1st CT users group meeting: 29/09/1999

The 1st meeting of the CT Users Group was held in Birmingham on 29/09/1999. The programme is shown below.

Please note: information provided in the slides is not peer-reviewed, is for educational use only and is explicitly not to be used for sales or marketing purposes. Any of the authors can be contacted, via the CTUG if no contact information is provided in the slides, to discuss the contents.

Meeting Programme

10:10 +/- Reference dosimetry for computed tomography - Paul Shrimpton, NRPB

The concept of reference doses is recognised as a useful and practical way of promoting optimisation of patient protection for diagnostic medical exposures. A robust methodology has been developed for establishing reference doses for the peculiar conditions of exposure in CT. This initiative represents an essential component of the European Guidelines on Quality Criteria for Computed Tomography published by the European Commission (EUR 16262, 1999) and available on the Internet (http://www.drs.dk/guidelines/ct/quality).
The reference dosimetry is based on measurements in standard head and body CT dosimetry phantoms of the computed tomography dose index made with a pencil ionisation chamber (CTDI100) and expressed in terms of absorbed dose to air. Appropriate combination of measurements made centrally and peripherally yields the weighted CTDI (CTDIw, mGy). This characterises the conditions of exposure for a single slice in serial scanning or per rotation in helical scanning, while the dose-length product (DLP, mGy cm) takes due account of the scope of serial or helical scanning in a complete examination.
Reference dose values for these two quantities are intended to represent thresholds to trigger investigations by radiology departments where typical practice, as indicated by mean values of the dose descriptors observed for representative groups of patients, is likely to be well away from the optimum. Initial reference dose values for some common procedures on adult patients have been set as the third quartiles of survey data from the UK and from a European pilot study of the quality criteria.
This practical system of dosimetry is also being further developed to allow comparison of dose during critical review of CT practice involving paediatric patients.
In order to provide support to the wide-scale implementation of reference dosimetry for CT, a database is being established of standard dosimetric data for a wide range of scanners. This will include normalised dose data (mGy (mA s)-1) from measurements of CTDI in air and in the standard dosimetry phantoms for different models of CT scanner. The database will be accessible via the Internet at : http://www.efomp.org.

10:30 Survey of patient effective doses and dose length product in CT - Arnold Rust, Royal Surrey Hospital

10:50 +/- HVL measurements to facilitate patient dose assessment for newer CT scanners using published normalised dose data - Anne Hill, Princess Margaret Hospital, Swindon

The National Radiological Protection Board (NRPB) method of calculating patient doses from CT can be applied to scanners not represented in normalized dose data sets, published in NRPB Report R250, if an appropriate existing data set is selected. A method of matching scanner models to data sets is currently being developed in a national scanner survey, organized by the Department of Health ImPACT CT evaluation group. This paper demonstrates the use of an alternative method of data set selection to that employed in the national study, by application to the Picker PQCT and Siemens +4 scanners. Half value layers (HVLs) evaluated at the isocentre and at intervals within the fan angle of the X-ray beam of either the Picker PQCT or Siemens +4 were compared with HVLs for three scanner models represented in the NRPB data sets. A good match indicates the suitability of a normalized dose data set, since data sets are dependent on the quality of the X-ray beam incident on a patient. The results demonstrate that Picker 1200SX data sets could be used with the Picker PQCT scanner. The most likely match for the Siemens +4, from the limited number of scanners investigated, was identified as being the Philips Tomoscan LX.

11:25 +/- Principles of multi-slice CT - Maria Lewis, ImPACT

The last big impact of technology on CT scanning occurred in 1989 with the introduction of helical CT into clinical practice. This was followed in 1991, by the launch of the CT Twin, the first 3rd generation dual detector bank scanner which at the time failed to make a great impression on the CT market. In 1998, four of the major CT manufacturers built on the experiences with this early multi-slice model and launched scanners capable of imaging four slices simultaneously. It appears that these models will herald big changes in CT practice in the near future.
The four new multi-slice models feature different designs of detector banks, based either on an adaptive array or matrix array principle. They also have varying specifications in terms of minimum scan time, range of available slice widths and maximum length imaged in a single rotation.
The concept of helical pitch must be reconsidered for multi-slice systems as two different definitions of this quantity have been adopted by the manufacturers. One definition utilises the nominal imaged slice width, whereas the other is based on the irradiated dose profile.
To make most efficient use of the data collected by the multi-array detector banks, interpolation algorithms have been extended beyond the 360° and 180° linear interpolations used commonly in single detector array models. Some manufacturers also recommend the use of non-integer pitch values to further improve image quality.
The effect of different pitches on image quality and dose needs to be evaluated, as with multi-slice systems there is no longer a straightforward relationship between pitch and image quality. Choice of pitch and imaged slice width will be dependent on the clinical application.
In clinical use, multi-slice systems can be utilised to advantage in the scanning of large volumes in a short time. Alternatively, certain applications may benefit from scanning with narrow slice widths in the same time as with wide slices on conventional systems, and without encountering tube cooling delays. The faster rotation times available on these systems also open up possibilities in the field of cardiology. Finally, the capability of using narrow slice widths routinely results in improved image quality for applications involving 3-D reconstructions.

