Final ReportSummary - BIO-CT-EXPLOIT (Innovative simulation tool for bone and bone biomaterials, based on enhanced CT-data exploitation)
The use of computer tomography (CT) imaging is steadily increasing in the ever growing bone implant / surgery and tissue engineering market, although commercial exploitation of CT data for structural design purposes is still based on trial-and-error approaches. This is because X-ray attenuation information is reduced to geometric grey level evaluation. However, on the academic stage, a transnational team of highly esteemed applied physicists, material scientists, engineering mechanicians, and mathematicians has recently pioneered concepts for extraction of chemical information from CT, and of its conversion, via micromechanics laws, into object-specific, inhomogeneous and anisotropic material properties. BIO-CT-EXPLOIT substantiated, through well-tailored R&D activities, this cutting-edge knowledge into unparalleled, highly reliable simulation tools for structural design purposes. Most of the work load related to these R&D activities was carried by the RTD partners, which transferred (upon reimbursement) the simulation tools to four highly innovative SME partners covering all aspects of the bone biomaterial / surgery preplanning market, being leaders in the fields of biomaterial production, of micro and nano-CT scanner development, of image-to-geometry / mesh conversion, and of finite element simulation technologies. As a result of the R&D activities being carried out in close cooperation with SMEs, the latter have become the owners of software prototypes designed for SME-specific needs, with rapid time-to-market characteristics. The uniqueness of these products will tremendously improve the strategic market positions of the SMEs, which are expected to generate annual revenues being already multiples of the singular EC contribution - when just considering the submarkets of preplanning dental / orthopaedic surgery and bone tissue engineering research. This may trigger SME growth rates exceeding 30 %, both in turnover and employment.
Project context and objectives:
CT is a powerful non-destructive evaluation technique for producing 3D images of an object. Numerically, the latter is defined through values related to small volume units, called voxels. Corresponding software primarily focuses on sophisticated evaluation of data, i.e. on impressive, clear, and easy-to-use visualisation tools. The latter allow for definition of topology and geometry of the investigated objects - and this information is the key to further data processing and understanding, such as mechanical simulations (e.g. based on finite elements) of bio-structures, the focus of the present application. It involved four SMEs with different commercial activities:
- Skyscan is specialised in CT hardware production, with inbuilt software;
- Simpleware is specialised in image-to mesh conversion;
- Cadfem is specialised in industrial and medical finite element analyses; and
- Inmatrixs is specialised in calcium phosphate biomaterial production.
These SMEs encountered repeatedly the following scientific / technological problem. While various options were at their disposal to derive, from the CT data, geometric models and corresponding finite element meshes, the mechanical properties of the material contained in the voxels remained largely obscure. Basically, they were guessed on the basis of experience or of some back-analysis; or some additional experimental work would be necessary to have access to them (which is sometimes expressed in terms of empirical regression functions between CT values and mechanical properties). However, without reliable information on material properties, FE analyses remain of rather qualitative nature, and their suitability for quantitative computer-aided design of biomaterial implants remains very limited.
On the other hand, the demand for reliable, cost-effective biomaterial design (including the wide realm of tissue engineering and regenerative medicine) is very high so that, once an SME would have a reliable access to material properties as input for FE analyses, this SME's competitive position on the market would be strongly strengthened. The same type of software would strongly affect the field of computer-aided radiology and surgery (as well as the field of simulation-supported surgery pre-planning), where complex, anisotropic and inhomogeneous material behaviour is standardly encountered; and the markets are huge in both cases.
The aforementioned access to voxel-specific reliable material property information, directly from the CT data themselves, and the integration of this information into commercial software were the central objectives of BIO-CT-EXPLOIT.
Therefore, latest conceptual research results of a highly qualified, interdisciplinary research consortium encompassing engineers, physicists, material scientists, and mathematicians were substantiated into clear and easy-to-understand computer codes which fit smoothly into the highly developed software products of one SME producing CT scanning instruments (Skyscan) and two SMEs developing simulation software (Simpleware and Cadfem) (with focus on image-to-mesh transformation and on finite element simulation technology, respectively). The results obtained from a first use of this new software for biomaterial design purposes was transferred to a fourth SME (Inmatrixs), active in biomaterial development and production.
Substantiation of cutting-edge, interdisciplinarily acquired concepts linking X-ray physics consistently to multiscale engineering mechanics and to applied mathematics, into robust, easy-to-understand software products is a veritable R&D task which can neither be attained by a single SME, nor by an SME consortium: Therefore, this task had necessarily to be outsourced to a multidisciplinary team of RTD partners representing their respective scientific fields at the world reputation level.
The following project objectives were tackled during the project duration of two years:
1. production and CT-scanning of various biomaterials, such as porous hydroxyapatite globules and rapid-prototyped tissue engineering scaffolds made of calcium phosphate-impregnated polymers;
2. collection and production of various bone material scans, comprising human and mouse femurs, and human mandibles;
3. development of software prototypes for the conversion of voxel-specific X-ray attenuation coefficients to elastic properties, falling into three categories:
- clearing of CT data from effects of instrumental resolution;
- identification of principal directions of anisotropic elasticity from object geometry / texture;
- translation of CT data into micromechanics-based elasticity tensors;
4. SME-specific image analysis tools, in form of software prototype releases concerning the following topics:
- clearing of CT data from effects of instrumental resolution;
- identification of principal directions of anisotropic elasticity from object geometry / texture;
- translation of CT data into micromechanics-based elasticity tensors;
5. comprehensive mechanical testing (quasi-static, nanoindentation, ultrasound) of polymer- and ceramic-based biomaterials; and
6. collection of software demonstration and clinically relevant 3D simulations of bone and bone biomaterial systems.
