Objective
For the first time in France a comprehensive quality assurance (QA) protocol in the framework of a breast cancer screening campaign has been established in the Bas-Rhin region. The protocol takes into account the particularities of the local organization of medical services involved in such an initiative (radiological private practices exclusively).
The four aspects of the mammography chain addressed were production of the image (X-ray generator);
image recording (X-ray film, cassettes);
image process (processor, dark room conditions);
image visualisation (viewing boxes).
The data analysis is still underway; a set of preliminary results have been presented to the radiologists involved.
These results, including image quality evaluation and dosimetry, clearly demonstrated the absolute necessity for a well established quality control program within the context of a screening program and for strict compliance with its technical requirements. As far as dosimetry is concerned, entrance surface dose assessment for a 5 cm tick perspex phantom was performed in each radiological centre by using specifically calibrated for mammography themoluminescent dosimetrys (TLD). A very large range of doses was found for all the centres. In a more general way, problems were identified at different levels of the entire mammography chain. In conclusion this first evaluation enables a management framework to be set up to facilitate higher quality screening, to identify technical corrective actions, to stimulate and improve multidisciplinary team work and to disseminate information to all those who are responsible for training and educational initiatives.
Data collected on radiological equipment from the Bas-Rhin screening initiative, together with results from a Parisian X-ray hospital department, represented the starting point upon which the multipartners collaboration on QA Expert Systems (ES) designing was established.
Initially, the task was to focus on the analysis of ava ilable information concerning film processing equipment.
A comprehensive database was created, and an attempt was made to identify the influence of X-ray processors' parameter variations on image quality through data analysis. For the purposes of this analysis, image quality was defined with the aid of phantom or measuring instruments. Such an approach enabled a better understanding of how to correlate quality control test results with the corrective actions to be taken in order to infer abnormal situations.
This first generation expert system prototype will be limited to mammography technique.
A comprehensive technical quality control programme has been established within the framework of the Mersey Region's breast screening programme.
Data analysis protocols have been established so that multicentre, multiparameter analysis can be performed on the results. To date work has concentrated on establishing the mechanism for this analysis and a preliminary evaluation of the results obtained.
The quality control performance involves daily assessment of imaging performance using a Leeds TOR MAS phantom and automatic processor performance using sensitometric strip measurements.
To date a preliminary detailed analysis of the image quality results have been undertaken in order to relate this performance to the radiographic conditions. Effort has concentrated on high contrast resolution large detail threshold contrast (1 cm diameter) and small detail threshold contrast (0.25 mm diameter).
The high contrast limiting resolution performance was assessed for each breast screening centre. The mean value for resolution varied form 8.9 lp/mm to 12.5 lp/mm with standard deviations from 0.9 lp/mm to 2.01 lp/mm.
Observer variations were found to play a small role in the observed variation in resolution. Further analysis demonstrated that phantom to phantom variations were minimal.
In order to assess possible causes of variations in resolution at different locations the high contrast resolution both parallel to and perpendicular to the anode cathode axis was investigated in relation to focal spot size.
These results leads one to conclude that for focal spots greater than approximately 0,5 mm, high contrast resolution is governed by focal spot size for the geometry normally employed in mammography. For focal spot sizes less than 0.5 mm resolution is to a large extent governed by screen and film unsharpness, which was demonstrated by performing measurements with the test phantom in contact with the screen and film combination.
The high contrast resolution was also assessed as a function of film density at a single screening centre and thus screen and film combination. The results indicated a clear relationship between film density and observed resolution with results varying from 11.1 lp/mm at density 0.4 to 14.3 lp/m at density 1.2 followed by a slight fall to 12.0 at higher densities. These results indicate that film density does play some part in the high contrast resolution measurement.
Collaboration between Linkoeping University and the Royal Marsden Hospital resulted in the development of a Monte Carlo computer code which combined the most effective parts of the codes previously developed at both institutions.
