CC BY
13
Научни трудове на Съюза на учените в България-Пловдив, Серия Г. Медицина, фармация и дентална медицина т. XIX. ISSN 1311-9427 юни 2016. Scientific works of the Union of Scientists in Bulgaria-Plovdiv, series G. Medicine, Pharmacy and Dental medicine, Vol. XIX, ISSN 1311-9427 Medicine and Dental medicine June 2016.
ТРИИЗМЕРНИ МОДЕЛИ НА КАРЦИНОМИ НА ГЪРДА
ПРЕДНАЗНАЧЕНИ ЗА ИЗСЛЕДВАНЕ В ОБЛАСТТА НА РЕНТГЕНОЛОГИЯТА Живко Близнаков, Янита Черногорова, Кристина Близнакова Технически университет - Варна
THREE DIMENSIONAL BREAST CANCER MODELS FOR X-RAY IMAGING RESEARCH
Zhivko Bliznakov, Yanita Chernogorova, Kristina Bliznakova Technical University of Varna
ABSTRACT: Nowadays, the development of realistic 3D physical and computational models of breast tumours with irregular shapes is an urgent requirement. The availability of such models is a powerful tool for the development of new technologies for precise definition of the boundaries of these cancers. Biomedical engineering unit at the Technical University of Varna (TUV) is present in this area both at modelling and simulation of computational breast phantoms and x-ray breast imaging techniques. To advance further in achievement of its innovative and challenging goals the TUV has collaborated with experts from top research institutions in the relevant field. This collaboration is presented within the MaXIMA project, with a main objective to increase the research and innovation capacity of the TUV in the field of computational modelling of breast tumours and their use in studies of advanced x-ray breast imaging techniques.
Key words: breast cancer, x-ray imaging, breast phantoms, 3D breast cancer models
I. Introduction
These days, the development of realistic three-dimensional physical and computational models of breast tumours with irregular shapes is essential. The existence of such models is a powerful tool in the hands of engineers, doctors and medical physicists and provides great perspective to researchers to use these models for development of new diagnostic x-ray technologies facilitating the precise delimitation of tumour edges.
Scientists from the biomedical engineering unit at the Technical University of Varna work in this area, both on modelling and simulation of computational breast phantoms and x-ray breast imaging techniques. The proposal "Three dimensional breast cancer models for X-ray Imaging research" had the European Commission approval and funding by the Horizon2020 programme. The MaXIMA project was highly evaluated by the Research Executive Agency (REA), which is an indisputable sign of the significance of this study [1].
II. Project Description
The main objective of MAXIMA action is to increase the research and innovation capacity of the Technical University of Varna in the field of computational modelling of breast tumours
98
(including cancers with irregular shape) and their use in studies of advanced x-ray breast imaging techniques, such as breast tomosynthesis and phase contrast imaging. For this purpose, a close and sustainable collaboration with other two European Universities: the Katholieke University of Leuven and the University of Naples - Federico II working in the specific domain of modelling and simulations dedicated for studies of x-ray breast imaging techniques was established.
For the purpose of better performance, management and monitoring of the results of scientific work, project activities are divided into five work packages (WP). The first WP „Management and Coordination" and the last WP „Dissemination and Exploitation" are required for project proposals under the „Horizon 2020" program. The remaining three work packages reflect the specific goals and outline the different stages of the MaXIMA project research. The diagram (Fig.1), describing the stages of the scientific work, clearly shows the formulated project work packages:
WP2 - Advancement in imaging biology samples.
WP3 - Advancement in developing innovative methods for tumour modelling.
WP4 - Reinforcing research capacity for effective use of cancer models in breast imaging.
Reference data used for the research are acquired from patient specific data from CT and tomosynthesis as well as images of histological samples. They are collected in the specifically created and designed for the project needs database shared amongst the project members.
Figure 1. Scientific work packages.
To succeed in creation of algorithms for modelling of tumours with irregular shapes, a number of scientific events were planned during the first year of the MaXIMA project. These include a training school on CT imaging of biology samples (WP2), regular work meetings with medical doctors, thematic scientific seminar, Special Session on "eHealth and Networking for Advanced Cancer Detection" as a part of the IEEE Black See Conference 2016. Specifically, the training summer school was organized at the Technical University of Varna with a main lector Prof. Paolo Russo from the University of Naples. The practical training of the young scientists will be accomplished in modern equipped laboratories of the Katholieke University of Leuven and the University hospital in Leuven, Belgium, in August 2016.
The concept for organizing these events throughout the project lifespan, in parallel with the scientific work, on one hand, will increase the research capacity of the Technical University of Varna team, and on the other, will ensure high quality of the results of the scientific goals. A Compromise with the last would not be acceptable both from the European Commission, as well as, by the local researchers and scientists, who follow strictly the implementation of each project stage.