Project results:
The principle milestones achieved during the project were the following:
1. production and CT-scanning of various biomaterials, such as porous hydroxyapatite globules and rapid-prototyped tissue engineering scaffolds made of calcium phosphate-impregnated polymers;
2. collection and production of various bone material scans, comprising human and mouse femurs, and human mandibles;
3. development of software prototypes for the conversion of voxel-specific X-ray attenuation coefficients to elastic properties, falling into three categories:
- clearing of CT data from effects of instrumental resolution;
- identification of principal directions of anisotropic elasticity from object geometry / texture;
- translation of CT data into micromechanics-based elasticity tensors;
4. SME-specific image analysis tools, in form of software prototype releases concerning the following topics:
- clearing of CT data from effects of instrumental resolution;
- identification of principal directions of anisotropic elasticity from object geometry / texture;
- translation of CT data into micromechanics-based elasticity tensors;
5. comprehensive mechanical testing (quasi-static, nanoindentation, ultrasound) of polymer- and ceramic-based biomaterials; and
6. collection of software demonstration and clinically relevant 3D simulations of bone and bone biomaterial systems.
Concerning items 3 and 4, the following details are noteworthy. Regarding (1), instrument-related statistical distribution of X-ray attenuation coefficients were represented through a spline smoothing approach, and then de-convolved from the CT-images, based on the technique of histogram equalisation. Improvement of the signal-to-noise ratio led to a robust and quick software prototype. Regarding (2), an algorithm for interpolation between directions on organ surfaces and directions of anatomical guiding lines was developed. It was implemented in form of a highly modular code structure, which can be used as new add-on module within same larger program system, or as a stand-alone software tool. Regarding (3), the volume average rule for X-ray attenuation coefficients was employed to convert CT-data into composition data, which gave, via micromechanics, access to voxel-specific elasticity values. Thereby, emphasis was laid on high predictive precision of micromechanics formulation (as validated through experimental data), and on compatibility with the diverse software environments used by the involved SMEs.
In more detail, the milestones were reached as part of a number of work packages (WP):
The key objectives of WP2 'Software and data interfaces: From RTDs to SMEs' were:
(1) definition of file formats in which SME partners provide input data for RTD-developed software;
(2) definition of software interface through which RTD partners read SME-delivered data, for their evaluation through the new software of WP3.
Concerning the first of these objectives, input data in form of CT images were provided in TIF, BMP, and DICOM file formats. The consortium compiled a collection comprising both projection and reconstructed CT images. This collection comprises micro- and nano-CT images of porous polymer-ceramic biomaterials fabricated at WUT, micro-CT images of a mouse femur provided by Skyscan, micro-CT images of a human femur provided by Simpleware and scanned by Skyscan, as well as standard CT images of male and female human mandibles, provided by HAW.
Concerning the second of the aforementioned objectives, the consortium agreed on the development of stand-alone software prototypes in form of MATLAB-based graphical user interfaces (GUIs) for clearing clearing of CT data from effects of instrumental resolution, and for translation of CT data into micromechanics-based elasticity tensors, and in form of C++ code for identification of principal directions of anisotropic elasticity from object geometry / texture, these prototypes allowing for input data in DICOM, TIF, JPG, PNG, and BMP formats (as regards any form of CT data), and in STL-format (as regards surface data of human mandibles).
The key objectives of WP3 'Physico-mechanical software development' left for investigation in the period from month 12 to 18 were:
(1) identification of instrument-related statistical distribution of X-ray attenuation coefficients ('line broadening');
(2) correction of spatial distribution of X-ray attenuation coefficients from instrument-induced effects;
(3) translation of CT data into voxel-specific material composition values of bone and bone biomaterials;
(4) translation of material composition into voxel-specific elastic properties of bone and bone biomaterials;
(5) reconstruction of trajectories of anisotropic elasticity for bone and bone biomaterials;
(6) material anisotropy definition in finite element models.
The first of these objectives relates to microCT images with an air-filled homogeneous pore space and an inhomogeneous solid phase. The probability density function (pdf) of corresponding X-ray attenuation coefficients is characterised by two peaks ('air peak' and 'solid peak', with certain widths). There, our work started, for the sake of mathematical simplicity, with approximating of both the line boardening of the air peak and of the solid peak, by Lorentzian functions. A corresponding low accuracy in some cases motivated the use of peak model functions (PMFs), which are more precise, but the numerical fit of the bimodal pdf still poses problems. As an interesting solution to this problem, a spline smoothing approach was developed, which allowed for separation of the two peaks, for precisely fitting the numerically defined pdf, and for smoothing out statistical fluctuations.
Also the second of the aforementioned objectives relates to microCT images with an air-filled homogeneous pore space and an inhomogeneous solid phase. Although the air properties are homogeneous, the air peak is no Dirac function (as it should be theoretically), and the actually observed distribution is used to correct, through de-convolution , the solid voxel distribution from errors related to the experimental device inclusive of the reconstruction algorithms applied to original CT projection data. In this context, the determination of spatial distribution of corrected specific attenuation values was a major challenge, which was finally met through the technique of histogram equalisation. Other important aspects were the identification of voxel outlines and image masking for the identification of the object of interest.
