Search

In recent years, there has clearly been a trend towards widespread dissemination of scientific results and achievements by sharing them in the form of educational materials. The application of the results of a research project in the education and knowledge development of students, young researchers and scientists, brings an added value to the outcomes of the scientific project. This work aims at contributing to the global process of free access to scientific materials by sharing the experience of a research group from the Medical University of Varna (MUV) in systematizing the results of three research projects, their transformation into educational resources and their application in innovative courses. The consolidated and systematized results of a research project are valuable assets for development of innovative educational materials. These resources, uploaded in a shared online repository, contribute to strengthen the international scientific community networking.

The processing of inputs in the first two layers of the probabilistic neural network (PNN) is highly parallel which makes it quite appropriate for hardware implementations with FPGA. One of the main inconveniences however remains the implementation of the nonlinear activation function of the pattern layer neurons. In the present study, we investigate the applicability of three approximations of the exponential activation function with look-up tables of different precision and the effect this has on the training process and the classification accuracy. Furthermore, seeking for a highly-parallel hardware-friendly algorithm for the automated adjustment of the spread factor $$\sigma _i$$, we investigated the performance of fifteen PNN training setups, which are based on the differential evolution (DE) or unified particle swarm optimization (UPSO) methods. The experimental evaluation was performed following a common experimental protocol, which makes use of the Parkinson Speech Dataset, as this research aims to support the development of portable medical devices that are capable to detect episodes with exacerbation in patients with Parkinson’s disease. The performance of the most successful setups is discussed in terms of error rates and from the perspective of the resources required for an FPGA-based implementation.

Purpose We present the development and the current state of the MaXIMA Breast Lesions Models Database, which is intended to provide researchers with both segmented and mathematical computer-based breast lesion models with realistic shape. Methods The database contains various 3D images of breast lesions of irregular shapes, collected from routine patient examinations or dedicated scientific experiments. It also contains images of simulated tumour models. In order to extract the 3D shapes of the breast cancers from patient images, an in-house segmentation algorithm was developed for the analysis of 50 tomosynthesis sets from patients diagnosed with malignant and benign lesions. In addition, computed tomography (CT) scans of three breast mastectomy cases were added, as well as five whole-body CT scans. The segmentation algorithm includes a series of image processing operations and region-growing techniques with minimal interaction from the user, with the purpose of finding and segmenting the areas of the lesion. Mathematically modelled computational breast lesions, also stored in the database, are based on the 3D random walk approach. Results The MaXIMA Imaging Database currently contains 50 breast cancer models obtained by segmentation of 3D patient breast tomosynthesis images; 8 models obtained by segmentation of whole body and breast cadavers CT images; and 80 models based on a mathematical algorithm. Each record in the database is supported with relevant information. Two applications of the database are highlighted: inserting the lesions into computationally generated breast phantoms and an approach in generating mammography images with variously shaped breast lesion models from the database for evaluation purposes. Both cases demonstrate the implementation of multiple scenarios and of an unlimited number of cases, which can be used for further software modelling and investigation of breast imaging techniques. The created database interface is web-based, user friendly and is intended to be made freely accessible through internet after the completion of the MaXIMA project. Conclusions The developed database will serve as an imaging data source for researchers, working on breast diagnostic imaging and on improving early breast cancer detection techniques, using existing or newly developed imaging modalities.

Anthropomorphic phantoms are models of real or virtual parts of the body, organ or tissue, represented by tissue-equivalent materials that aim to provide a realistic and accurate representation of their anatomy and properties. The aim of this study is to evaluate experimentally the suitability of 3D printed materials in the production of both, physical breast phantoms and abnormalities, to be used in optimization tasks in breast imaging. For this purpose, we designed three computational breast models, composed of skin, duct tree, adipose compartments and lesions. Subsequently, they were printed by using two 3D printing technologies and different printing materials, which were previously studied in details. The physical phantoms were scanned at a mammography machine, which allows 2D and 3D mammography (tomosynthesis) modes. The images were evaluated from an experienced radiologist. The results showed that tomosynthesis images are characterized with better realism compared to 2D mammography images. Next step is improvement in the printing quality of tumour formations as well as quantitative evaluation of the obtained results.