11:50 Business meeting

13:50 +/- Should diagnostic reference levels (DRLs) be adopted for dental CT scans? - Anthony Reynolds, Image Diagnostic Technology Ltd

Permanent dental implants offer a stable, comfortable and enduring alternative to removable dentures, but the surgical procedure demands considerable skill and detailed planning. Pre-operative imaging informs the quantitative assessment of jaw bone thickness and density at potential implant sites, and helps the surgeon avoid sensitive structures such as nerves. Reformatted computed tomography (CT) scans have become the imaging modality of choice for pre-implant assessment, because the non-distorted images are suitable for direct measurement, and the bone density can be estimated directly from the CT numbers. Over a thousand CT scans are carried out in the UK annually for pre-surgical planning prior to the placement of dental implants.
For many years the main disadvantage of CT was considered to be a relatively high radiation dose to the patient. One reason for this perception was that the earliest dosimetry studies were based on outdated equipment (by today’s standards) whereas significant improvements in CT Scanner design, beam collimation, and detector efficiency have recently taken place. Another reason is that doses from dental CT scanning can vary by a factor of 10 or more from one hospital to another. This is in part because the lower dose limits at which images become clinically unacceptable are not well understood.
Image Diagnostic Technology Ltd (IDT) provides a commercial service to dental implantologists by arranging for patients to be CT scanned at participating hospitals, retrieving the computer data, and generating reformatted images and 3-D views. To manage the variability in patient doses IDT establishes strict scanning protocols to which participating hospitals adhere. To optimise these protocols IDT obtains Effective Doses (EDs) for each scanner model, following methodologies developed by the ImPACT group.
The ED is calculated from measured CT Dose Index (CTDI) values using the CTDOSE software and the NRPB SR250 datasets, with matching of new scanners to the existing datasets done by ImPACT. Results obtained to date show that adequate image quality for surgical planning can be achieved for EDs of 0.1 - 0.3 mSv for a typical maxilla and 0.2 - 0.6 mSv for a typical mandible, for a range of scanner models. These numbers could form the basis for DRLs for dental CT scanning.

14:15 +/- An evaluation of SmartHelical ™ software on the IGE CTi Advantix - Alexis Moore, Leeds General Infirmary

The IGE HiSpeed CT/i scanner has several patient dose reducing features. SmartHelical™ is the latest of these features and is a new piece of software which IGE claim allows the patient dose to be reduced by 20-30% whilst maintaining image quality. When SmartHelical™ is selected the CT machine scans a slightly longer section than is defined by the operator, and the effective slice thickness is increased by 10-20%. An improved reconstruction algorithm is used to improve the image quality for a given tube current. The General Infirmary at Leeds obtained a copy of the software on “flex trial” in order to evaluate the software clinically and to verify the predicted dose reduction. Noise measurements were made using one of the phantoms which was available with the scanner. Data was acquired using a water section for the different tube voltages, slice thicknesses, pitches and software filters using varying mAs values in both SmartHelical™ and standard Helical modes. A second set of data was collected for Teflon, plex and polystyrene with varying slice thicknesses and mAs values in both modes. An exponential decay curve was fitted to each data set, and the value of the mAs where the curve reached a plateau determined. The noise was determined for this mAs in Helical mode and the value of the mAs with a similar amount of noise in SmartHelical™ mode was determined. This enabled the prediction of the amount of mAs reduction achievable in SmartHelical™ mode whilst maintaining image quality.

14:35 +/- A comparison of fluoroscopic CT Systems - Nick Keat, ImPACT

CT fluoroscopy first appeared as an option on a commercially available CT scanner in 1996. Since then, most CT scanner manufacturers have introduced or are developing capabilities to allow their systems to produce near real time, continuously updated images, analagous to conventional fluoroscopy. Currently, the main applications for these systems are in two areas:
Biopsy and other interventional techniques such as lesion drainage.
As a method to synchronise the start of conventional CT imaging runs with optimal contrast agent enhancement.
This talk will identify the advantages, and problems associated with fluoroscopic CT. The image reconstruction process will be described. A comparison of differences between the manufacturers' systems will be made. The doses to both the patient and the operator (usually a radiologist) will be evaluated, using data from manufacturers, published papers and measurements from ImPACT’s assessment program. These doses will be put into perspective with other interventional radiological techniques, and compared to statutory dose limits, as well as thresholds for physiological damage.

14:55 CT scanner dose survey update - Sue Edyvean, ImPACT

15:15 Round table session