As regards the third of the aforementioned objectives, our starting point was the linear relation between grey values given by standard CT scanners like the ones produced by Skyscan and X-ray attenuation coefficients. Combination of this relationship with the volume average rule for X-ray attenuation coefficients (the X-ray attenuation coefficient of a voxel is the volume average of the X-ray attenuation coefficients of all material components building up this voxel) led to an average rule for grey value. This average rule was then applied to three material classes with distinct chemical composition characteristics: porous ceramic biomaterial, extracellular bone matrix, and macroscopic bone material. This allowed for determination of voxel-specific composition, in terms of nanoporosity, of mineral and collagen content, and of vascular porosity. Respective key input parameters are the nanoporosity of the densest voxel in a ceramic biomaterial, the mean mass density of extracellular bone matrix, and the attenuation coefficient related to extracellular bone matrix.
Concerning the fourth of the aforementioned objectives, homogenisation schemes in the framework of continuum micromechanics were used to translate porosities or mineral / collagen contents into corresponding elastic properties. Micromechanical models appropriate for ceramic biomaterials, as well as for bone at the micrometer and at the milimeter scale were implemented in a computationally efficient manner.
As concerns the fifth of the aforementioned objectives, principal material directions were derived from interpolation between surface directions and directions of anatomical guiding lines within arbitrary organs. Additional software was provided for three types of guiding line construction in human mandibles:
(1) 'anatomical' guiding lines respecting full anatomy;
(2) 'rudimentary' guiding lines based on characteristic anatomical landmarks; and
(3) guiding lines which were adapted from a gallery of typical mandibles, to the actually considered organs.
Emphasis was laid on an efficient implementation in the C++ class structure.
As concerns the sixth of the aforementioned objectives, the output of the aforementioned software tools is designed in a way which is compatible with standard finite element packages used in our consortium, such as ANSYS and ABAQUS.
These issues were tackled in the first half of the project (months 1 to 12). Afterwards, the period from month 12 to 18 was devoted to a feasibility study on further development of corresponding software tools, beyond the level of the prototypes delivered to the SME partners in WP4. The aforementioned feasibility study extended the discussion on material anisotropy from the level of natural bone systems, to that of tissue engineering systems. Thereby, both the aspects of tissue engineering scaffold design and those of clinical realisation of tissue engineering solutions were tackled in an unprecedented fashion: as concerns biomaterial design, micro finite element analyses derived from micro-CT scans were identified as the proper technological solution for the determination of scaffold anisotropy, whereby a new way to arrive at the elastic properties of polymer-ceramic scaffolds was developed: their voxel-specific attenuation coefficients were transformed, based on X-ray attenuation averaging rules, into voxel-specific ceramic and polymer volume fractions, and the latter entered a new micromechanical formulation representing the micro and nanocrystals dispersed in a contiguous polymer matrix, this formulation delivering the sought voxel-specific elastic properties. As concerns clinical tissue engineering solutions, a clinical case provided by Inmatrixs, namely surgery and biomaterial treatment of a cystic mandible, was analyzed with respect to tissue inhomogeneity and anisotropy: sequential analysis of CT data sets revealed strong development of new bone tissue induced by the full load carrying biomaterial, to occur around six months after surgical intervention. This bone development was spatially inhomogeneous, resulting in the evolution of a new 'cortical shell', as opposed to a less dense, more trabecular bone-like material in the inner part of biomaterial-filled cystic volume. Furthermore, the algorithm for material direction identification, developed during months 1 to 12 for mandibles without large defects, turned out to be fully applicable to the investigated mandible with a large cystic defect - this being a clear step beyond the aims originally foreseen for BIO-CT-EXPLOIT.
The key objectives of WP4 'Software and data interfaces from RTDs to SMEs' were:
(1) definition of file formats in which RTD partner-developed software provides input data for the commercial software environments of the SME partners;
(2) definition of commercial software interfaces through which SME partners read data produced by RTD partner-developed software.
These two objectives were met by the RTD partners through development of SME-specific solutions for integration of newly developed software prototypes into the commercial software of the SMEs:
As regards the interface to ScanFE, a software product developed and distributed by Simpleware, the BIO-CT-EXPLOIT software for clearing CT data from effects of instrumental resolution has been developed in the MATLAB environment from where it was compiled for distribution as a MATLAB-based stand-alone executable. In the context of a GUI, the software is fully compatible with various image data formats, and performs fully automated preprocessing steps such as data reduction, outlier removal and object of interest identification. In addition to the corrected image files, various informations on the scanned objects themselves, such as pore radius frequency plots, are provided in different formats, as specified by the user. In full compatibility with ScanFE's material property input box, the BIO-CT-EXPLOIT translation of CT data into voxel-specific complete micromechanics-based elasticity tensors was provided in form of power-functions relating attenuation coefficients and mass densities, as well as mass densities and elastic properties. Corresponding factors and powers were derived from combined X-ray physics-micromechanics methods, and as a result, fully explicit scalar formulae, as well as GUI-produced data sets, were provided. Since ScanFE is tailored for image-to-mesh conversion, the BIO-CT-EXPLOIT software for identification of principal directions of anisotropy in human mandibles may provide these material directions directly on a voxel-per-voxel basis.
As regards the interface to ANSYS, a software product distributed by CADFEM, the error clearing aspect of the BIO-CT-EXPLOIT prototypes is dealt with similarly as described before for ScanFE, while the image-to-property conversion (concerning the effects of both material inhomogeneity and anisotropy) is realised such that the software provides ready-to-use ANSYS code, be it for ceramic biomaterials, or for human mandibles exhibiting inhomogeneous and anisotropic material properties. Therefore, the user is equipped with GUIs built on MATLAB code, available as stand-alone executables.