A web-based interactive platform dedicated to developing and visualizing regional health profiles of Bulgaria is presented in this paper. The platform is designed to provide simplicity and ease of use for the user, while accumulating and presenting vast amounts of data. In order to achieve that choropleth maps and clean design are considered. The platform’s back-end is developed with the high-level Python web framework Django, while the front-end is implemented with HTML, CSS and JavaScript. The platform will be further updated and exploited in the routine research work of experts in the field of public health as well as made available to the general public.

In the present work, we investigate the behavior of the Locally Recurrent Probabilistic Neural Network (LRPNN) with different activation functions in the recurrent layer neurons. Specifically, we evaluate the performance of the modified activation function proposed here, which belongs to the family of Rectified Linear Units (ReLU), and compare it with other ReLU-based functions, the traditional sigmoid activation function, as well as with the Swish and E-Swish activation functions. Furthermore, we investigate the efficiency of a training procedure which simultaneously adjusts the spread factor sigma and the weights in the recurrent layer of the LRPNN. This training helps for coping with practical tasks, such as the recognition of Parkinson condition from speech signals, which operate under limited amount of training data.

The use of anthropomorphic phantoms finds an increased application in the last years. From their use in virtual studies to assessment of new imaging technologies and reconstruction algorithms. The current study proposes a resource saving approach, in terms of time and money for assessing the creation of an anthropomorphic model by means of 3D printing. A four component object consisting of segmentations of adipose, gland, skin and lesion tissues was created from an MRI image set. From the resulting object two computational breast models were created – one with the elemental composition of the real breast tissues and a second one with the elemental composition of the low cost 3D printing materials planned to be used for the phantom manufacturing. Then, an in-house developed software was used to generate mammography images, which were subjected to both visual assessment and profile comparison. In this study we used nylon and acrylonitrile butadiene styrene 3D printing materials to represent the x-ray properties of the different breast tissues. The results showed that these 3D materials well represent the x-ray absorption characteristics of both glandular, adipose and skin tissues, while further investigation on suitable materials for representing the lesion characteristics is needed.

This paper presents a web-based platform dedicated to evaluation studies in x-ray breast imaging. The backend development is based on ASP.NET and C#, while the frontend is implemented under HTML, CSS and JavaScript. The platform is designed to own simple and intuitive graphical user-friendly interface. Questions to be evaluated are entered by the users and stored in the application database. In the same database, two other features are kept as well: the results of the evaluation and the regions of interest taken from patient and simulated mammography images. The content of the database is subject to modification, as well as further addition of data. The designed web-based platform was tested in a particular evaluation study, which concerned the realism of the generated simulated mammograms from computational breast models. The new approach in evaluating images was estimated to be more convenient and faster compared to our previous experience. The system will be further updated and will be exploited in the routine research work of the laboratory as well as will be shared with other research groups.

Phase-contrast imaging is an imaging technique, which provides an advantage over the traditional x-ray imaging by improving the contours of the tissues, thus providing higher contrast between tissues. To advance in this filed, the researchers from the Laboratory of Computer Simulations in Medicine are building a custom optical setup to generate diffraction pattern corresponding to the phase contrast data, but in the optical range. This paper reports on the current state of the optical system prototype and its initial validation and use.

In the present study we evaluate the performance of various training schemes for the locally recurrent probabilistic neural network and seek for advantageous tradeoffs between required training time and classification accuracy. Specifically, we consider training schemes which make use of a simple incremental procedure for adjusting sigma, as well as methods based on particle swarm optimization or differential evolution in different configurations. The experimental evaluation was carried out in common experimental protocol based on the Parkinson speech dataset. The experimental results show that with a proper training configuration a high accuracy can be achieved even with limited training data.

The development of new techniques for detection of breast cancer often relies on virtual clinical trials. These trials however require vast number of x-ray images of the breast containing realistically looking pathologies. For this purpose, as many as possible and as realistic as possible physical models of breast lesions are needed. The goal of this study is to create a methodology for manufacturing various breast abnormalities, with a detail description of the stages needed for their development. These models are to be used for development of new breast cancer detection technologies.

