The research focus of the current paper are modern algorithms for solving the problem of automatic diagnostic of thorax diseases based on X-ray images. Special attention was paid to image preprocessing, classification together with calculation of features and their comparison in terms of efficiency. Approaches mentioned in this paper are used for development of new algorithm for automatic diagnostic of medical images.

Objectives: Cone beam CT (CBCT) in dentistry and maxillofacial surgery is a widely used imaging method for the assessment of various maxillofacial and dental pathological conditions. The objective of this study was to summarize the results of a multinational retrospective–prospective study that focused on patient exposure in this modality. Methods: The study included 27 CBCT units and 325 adult and paediatric patients, in total. Data on patients, clinical indications, technical parameters of exposure, patient dose indicator, or, alternatively, dose to phantom were collected. The dose indicator used was air kerma–area product, PKA. Results: In most scanners operators are offered with a variety of options regarding technical parameters, especially the field of view size. The median and the third quartile value of PKA for adult patients in 14 different facilities were 820 mGy cm² and 1000 mGy cm² (interquartile range = 1058 mGy cm²), and 653 mGy cm² and 740 mGy cm² (interquartile range = 1179 mGy cm²) for children, as reported by four different institutions. Phantom dose data were reported from 15 institutions, and median PKA ranged from 125 mGy cm² to 1951 mGy cm². Median PKA values varied by more than a 10-fold between institutions, mainly due to differences in imaging protocol used, in particular field of view and tube current-exposure time product. Conclusions: The results emphasize the need for a cautious approach to using dental CBCT. Imaging only when the clinical indications are clear, accompanied with the appropriate radiographic techniques and the optimum imaging protocol, will help reduce radiation dose to patients.

In recent years, starting from 2001, conventional film-screen radiography has been rapidly replaced by digital (mostly by computed radiography, CR) techniques in Estonia. New strategies for optimisation and quality assurance for digital radiography have been introduced by the DIMOND III and SENTINEL partners recently. It includes consideration to diagnostic requirements of a given clinical situation, but also objectivation and standardisation of image quality, e.g. using CDRAD test phantom, and constancy testing. The aim of this work is to evaluate the performance of an automatic exposure control (AEC) at different sensitivities when used with the Fujifilm CR system. Image quality, using threshold contrast-detail detectability (TCDD) and signal-to-noise ratio squared related to dose (SNR2/D), was also investigated for different sensitivities and tube potential values for this combination. Based on the image evaluation data, optimum speed modes for the organ programme settings have been proposed.

The University of Tartu provides a quality control service to the majority of diagnostic X-ray departments in Estonia. Its methodology has been adopted from the IEC and other relevant standards. Recently the Testing Centre of the University of Tartu was accredited on this methodology by ISO/IEC 17025. Besides the implementation of the quality management system, participation in interlaboratory comparison (ILC) was one of the prerequisites for the accreditation. Tests for estimating reproducibility of tube voltage and dose rate, accuracy of the voltage and accuracy of exposure time were carried out on a diagnostic X-ray unit in the Radiation and Nuclear Safety Authority in Helsinki. The measurement performance was judged by calculating deviation E(n) normalised with respect to the stated uncertainties. E(n) values for all tests were less than unity and by the common ILC criteria the testing performance could be considered as acceptable.

No abstract