As regards the interface to CTan(alyzer), a software product developed by Skyscan and distributed with the sale of microCT desktop machines, both the error clearing concept and the image-to-property conversion tool were re-programmed from scratch, as to be fully compatible with the C++ source code at the foundation of CTan. As an interesting side product of this re-coding step, a considerable speed-up needs to be mentioned.
The key objectives of WP5 'Processing and CT imaging of biomaterials' were:
(1) processing of state-of-the-art bone biomaterials and tissue engineering scaffolds (ceramic and polymer-ceramic composites);
(2) micro and nano CT imaging of bone biomaterials.
As concerns the first of these objectives, porous hydroxyapatite granules at different sizes, and calcium phosphate-impregnated polymer scaffolds were produced.
As concerns the second of these objectives, the aforementioned materials were scanned in micro and nano-CT scanners, and the resulting data were distributed to the consortium partners.
The key objectives of WP6 'Mechanical testing' were:
(1) nanoindentation tests on state-of-the-art bone biomaterials;
(2) uni- / triaxial (poromechanical) tests on bone biomaterials;
(3) acoustic (ultrasonic) tests on bone biomaterials.
These objectives were met in two parallel experimental programs, devoted to biomaterials consisting of calcium phosphate ceramics only, and to rapid-prototyped ceramic-polymer-based biomaterials, respectively.
Due to their spherical shape, the ceramic globules produced by Inmatrixs could not be readily subjected to standard biomechanical testing protocols, and a novel protocol was developed (going beyond the original plan given in BIO-CT-EXPLOIT Annex I). The globules were subjected to quasi-static micro-splitting tests, which were accompanied by sequential microCT scans. In this way, thanks to recent equipment developments undertaken by SKYSCAN, the 3D fracturing behaviour of these biomaterials under mechanical load were, for the first time, investigated, and the corresponding test data provide a valuable basis for validation of any numerical simulations on hydroxyapatite globules. As a rule, fracturing processes were initialised at comparatively large pores inside the globules, as well as at microcracks stemming from the production process.
On the other hand, the rapid-prototyped, ceramic-polymer-based biomaterials were subjected to a multitude of different mechanical testing procedures (nanoindentation tests; quasi-static loading / unloading as well as poromechanical tests, with and without simultaneous microCT imaging; as well as ultrasonic tests at different frequencies), providing in this way a rarely seen completeness of mechanical characterisation. In particular, all methods were investigated concerning their consistency, and the unloading experiments were identified as those revealing elasticity (confirmed by acoustic testing), whereas loading tests are characterised by significant inelastic processes.
The key objective of WP7 'Demonstration' was the demonstration of the new software modules. These demonstration activities fell into two parts:
(i) showing the proper functioning of the newly developed software prototypes; and
(ii) numerical simulations based on these software prototypes, tackling real-life problems.
As regards proper functioning of BIO-CT-EXPLOIT software, the error clearing tool has fully met and partially even surpassed the original requirement catalogue of Annex I, including the following aspects: multipeak materials can be treated, high-speed C++ versions are available, clearing option is extended to polygon-based artefact removal. The code for principal direction identification works smoothly on images of strongly varying physiological cases. Attenuation-to-elasticity conversion was successfully demonstrated for a human mandible affected by a cyst, for a mouse femur, and for a ceramic biomaterial globule.
As regards demonstration of the new software for real-life applications, several large-scale simulations were performed:
(i) a fully inhomogeneous and anisotropic simulation of biting with a cyst-affected human mandible, delivering stress and strain profiles as function of simulation sophistication;
(ii) a high resolution microFE analysis of a mouse femur, showing the relevance of composition-driven inhomogeneity of elastic properties, for realistic estimation of stress profile in a physiological load case;
(iii) various high resolution simulations of ceramic, as well as ceramic-polymer biomaterial and tissue engineering scaffolds, again showing the relevance of composition-driven inhomogeneity of elastic properties, for realistic estimation of stress profiles under physiological strain.
In all these case, emphasis was also put on proper visualisation of the results.
The key objectives of WP8 'Dissemination and training' are:
(1) training of SME partners;
(2) identify and organise activities to promote the commercial exploitation of the results and the widest dissemination of knowledge possible;
(3) exploitation of the results in the health sector (clinics, implant manufacturing).
These objectives were tackled in a variety of activities:
- Three workshops were organised, each of them focussing on the use of the newly developed software in one of the three SME partners Skyscan, Simpleware, and Cadfem. In order to also maximise the audience from the side of the SMEs' customers, all three workshops were held in the framework of the SMEs' annual users meetings.
- The SME partners were provided with manuals describing the functionality of the new software prototypes, and how they can be used within their commercial software environment. In addition, the CT-scanning experience gained during the project was summarised in terms of guidelines for effective micro- / nanoCT scanning.
- Wide dissemination of knowledge gain was achieved by completion and publication of several papers for international scientific journals, and by presentations on BIO-CT-EXPLOIT topics at international conferences. The key challenge, namely not to present confidential details harming the market position of the SMEs, and to have results published within the relatively short project duration of two years, was successfully met.
- The wide public was informed, by press release, about the successful completion of the BIO-CT-EXPOIT project, and the message on the project has been well spread also during the project duration, as it has entered several brochures which are publicly available on the internet. In particular, BIO-CT-EXPLOIT was featured in the fit-for-health campaign (please see http://www.fitforhealth.eu/success-stories.aspx for details).