This paper presents a method for creation of computational models of breast lesions with irregular shapes from patient Digital Breast Tomosynthesis (DBT) images or breast cadavers and whole-body Computed Tomography (CT) images. The approach includes six basic steps: (a) normalization of the intensity of the tomographic images; (b) image noise reduction; (c) binarization of the lesion area, (d) application of morphological operations to further decrease the level of artefacts; (e) application of a region growing technique to segment the lesion; and (f) creation of a final 3D lesion model. The algorithm is semi-automatic as the initial selection of the region of the lesion and the seeds for the region growing are done interactively. A software tool, performing all of the required steps, was developed in MATLAB. The method was tested and evaluated by analysing anonymized sets of DBT patient images diagnosed with lesions. Experienced radiologists evaluated the segmentation of the tumours in the slices and the obtained 3D lesion shapes. They concluded for a quite satisfactory delineation of the lesions. In addition, for three DBT cases, a delineation of the tumours was performed independently by the radiologists. In all cases the abnormality volumes segmented by the proposed algorithm were smaller than those outlined by the experts. The calculated Dice similarity coefficients for algorithm-radiologist and radiologist-radiologist showed similar values. Another selected tumour case was introduced into a computational breast model to recursively assess the algorithm. The relative volume difference between the ground-truth tumour volume and the one obtained by applying the algorithm on the synthetic volume from the virtual DBT study is 5% which demonstrates the satisfactory performance of the proposed segmentation algorithm. The software tool we developed was used to create models of different breast abnormalities, which were then stored in a database for use by researchers working in this field. ispartof: PHYSICA MEDICA-EUROPEAN JOURNAL OF MEDICAL PHYSICS vol:57 pages:80-87 ispartof: location:Italy status: published

Introduction : 3D anthropomorphic models of human tissues have become a requirement for conducting realistic virtual studies. One of the current directions in the research of X-ray imaging is the development of physical models with 3D printing techniques using specific materials aiming to obtain replica of the human body tissues with similar radiological characteristics. Aim : The aim of this study is to create a calibration phantom for establishing the X-ray properties of different cartridge infills and their suitability to represent the X-ray properties of different breast types. Materials and Methods : A physical calibration model consisting of 22 objects was designed and printed by using an inkjet printer. A mixture was obtained from 5 mL printer ink and 3 g of potassium iodide (KI), which was used to fill the printer’s cartridge and to print the model on a set of plain office paper. Experimental X-ray images of the physical model were acquired on radiographic system SEDECAL X PLUS LP+. The obtained attenuation coefficient of the printing mixture was evaluated and compared to the breast tissue coefficients corresponding to the used X-ray energy. Results and Discussion : The physical model was printed on ten office sheets and stacked above one another. The obtained attenuation coefficient of the printing mixture was found very similar to that of the glandular tissue of the breast for the used X-ray energy. Conclusion: The obtained printer ink-KI mixture is suitable for representing the glandular part of breast tissue. The method has the potential to be used for creation of a realistic physical breast model.

The current contribution presents an approach for creation of complex computational breast phantoms, with included breast abnormalities. In-house developed and validated software tool, called BreastSimulator, is used to create different models of healthy breasts, by varying parameters like shape, size, duct tree features, Cooper ligaments, skin, etc. Separately, different 3D tumour models are constructed from segmented tumour regions on sets of real tomosynthesis images from real patients. A semi-automatic algorithm was developed to segment the abnormalities in the images. The algorithm applies a series of image processing operations and region-growing techniques with minimal interaction from the user to find the areas of the lesions. The flat tumour images are then correctly stacked and stored in a 3D voxel matrix, matching the desired resolution. Relevant information about the lesion (voxel size, matrix size, geometrical centre, etc.) is also saved. Selected tumour models are then inserted at various locations into the computational models of the healthy breasts. Examples of simulated mammographic and tomosynthesis images from such breast phantoms are presented to illustrate the potential of the proposed approach. Combining mathematical breast modelling with segmented real 3D tumour shapes results in the creation of realistic breast phantoms. The approach allows implementation of multiple scenarios and unlimited number of cases, which can be used for further software modelling and investigation of breast imaging techniques.