The code calculates:
contrast (conventional imaging) and signal to noise ratio (digital imaging) at a specified position in the image plane (detector) and for a given contrasting detail;
the mean absorbed dose (energy imparted per unit mass) in the phantom;
contrast improvement factor (CIF), signal to noise ratio improvement factor (SIF) and dose increase factor (DIF) using an antiscatter grid or an airgap and for a particular contrasting detail.
The program has been made flexible to allow easy variation of input parameters such as: photon energy spectrum, beam filtration, phantom size and atomic composition, focal distance, entrance field area, position of interest in the image plane, grid parameters (including cover and interspace material), airgap length, thickness and compostion of contrasting detail and detector.
Data about commersially available grids have been obtained from the grid manufacturers Siemens, Philips, Mitaya, and Toshiba. In all, data on about 100 grids are available, some containing reports about measurements. The code has been tested against some of the measurements as well as against some measurements in the literature using airgaps.
A first preliminary run of the code has been made in a simulated paediatric procedure. 3 grids were compared. The SIF was calculated as a function of the tube potential is geometries with small (child) and large (adult) scattering volumes. With the large scattering volume (adult), the signal to noise ratio increased significantly when using the grid due to removed scattered radiation. In the child geometry, SIF was less than 1 when using the same grid. Scattered radiation was removed but the primary radiation was also attenuated contributing to reduced signal to noise ratio . The low ratio grid with cover and interspace material of carbon fibre had little effect on primary transmission and contributed to a SIF of more than 1 in the child geometry.
A computer program has been developed which models the use of antiscatter grids in diagnostic radiology.
The computerprogram developed reduces the amount of computer time needed for calculation of contrast, scatter or dose at a point by making use of the collision density estimator extended to allow for the calculation of scatter at a point.
Careful consideration was given to the choice of differential scattering cross sections. It proved to be unnecessary to use cross section which allow for interatomic interactions, the use of free atom cross sections allowing adequate accuracy in all cases.
The resulting computer program can estimate the scatter, contrast and signal to noise ratio at any point in the image plane when using a grid or airgap as an antiscatter technique. The program also calculates the energy imparted to the patient, simulated as a rectangular block of tissue whose size and composition can be readily varied. The programs have been structured so that all input parameters are easily varied including the test detail used for the contrast calculation, the grid parameters and the X-ray tube potential and the filter.
Antiscatter grids remove both primary and secondary photons from the radiation field which exposes the image receptor. This has been modelled using the method developed by Day and Dance (1983). The method uses an analytic rather than stochastic approach to calculate the average transmission and has important advantages in the computation time required to achieve a given statistical precision on any calculated quantity.
Software has been developed which calculates, by direct analogue simulation, the effects of secondary particles generated within the grid and grid covers.
Careful checks have been applied to all the codes developed.
Results show that the effect of secondary photons produced in the grid is to increase the secondary transmission for grids with an aluminium interspace by about 22%. Most of this contribution arises fro m scattered photons, the contribution of L-fluorescent photons from lead being negligible and the contribution of K-fluorescent photons from lead amounting to about 5% for photon energies well above the K-edge.
An electronic instrument is being developed for measuring waveform, tube potential (kVp), ripple, HVL, exposure time and exposure in the mammography energy range.
A probe with 2 couples of silicon PIN detectors having a sensitive surface of 7.34 mm{2} and 100 mm{2} respectively was built. The electronic needed to acquire the relevant signals was designed and implemented.
The 4 detectors in the probe have, in couples, different functions. Those having a wider area are for measuring kVp. The smaller detectors measure exposure above and under an 0.5 mm thick aluminium filter.