Potential impact:
Today the finite element method (FEM) is an integrated part of the development process in countless industrial activities, where its application helps the engineer to faster develop new designs, to increase the quality of new products, to reduce the development costs, and to bring goods faster to the market. A potential new market for FEM technologies has emerged: the health care market. There, FEM is used particularly in patient / individuum-specific simulations for diagnosis and therapy. However, FEM programs are still complex to use, and only trained people with a technical background can use the software. However, in order to enter the broad medical market, also the 'normal' doctor has to be directly addressed. Therefore, the FEM software has to be first of all easy-to-use, possibly automated, robust, and fast to apply. To meet this goal, CADFEM has established, in 2007, a new business unit: CADFEM Medical. Its aim is to identify surgeries where FEM can bring a benefit. For these surgeries, vertical applications based on standard and established FE packages like ANSYS are developed. These vertical applications are intended to fulfil the aforementioned requirements, such as easy use, high automation, robustness, and speed. However, CADFEM is facing new problems when addressing medical applications, problems which are not encountered in the 'classical' engineering field. The biggest problem is for sure the determination of reliable patient / individuum-specific material properties. And this is exactly where the project results of BIO-CT-EXPLOIT, namely the determination of realistic material properties out of CT data, turn out as the critical key to successful and significant penetration of the biomedical markets; and finally to establishment of the FEM in these markets - with CADFEM at the very forefront.
While the eminent strength of CADFEM lies in finite element solutions, our second simulation tool producer, Simpleware, is world-leading expert in conversion image data into geometric information and meshes. Simpleware software (ScanIP and ScanFE) provides state-of-the-art tools required for processing 3D imaging data as obtained by imaging modalities such as MRI and CT, for segmentation and identification of volumes of interest, as well as for automated generation of finite element meshes to be used in a wide range of third party solvers such as ANSYS (CADFEM). On the other hand, as for the traditional product line of the Simpleware, material property determination is largely left to the user. With BIO-CT-EXPLOIT having targeted at the determination of realistic elasticity properties from CT-data based micromechanics, Simpleware's portfolio in realistic simulation tools for complex problems related to BIO-structures has been significantly extended, opening new commercialisation routes not only in the biomedical, but also in the materials science market. In fact, very recently the latter market has gained significant importance for Simpleware.
BIO-CT-EXPLOIT also fits well into the market strategy of our CT instrument producer SKYSCAN, being quite distinct from those of the simulation tool providers. As a micro and nano-CT manufacturer Skyscan does not sell its software as a self-contained product, but as complement to CT instruments. Nevertheless, software can become one of the most important topics in a sale. Desktop microtomography systems always use polychromatic X-ray sources. The latter provide the possibility to use a micro-CT equipment in a laboratory environment, as an interesting alternative to a huge synchrotron facility. Unfortunately, polychromatic sources compromise the attenuation coefficients, making quantification difficult. BIO-CT-EXPLOIT has opened up new avenues for better interpretation of polychromatic X-ray values, based on the correction scheme for clearing CT data from instrumental effects including reconstruction algorithms. This is expected to already attract new customers deciding for desktop systems, while it will also be of great help for existing users. BIO-CT-EXPLOIT has evidenced the beneficial effects of new developments, such as special stages for performing in-situ measurements on objects being deformed in compression and/or tension. These in-situ measurements will also open up an important new market. In addition, the BIO-CT-EXPLOIT project results will enable SKYSCAN to offer new services to existing and new users within a short time frame.
Again a different market strategy is followed by our biomaterial producer, Inmatrixs, a highly innovative, young company which focuses on the processing of calcium phosphate scaffolds for stem cell research and tissue engineering research. The structure-property relations revealed through BIO-CT-EXPLOIT allow for optimisation of the mechanical performance and biological behaviour of these biomaterials. Use of entirely new, unparalleled design tools will allow for a significantly accelerated development cycle. This, in turn, will reduce both overall costs and time to market of Inmatrixs products, which will be directly used in the production (at least on pilot scale) and also in a clinical context.
Taking a larger view, BIO-CT-EXPLOIT has helped to fulfil a number of societal objectives in the European Community. Improved surgery planning or tailoring of tissue implants and scaffolds - through improved CT-data exploitation for simulation purposes - improves health and life quality, by significant reduction or repair of patients' disabilities or handicaps, together with decreased medical complications due to e.g. lower probability for further traumatical interventions. This affects particularly the largely growing population of elderly people in Europe. It improves the safety and mobility of this part of the population, enabling them to take a more active part in attacking Europe's challenges. Development of improved biomaterials for more economic and safer performance in living systems creates new industrial branches, helping to increase employment rates. Optimising materials leads to longer service life of permanent implants or of biological tissue grown on resorbable scaffolds, which both alleviates adverse effects on the environment. Finally, engineering within a living system requires its utmost comprehension: The highly relevant new knowledge obtained through BIO-CT-EXPLOIT will improve Europe's education and training standards. This is supported by scientific publications in international journals, guaranteeing the spreading of BIO-CT-EXPLOIT knowledge throughout the world-wide research and higher education scene.
World-wide dissemination of the project results is also eased through a world-wide network of companies, called TechNet Alliance (see http://www.technet-alliance.com for details), built up by CADFEM and including also our SME partner Simpleware. The members are usually companies offering in their local market a portfolio similar to those of CADFEM and Simpleware. If one TechNet partner feels to have a promising product the TechNet members can be used as a world-wide distribution channel, comprising countries such as Austria, Belgium, Brazil, China, Czech Republic, France, Greece, India, Ireland, Italy, Japan, Luxemburg, the Netherlands, Poland, Spain, Switzerland, Turkey, UK, or USA. All in all TechNet has more than 50 members in more than 20 countries. Linking all these partners to the biomedical market will guarantee an optimum world-wide spreading of the results obtained in the framework of BIO-CT-EXPLOIT.