The accumulated database of 3D images of breast tumours and the increased knowledge level
99
will enable the team to proceed with WP3 - Advancement in developing innovative methods for tumour modelling. This is the scientific field in which engineers from the Technical University of Varna have high experience. Within successfully completed European project PHASETOMO [2], scientists from the Technical University of Varna updated the BreastSimulator software platform [3] with additional modules for phase contrast mammography and tomography, which opens possibilities to study the performance of these imaging techniques with generated computer models of uncompressed and compressed breasts (Fig.2). This achievement has opened new opportunities for further research and development in the field, and MaXIMA action is its natural continuation.
Using the available patient specific data, first attempts to segment breast tumours were accomplished. The resulting images, achieved by the new algorithm for tumours' segmentation provide important information about the tumours' location, size, and visibility. Evaluation of the first segmented models was provided by medical experts in this field based on criteria such as degree of extraction of the tumour (complete or partial), extraction accuracy (only tumour formation or other tissue) and tumour shape. Periodic inspections and discussions with medical doctors are planned during the whole project life-time, which is expected to result in robust algorithms for the creation
of realistic computational tumour models. The requirement for proceeding to the next stages of the project development will be the favourable opinion of the evaluators.
Figure 2. Computer breast model, generated with the BreastSimulator.
The successful tumour modelling will speed up the research in two directions: (a) Computer modelling of tumours with irregular shapes and their inclusion within the computer models of breasts, generated by the BreastSimulator software platform for the purposes of design, testing and validating through simulation work of phase contrast breast imaging (WP4), (b) Creating of physical models of tumours and breasts using the available 3D printer technologies. Two of the technologies: the fuse deposition modelling and the stereolithography are well known printing technologies and used by the team of Technical University of Varna. Recently, these two technologies were used for manufacturing various physical phantoms of compressed breast models (Fig. 3). These models will be used to design, test and validate experimentally the phase contrast imaging technique (WP4).
The role of physicists and biomedical engineers involved in the project is to determine the most suitable materials, which own similar absorption and phase characteristics as these of the human tissue for diagnostic x-rays. These materials then will be used for the manufacturing of realistic breast phantoms with realistic healthy and cancerous tissues. The team is currently considering an innovative approach to improve the technology for manufacturing the complete breast by simultaneously printing with two
Figure 3. Phantoms printed by TUV team.
different materials.
The final stage of the research is related to the testing of the results obtained in the project study. Both, computer simulations and real experiments with physical phantoms are planned. To verify the accuracy of the first simulation results from phase contrast breast tomosynthesis, as well as, to check the absorption and the phase characteristics of a set of materials, a proposal to the European Synchrotron Radiation Facility in Grenoble, France was submitted in March and recently approved.
III. Discussion and Conclusions
A characteristic feature of the MaXIMA project is its complexity. The relation between different scientific stages is both horizontal within the activities of a specific work package itself, and vertical - amongst several work packages. Parallel developments (computer simulation and real experiments) are used in order to define and verify the accuracy of the results.
The excellence and sustainability of the project consists in the opportunity for wider application of the developed tools and obtained results in future research application. The results of a given scientific task in the corresponding WP may be combined into a product that can be used for teaching and training purposes in Bachelor and Master of Science programs for students and young researchers in the field. For example, the software platform BreastSimulator is very useful for training of engineers and young physicians in computer modelling, visualizing and determining the edges of tumours with irregular shapes. Phase-contrast computer simulator has the potential to become an indispensable tool for training of radiologists and medical physicists.
The lack of realistic breast phantoms results in limited possibilities for testing new imaging technologies, such as phase-contrast tomography and mammography. The challenge of constructing variety of ratios for tissue composition of the breast has not yet been overcome. Therefore, biomedical engineers, physicians, medical physicists and computer scientists work together within the MaXIMA team, complimenting each other, in order to develop as realistic as possible physical models of the woman breast. The success of this final phase of the project and the potential verification of complex physical phantom would be a new impulse for X-ray imaging research.
Acknowledgments
This work is supported by MaXIMA action: Three dimensional breast cancer Models for X-ray IMAging research. This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 692097.
References
[1] Three dimensional breast cancer Models for X-ray IMAging research - MaXIMA project from the H2020-TWINN-2015 (Project Number: 692097)
[2] Development of a three-dimensional reconstruction algorithm for phase contrast breast tomosynthesis - PHASETOMO - Marie Curie Career Integration Grant within the 7th European Community Framework Programme (PCIG09-GA-2011-293846)
[3] Bliznakova K, Sechopoulos I, Buliev I, Pallikarakis N, 2012, BreastSimulator: A software platform for breast x-ray imaging research. Journal of Biomedical Graphics and Computing, 2(1), pp. 1-14, DOI: 10.5430/jbgc.v2n1p1,
Contact details
Author: Kristina Bliznakova Organization: Technical University of Varna
Address: 1, Studentska Str.
City: Varna, 9010
Country: Bulgaria
E-mail: [email protected]