The tests for evaluating the performances of the device for the exposure measurement were carried out in 2 stages. The output signal of the smallest unfiltered detector, time integrated, was compared with the response of an ionization chamber for mammography. This analysis pointed out that the device, without any correction, is able to measure exposure or air kerma within +/- 7% in the mammography range 25 to 35 kVp. The detector was then embedded in the surface layer of a phantom having an X-ray transmission close to that of a standard breast. The electronic equipment was made portable. In this way it was possible to test the device and the associated electronics in actual mammography units.
The project deals with two important areas of Radiation Protection in medicine, namely the improvement of quality assurance (QA) in diagnostic radiology, and expert systems (ES).
Improvement of QA in diagnostic radiology :
This directly follows the Community wide trial carried out in 1988, checking the impact of the quality criteria in routine practice, developing new test procedures and more updated calculation codes which are better able to fulfil the proposed guidelines, and extending the quality criteria to further examinations, paediatric and digital radiology, etc.
Expert systems (ES) as a guide for reducing patient dose and optimizing image quality; there are three parts in this area :
Identification and definition of the applicable clinical domains (mammography, paediatric radiology, digital radiology). The ES will be conceived in a way that subsequent extension without any major revision will be possible.
Development of a package software for the assessment of the patient dose according to current radiological techniques. Implementation of a system for automatic data acquisition.
Selection of an adequate knowledge acquisition tool, implementation and revision of a prototype.
The work of the project is based on the collaboration of all coordinated partners together with prolonged partners:
CAATS-INSERM together with USL number 7 and the IRS Ltd are being involved in the expert system approach for QA in mammography, computed tomography and digital radiology;
the University of Madrid is contributing to the analysis of a series of relevant parameters influencing both patient dose and image quality;
the University of Linkoping together with the Royal Marsden Hospital in London are mainly concerned with Monte Carlo calculations with a view to evaluating the optimal choice of the grid under various scatter conditions. Theoretical considerations in this area might be used as a part of the Expert System;
the University of Ferrara is designing a real time instrument which will be useful for quality control in mammography. This could be checked either within the work programme of the CAATS;
INSERM project or the common Greek one of the University of Patras and Anti-cancer Hospital;
Each partner will be involved in the development of a framework which will serve as a guideline for the optimization of medical radiation protection either in adult radiology or in paediatric radiology: already established links with the group of the University of Munich will be particularly reinforced.
Fields of science (EuroSciVoc)
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques. See: The European Science Vocabulary.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques. See: The European Science Vocabulary.
- natural sciences computer and information sciences data science
- medical and health sciences clinical medicine radiology
- natural sciences computer and information sciences artificial intelligence expert systems
- medical and health sciences clinical medicine oncology breast cancer
- natural sciences physical sciences theoretical physics particle physics photons
You need to log in or register to use this function
Programme(s)
Multi-annual funding programmes that define the EU’s priorities for research and innovation.
Multi-annual funding programmes that define the EU’s priorities for research and innovation.
Topic(s)
Calls for proposals are divided into topics. A topic defines a specific subject or area for which applicants can submit proposals. The description of a topic comprises its specific scope and the expected impact of the funded project.
Data not available
Calls for proposals are divided into topics. A topic defines a specific subject or area for which applicants can submit proposals. The description of a topic comprises its specific scope and the expected impact of the funded project.
Call for proposal
Procedure for inviting applicants to submit project proposals, with the aim of receiving EU funding.
Data not available
Procedure for inviting applicants to submit project proposals, with the aim of receiving EU funding.
Funding Scheme
Funding scheme (or “Type of Action”) inside a programme with common features. It specifies: the scope of what is funded; the reimbursement rate; specific evaluation criteria to qualify for funding; and the use of simplified forms of costs like lump sums.
Funding scheme (or “Type of Action”) inside a programme with common features. It specifies: the scope of what is funded; the reimbursement rate; specific evaluation criteria to qualify for funding; and the use of simplified forms of costs like lump sums.
Coordinator
75019 Paris
France
The total costs incurred by this organisation to participate in the project, including direct and indirect costs. This amount is a subset of the overall project budget.