Project website: http://bio-ct-exploit.imws.tuwien.ac.at
Project context and objectives:
CT is a powerful non-destructive evaluation technique for producing 3D images of an object. Numerically, the latter is defined through values related to small volume units, called voxels. Corresponding software primarily focuses on sophisticated evaluation of data, i.e. on impressive, clear, and easy-to-use visualisation tools. The latter allow for definition of topology and geometry of the investigated objects - and this information is the key to further data processing and understanding, such as mechanical simulations (e.g. based on finite elements) of bio-structures, the focus of the present application. It involved four SMEs with different commercial activities:
- Skyscan is specialised in CT hardware production, with inbuilt software;
- Simpleware is specialised in image-to mesh conversion;
- Cadfem is specialised in industrial and medical finite element analyses; and
- Inmatrixs is specialised in calcium phosphate biomaterial production.
These SMEs encountered repeatedly the following scientific / technological problem. While various options were at their disposal to derive, from the CT data, geometric models and corresponding finite element meshes, the mechanical properties of the material contained in the voxels remained largely obscure. Basically, they were guessed on the basis of experience or of some back-analysis; or some additional experimental work would be necessary to have access to them (which is sometimes expressed in terms of empirical regression functions between CT values and mechanical properties). However, without reliable information on material properties, FE analyses remain of rather qualitative nature, and their suitability for quantitative computer-aided design of biomaterial implants remains very limited.
On the other hand, the demand for reliable, cost-effective biomaterial design (including the wide realm of tissue engineering and regenerative medicine) is very high so that, once an SME would have a reliable access to material properties as input for FE analyses, this SME's competitive position on the market would be strongly strengthened. The same type of software would strongly affect the field of computer-aided radiology and surgery (as well as the field of simulation-supported surgery pre-planning), where complex, anisotropic and inhomogeneous material behaviour is standardly encountered; and the markets are huge in both cases.
The aforementioned access to voxel-specific reliable material property information, directly from the CT data themselves, and the integration of this information into commercial software were the central objectives of BIO-CT-EXPLOIT.
Therefore, latest conceptual research results of a highly qualified, interdisciplinary research consortium encompassing engineers, physicists, material scientists, and mathematicians were substantiated into clear and easy-to-understand computer codes which fit smoothly into the highly developed software products of one SME producing CT scanning instruments (Skyscan) and two SMEs developing simulation software (Simpleware and Cadfem) (with focus on image-to-mesh transformation and on finite element simulation technology, respectively). The results obtained from a first use of this new software for biomaterial design purposes was transferred to a fourth SME (Inmatrixs), active in biomaterial development and production.
Substantiation of cutting-edge, interdisciplinarily acquired concepts linking X-ray physics consistently to multiscale engineering mechanics and to applied mathematics, into robust, easy-to-understand software products is a veritable R&D task which can neither be attained by a single SME, nor by an SME consortium: Therefore, this task had necessarily to be outsourced to a multidisciplinary team of RTD partners representing their respective scientific fields at the world reputation level.
The following project objectives were tackled during the project duration of two years:
1. production and CT-scanning of various biomaterials, such as porous hydroxyapatite globules and rapid-prototyped tissue engineering scaffolds made of calcium phosphate-impregnated polymers;
2. collection and production of various bone material scans, comprising human and mouse femurs, and human mandibles;
3. development of software prototypes for the conversion of voxel-specific X-ray attenuation coefficients to elastic properties, falling into three categories:
- clearing of CT data from effects of instrumental resolution;
- identification of principal directions of anisotropic elasticity from object geometry / texture;
- translation of CT data into micromechanics-based elasticity tensors;
4. SME-specific image analysis tools, in form of software prototype releases concerning the following topics:
- clearing of CT data from effects of instrumental resolution;
- identification of principal directions of anisotropic elasticity from object geometry / texture;
- translation of CT data into micromechanics-based elasticity tensors;
5. comprehensive mechanical testing (quasi-static, nanoindentation, ultrasound) of polymer- and ceramic-based biomaterials; and
6. collection of software demonstration and clinically relevant 3D simulations of bone and bone biomaterial systems.
Project results:
The principle milestones achieved during the project were the following:
1. production and CT-scanning of various biomaterials, such as porous hydroxyapatite globules and rapid-prototyped tissue engineering scaffolds made of calcium phosphate-impregnated polymers;
2. collection and production of various bone material scans, comprising human and mouse femurs, and human mandibles;
3. development of software prototypes for the conversion of voxel-specific X-ray attenuation coefficients to elastic properties, falling into three categories:
- clearing of CT data from effects of instrumental resolution;
- identification of principal directions of anisotropic elasticity from object geometry / texture;
- translation of CT data into micromechanics-based elasticity tensors;
4. SME-specific image analysis tools, in form of software prototype releases concerning the following topics:
- clearing of CT data from effects of instrumental resolution;
- identification of principal directions of anisotropic elasticity from object geometry / texture;
- translation of CT data into micromechanics-based elasticity tensors;
5. comprehensive mechanical testing (quasi-static, nanoindentation, ultrasound) of polymer- and ceramic-based biomaterials; and
6. collection of software demonstration and clinically relevant 3D simulations of bone and bone biomaterial systems.
