)) was evaluated using survival analysis for women, men, different age categories, and the total number of CT examinations received. Baseline characteristics and the probability of survival up to 4.5 years were assessed for high-dose patients, taking clinical indication into account.

Results:

An effective radiation dose was available for 100,672 CT examinations performed amongst 49,978 patients, 649 of whom received a high radiation dose. The estimated high radiation dose probability was 0.43% for a single examination and 1.5% (men) to 1.9% (women) for repeated examinations over a period of 4.5 years. Factors increasing this probability are female sex, especially in the age category 51-64 (probability=3.02%), and >6 CT examinations (probability>16%). The impact on survival is limited because high-dose patients are often severely ill: 65% of high-dose CT examinations were oncology-related and 48% of high-dose patients died within 4.5 years from pre-existing medical conditions.

Conclusion:

The probability of receiving a high radiation dose from CT examinations is small but not negligible. High-dose patients often had pre-existing conditions with a high risk of premature death. Consequently, the impact of a high radiation dose is expected to be small on a population level.

Limitations:

The data originates from a single centre.

Ethics committee approval

The study was approved and written informed consent was waived.

Funding:

No funding was received for this work.

Purpose:

To investigate the water equivalent diameter (WED) and size-specific dose estimates (SSDE) for paediatric head computed tomography (CT) examinations on two CT devices.

Methods and materials:

Paediatric brain CT examinations from one year were investigated for two Siemens devices, with one located at the emergency (ER) department (CT1 and CTER). The WED of the middle slice of the CT series was determined and used to calculate the SSDE according to AAPM report 293. The paediatric patients were categorised into 5 age groups (A: 0-1y, B: 1-5y, C: 5-10y, D: 10-15y, and E: 15-18y). A Mann-Whitney test was used to compare the WED and SSDE of the age groups for both scanners.

Results:

At CT1 and CTER, respectively, 128 and 207 scans were carried out. The correlation between patient age and WED was significant (CT1: r2=0.73, CTER: r2=0.71). The mean WED and SSDE of the age groups CT1 were A: 14 cm, 16 mGy; B: 16 cm, 16 mGy; C: 17 cm, 19 mGy; D: 18 cm, 21 mGy; E: 18 cm, 29 mGy. For the same age groups, the WED was not significantly different (p>0.05) between the two devices. For the age groups up to 10 years (A-C), the SSDE was also not significantly different (p>0.05), however, SSDE differed significantly (p<0.05) for the age groups above 10 years (D-E), with a mean SSDE of 24 mGy and 31 mGy for CTER.

Conclusion:

Several European documents suggest the use of patient weight rather than age for dose investigation purposes. As weight may not be available at the moment of dose monitoring, SSDE is a valid alternative. It can be calculated and scaled with CTDIvol (and therefore attenuation). We provide the first set of reference values.

Limitations:

No access to the clinical indication of the head CT examinations.

Ethics committee approval

n/a

Funding:

PhD grant from VLAIO [Grant No. HBC.2016.0233].

Purpose:

To compare the small-size detail detection of combined modalities DBT+S (digital breast tomosynthesis and synthetic 2D) and DBT+FFDM (DBT and conventional full-field digital mammography) for mammography systems of different models and vendors.

Methods and materials:

15 different mammography systems of 9 different models were compared. We used a home-made phantom composed by CDMAM (Artinis), homogeneous PMMA slabs, and BR3D tissue-equivalent slabs (Cirs) containing an anatomic noise in order to obtain different phantom configurations equivalent to 32, 60, and 90 mm breast thicknesses.

The CDMAM phantom was composed of an aluminium slab with gold details of different diameter and thickness. For this study, we focused only on small details ranging from 0.5 to 0.1 mm that simulate microcalcifications.

We exposed at 2 different dose levels (one obtained with a clinical automatic exposure programme and the other with manual “fixed” parameters) and the processed images were successively divided into squared sub-images, each containing a detail of different size and contrast. The sub-images were rotated and displayed in random order on a 5MP calibrated monitor to 4 trained readers.

The percentage correct rate for DBT+S and DBT+FFDM was evaluated.

Results:

As an example, the percentage correct rate for 4 different systems for details 0.4 mm and 0.1 mm is shown.

Conclusion:

For details greater than 0.4 mm, the combined modalities DBT+S and DBT+FFDM show a similar detectability in all the systems. For smaller details, the difference in the detection between the 2 combined modalities depends on the system.

Limitations:

The use of a physical phantom, composed of aluminium and gold details, could affect the reconstruction algorithm for DBT and s-2D.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

To use a software package designed for the automation of routine quality control (QC) tests in digital mammography to detect phantoms with manufacturing defects.

Methods and materials:

24 Leeds TORMAS phantoms to be employed in routine tests of mammography units used within a regional screening programme were preliminarily tested to exclude the possible presence of defects. Two images of each test object were acquired using the same digital mammography equipment, one in a manual exposure mode to enhance possible differences while minimising variability due to the imaging system itself and one in an automatic exposure mode (AEC) with the phantom on the top of a 35 mm polymethyl-methacrylate (PMMA) stack to reproduce the same conditions as in the routine use. 15 image quality indices (IQIs) automatically calculated by the software tool used for routine tests were selected to evaluate phantom homogeneity and detect possible defects. Outliers were computed for each IQI and their impact on phantom homogeneity analysed.

Results:

8 outliers were found in 5 different IQIs from the images acquired in the manual mode associated to 7 phantoms. The impact of 4/7 outliers (57%) was mitigated in the images acquired with the PMMA in the AEC mode as used for routine tests; the 4 phantoms generating those outliers were thereby assumed to be acceptable. The 3 phantoms including outliers measurable from both datasets (without and with PMMA) were considered defective and were replaced with new phantoms.

Conclusion:

Software tools employed for routine QC tests can be successfully used to test phantom quality and detect possible manufacturing defects.

Limitations:

The limited number of images acquired per phantom due to time constraints.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

In Germany, a specific qualification is required to get permission to justify medical exposure in accordance with German regulations to diagnose and applicate x-ray. For this, specific courses on radiation protection are legally mandatory. One focus is on the correct use of equipment for radiation protection and justifying medical exposure. For comparison, the legal bases of the course concepts in Switzerland and Austria were also examined.

Methods and materials:

Investigators attended all mandatory courses on radiation protection for their entire length (basic, advanced, and up-to-date refresher courses). A total of 15 different courses from 6 different organisers were evaluated. Structured interviews with course organisers were conducted. The participants’ prior and gained knowledge during the attended courses was determined separately by multiple-choice questions. Structured checklists were developed, assuring consistent evaluations of the courses. The investigator’s methods and checklists were developed and affirmed by experts in workshops. Representatives of professional societies were interviewed.

Results:

Following the evaluation, it was concluded that the quality of knowledge transfer was divergent. An increase of knowledge could always be confirmed statistically, but often the clinical relevance and up-to-date contents were missing. All investigated courses complied with the mandatory content guidelines. The compulsory up-to-date refresher courses were found to be refreshing basic knowledge mostly instead of presenting new developments and technologies in radiation protection.

Conclusion:

To achieve a more appropriate learning situation in the courses, proposals are made, including a new target-group-specific course configuration with reduced attendance times. A specific course for radiation safety officers similar to Austria’s concept is suggested. Subject catalogues for the redesigned courses were created and could be used for harmonisation throughout Europe.

Limitations:

Only courses on radiation protection in Germany were evaluated.

Ethics committee approval

n/a

Funding:

Funding by Bundesamt für Strahlenschutz.