The Locally Recurrent Probabilistic Neural Network (LRPNN) consists of an input layer, three hidden layers and an output layer. The first two hidden layers are derived from the original PNN, while the third layer referred as recurrent layer is capable to model correlations within temporal sequences of observations. In the present study, we investigate the feasibility of FPGA-based implementation of the locally recurrent layer of LRPNN. An important consideration due to the specifics of this architecture is the use of modules with very high precision in the hardware design. Although expensive in terms of available resources in the FPGA chip, this is necessary, in order to compensate for the added error of quantization due to the multiple feedbacks from neurons in the neural network. The weights for the recurrent layer of the LRPNN are automatically computed from the available training data and translated into the hardware design. The experimental evaluation was carried out on the DEAP database, where two classes of emotional states were considered. The design makes use of a computed short-term energy from a 32-channel electroencephalographic (EEG) signal as an input. Results of an extensive experimental validation show that there is approximately one percent difference between the accuracy achieved with CPU-based software and FPGA-based hardware implementation of the LRPNN.

Goal of this study is to compare five statistical algorithms used for evaluation of software packages based on Monte Carlo techniques. These meth-ods are the following analysis: regression, correlation and Bland Altman as well as Wilcoxon rank sum and Kolmogorov-Smirnov tests. The methods were ap-plied for the case study of validating a dedicated computer code for calculation of scattered radiation reaching the eyes of the operator during interventional pro-cedures. The transport of x-rays and their interactions with the matter have been accomplished by the use of Monte Carlo techniques. Results were presented in the form of number of registered photons as a function of their energy and further compared to data from literature. From the five statistical methods for compari-son, the most suitable for our application turns out to be the Bland-Altman method. Further effort is related to the development of a specific software appli-cation for evaluation of data generated from general purpose x-ray simulations, as well as with analysis of more data obtained from different incident x-ray spec-tra.

A common approach in the development and improvement of diagnostic imaging techniques is the use of anthropomorphic phantoms. These phantoms can be physical or computational. In this study the creation of computational breast phantoms with included pathological formations is presented. The creation of the realistic phantoms is achieved by utilizing real patient data in the form of tomographic images. The 3D tumour models are generated by segmenting the regions containing tumour formations in the patient images. The segmentation is performed with a developed software tool based on a semi-automatic algorithm, which makes use of a series of image processing and region growing techniques. The software tool also provides the user an opportunity for corrections after the automated segmentation. Then the acquired flat images are stacked in a 3D voxel matrix. Creation of the computational healthy breast model as well as the compression procedure is achieved with a software tool called BreastSimulator. The healthy breast model and the segmented tumour formation are then interactively combined with a software tool called XRAYImagingSimulator. While the user can select a location for the tumour formation, also an automatic software processing is applied for integration between the two computational models. The simulation procedure for acquiring tomographic images from the created realistic breast phantom with included tumour formation is performed with the XRAYImagingSimulator software tool. Finally, the acquired simulation images are reconstructed with a software tool called FDKR. The combination of mathematical models of the breast and tumour models segmented from real patient data leads to the creation of realistic breast phantoms, which can be used in X-ray imaging simulation studies. The presented approach gives an opportunity for generation of multiple cases of breast cancer; thus allowing for further progress in already existing software models and techniques in diagnostic imaging. Acknowledgements This research is supported by the Bulgarian National Science Fund under grant agreement DN17/2. This project also has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 692097.

Hardware implementation of machine learning methods is often considered challenging, however, it brings major benefits, such as speed of operation and energy efficiency. Having in mind these benefits, we evaluate the performance and complexity of four probability density estimators (PDE) that we consider interesting for hardware implementation of the probabilistic neural network (PNN). We report results of a comparative evaluation of these PDEs for three different sizes of the PNN, evaluated in a common experimental protocol that makes use of the SPECT dataset. We conclude that two of the estimators are more appropriate for hardware implementation in FPGA.

High efficiency is among the most important targets in power electronic converters. A possible approach to obtain this goal is the usage of better switches. This paper is focused at the study and comparison of SiC and Si power MOSFETS in soft switching resonant DC/DC converter. The latter converter being a part of light battery charger used in ultra-light electric vehicles is realized and investigated. The losses in the power switches are compared mathematically, with simulations and experimental verification. Two types of transistors are considered. First traditional state of the art Si-technology using a FREDFET (fast recovery diode MOSFET) and it is compared with what the SiC technology offers, using a fast Schottky antiparallel diode. As an example a 1500 W series resonance converter with input voltage of 400 V and output voltage of 100 V is used. The latter voltage permitting the use of Schottky diodes at the secondary side. The losses in both converters are compared via calorimetic measurements to avoid accuracy problems as in loss measurements with an oscilloscope for example.