Concerning items 3 and 4, the following details are noteworthy. Regarding (1), instrument-related statistical distribution of X-ray attenuation coefficients were represented through a spline smoothing approach, and then de-convolved from the CT-images, based on the technique of histogram equalisation. Improvement of the signal-to-noise ratio led to a robust and quick software prototype. Regarding (2), an algorithm for interpolation between directions on organ surfaces and directions of anatomical guiding lines was developed. It was implemented in form of a highly modular code structure, which can be used as new add-on module within same larger program system, or as a stand-alone software tool. Regarding (3), the volume average rule for X-ray attenuation coefficients was employed to convert CT-data into composition data, which gave, via micromechanics, access to voxel-specific elasticity values. Thereby, emphasis was laid on high predictive precision of micromechanics formulation (as validated through experimental data), and on compatibility with the diverse software environments used by the involved SMEs.
In more detail, the milestones were reached as part of a number of work packages (WP):
The key objectives of WP2 'Software and data interfaces: From RTDs to SMEs' were:
(1) definition of file formats in which SME partners provide input data for RTD-developed software;
(2) definition of software interface through which RTD partners read SME-delivered data, for their evaluation through the new software of WP3.
Concerning the first of these objectives, input data in form of CT images were provided in TIF, BMP, and DICOM file formats. The consortium compiled a collection comprising both projection and reconstructed CT images. This collection comprises micro- and nano-CT images of porous polymer-ceramic biomaterials fabricated at WUT, micro-CT images of a mouse femur provided by Skyscan, micro-CT images of a human femur provided by Simpleware and scanned by Skyscan, as well as standard CT images of male and female human mandibles, provided by HAW.
Concerning the second of the aforementioned objectives, the consortium agreed on the development of stand-alone software prototypes in form of MATLAB-based graphical user interfaces (GUIs) for clearing clearing of CT data from effects of instrumental resolution, and for translation of CT data into micromechanics-based elasticity tensors, and in form of C++ code for identification of principal directions of anisotropic elasticity from object geometry / texture, these prototypes allowing for input data in DICOM, TIF, JPG, PNG, and BMP formats (as regards any form of CT data), and in STL-format (as regards surface data of human mandibles).
The key objectives of WP3 'Physico-mechanical software development' left for investigation in the period from month 12 to 18 were:
(1) identification of instrument-related statistical distribution of X-ray attenuation coefficients ('line broadening');
(2) correction of spatial distribution of X-ray attenuation coefficients from instrument-induced effects;
(3) translation of CT data into voxel-specific material composition values of bone and bone biomaterials;
(4) translation of material composition into voxel-specific elastic properties of bone and bone biomaterials;
(5) reconstruction of trajectories of anisotropic elasticity for bone and bone biomaterials;
(6) material anisotropy definition in finite element models.
The first of these objectives relates to microCT images with an air-filled homogeneous pore space and an inhomogeneous solid phase. The probability density function (pdf) of corresponding X-ray attenuation coefficients is characterised by two peaks ('air peak' and 'solid peak', with certain widths). There, our work started, for the sake of mathematical simplicity, with approximating of both the line boardening of the air peak and of the solid peak, by Lorentzian functions. A corresponding low accuracy in some cases motivated the use of peak model functions (PMFs), which are more precise, but the numerical fit of the bimodal pdf still poses problems. As an interesting solution to this problem, a spline smoothing approach was developed, which allowed for separation of the two peaks, for precisely fitting the numerically defined pdf, and for smoothing out statistical fluctuations.
Also the second of the aforementioned objectives relates to microCT images with an air-filled homogeneous pore space and an inhomogeneous solid phase. Although the air properties are homogeneous, the air peak is no Dirac function (as it should be theoretically), and the actually observed distribution is used to correct, through de-convolution , the solid voxel distribution from errors related to the experimental device inclusive of the reconstruction algorithms applied to original CT projection data. In this context, the determination of spatial distribution of corrected specific attenuation values was a major challenge, which was finally met through the technique of histogram equalisation. Other important aspects were the identification of voxel outlines and image masking for the identification of the object of interest.
As regards the third of the aforementioned objectives, our starting point was the linear relation between grey values given by standard CT scanners like the ones produced by Skyscan and X-ray attenuation coefficients. Combination of this relationship with the volume average rule for X-ray attenuation coefficients (the X-ray attenuation coefficient of a voxel is the volume average of the X-ray attenuation coefficients of all material components building up this voxel) led to an average rule for grey value. This average rule was then applied to three material classes with distinct chemical composition characteristics: porous ceramic biomaterial, extracellular bone matrix, and macroscopic bone material. This allowed for determination of voxel-specific composition, in terms of nanoporosity, of mineral and collagen content, and of vascular porosity. Respective key input parameters are the nanoporosity of the densest voxel in a ceramic biomaterial, the mean mass density of extracellular bone matrix, and the attenuation coefficient related to extracellular bone matrix.
Concerning the fourth of the aforementioned objectives, homogenisation schemes in the framework of continuum micromechanics were used to translate porosities or mineral / collagen contents into corresponding elastic properties. Micromechanical models appropriate for ceramic biomaterials, as well as for bone at the micrometer and at the milimeter scale were implemented in a computationally efficient manner.
As concerns the fifth of the aforementioned objectives, principal material directions were derived from interpolation between surface directions and directions of anatomical guiding lines within arbitrary organs. Additional software was provided for three types of guiding line construction in human mandibles:
(1) 'anatomical' guiding lines respecting full anatomy;
(2) 'rudimentary' guiding lines based on characteristic anatomical landmarks; and
(3) guiding lines which were adapted from a gallery of typical mandibles, to the actually considered organs.
Emphasis was laid on an efficient implementation in the C++ class structure.
As concerns the sixth of the aforementioned objectives, the output of the aforementioned software tools is designed in a way which is compatible with standard finite element packages used in our consortium, such as ANSYS and ABAQUS.
These issues were tackled in the first half of the project (months 1 to 12). Afterwards, the period from month 12 to 18 was devoted to a feasibility study on further development of corresponding software tools, beyond the level of the prototypes delivered to the SME partners in WP4. The aforementioned feasibility study extended the discussion on material anisotropy from the level of natural bone systems, to that of tissue engineering systems. Thereby, both the aspects of tissue engineering scaffold design and those of clinical realisation of tissue engineering solutions were tackled in an unprecedented fashion: as concerns biomaterial design, micro finite element analyses derived from micro-CT scans were identified as the proper technological solution for the determination of scaffold anisotropy, whereby a new way to arrive at the elastic properties of polymer-ceramic scaffolds was developed: their voxel-specific attenuation coefficients were transformed, based on X-ray attenuation averaging rules, into voxel-specific ceramic and polymer volume fractions, and the latter entered a new micromechanical formulation representing the micro and nanocrystals dispersed in a contiguous polymer matrix, this formulation delivering the sought voxel-specific elastic properties. As concerns clinical tissue engineering solutions, a clinical case provided by Inmatrixs, namely surgery and biomaterial treatment of a cystic mandible, was analyzed with respect to tissue inhomogeneity and anisotropy: sequential analysis of CT data sets revealed strong development of new bone tissue induced by the full load carrying biomaterial, to occur around six months after surgical intervention. This bone development was spatially inhomogeneous, resulting in the evolution of a new 'cortical shell', as opposed to a less dense, more trabecular bone-like material in the inner part of biomaterial-filled cystic volume. Furthermore, the algorithm for material direction identification, developed during months 1 to 12 for mandibles without large defects, turned out to be fully applicable to the investigated mandible with a large cystic defect - this being a clear step beyond the aims originally foreseen for BIO-CT-EXPLOIT.
The key objectives of WP4 'Software and data interfaces from RTDs to SMEs' were:
(1) definition of file formats in which RTD partner-developed software provides input data for the commercial software environments of the SME partners;
(2) definition of commercial software interfaces through which SME partners read data produced by RTD partner-developed software.
These two objectives were met by the RTD partners through development of SME-specific solutions for integration of newly developed software prototypes into the commercial software of the SMEs:
As regards the interface to ScanFE, a software product developed and distributed by Simpleware, the BIO-CT-EXPLOIT software for clearing CT data from effects of instrumental resolution has been developed in the MATLAB environment from where it was compiled for distribution as a MATLAB-based stand-alone executable. In the context of a GUI, the software is fully compatible with various image data formats, and performs fully automated preprocessing steps such as data reduction, outlier removal and object of interest identification. In addition to the corrected image files, various informations on the scanned objects themselves, such as pore radius frequency plots, are provided in different formats, as specified by the user. In full compatibility with ScanFE's material property input box, the BIO-CT-EXPLOIT translation of CT data into voxel-specific complete micromechanics-based elasticity tensors was provided in form of power-functions relating attenuation coefficients and mass densities, as well as mass densities and elastic properties. Corresponding factors and powers were derived from combined X-ray physics-micromechanics methods, and as a result, fully explicit scalar formulae, as well as GUI-produced data sets, were provided. Since ScanFE is tailored for image-to-mesh conversion, the BIO-CT-EXPLOIT software for identification of principal directions of anisotropy in human mandibles may provide these material directions directly on a voxel-per-voxel basis.
As regards the interface to ANSYS, a software product distributed by CADFEM, the error clearing aspect of the BIO-CT-EXPLOIT prototypes is dealt with similarly as described before for ScanFE, while the image-to-property conversion (concerning the effects of both material inhomogeneity and anisotropy) is realised such that the software provides ready-to-use ANSYS code, be it for ceramic biomaterials, or for human mandibles exhibiting inhomogeneous and anisotropic material properties. Therefore, the user is equipped with GUIs built on MATLAB code, available as stand-alone executables.
As regards the interface to CTan(alyzer), a software product developed by Skyscan and distributed with the sale of microCT desktop machines, both the error clearing concept and the image-to-property conversion tool were re-programmed from scratch, as to be fully compatible with the C++ source code at the foundation of CTan. As an interesting side product of this re-coding step, a considerable speed-up needs to be mentioned.
The key objectives of WP5 'Processing and CT imaging of biomaterials' were:
(1) processing of state-of-the-art bone biomaterials and tissue engineering scaffolds (ceramic and polymer-ceramic composites);
(2) micro and nano CT imaging of bone biomaterials.
As concerns the first of these objectives, porous hydroxyapatite granules at different sizes, and calcium phosphate-impregnated polymer scaffolds were produced.
As concerns the second of these objectives, the aforementioned materials were scanned in micro and nano-CT scanners, and the resulting data were distributed to the consortium partners.
The key objectives of WP6 'Mechanical testing' were:
(1) nanoindentation tests on state-of-the-art bone biomaterials;
(2) uni- / triaxial (poromechanical) tests on bone biomaterials;
(3) acoustic (ultrasonic) tests on bone biomaterials.
These objectives were met in two parallel experimental programs, devoted to biomaterials consisting of calcium phosphate ceramics only, and to rapid-prototyped ceramic-polymer-based biomaterials, respectively.
Due to their spherical shape, the ceramic globules produced by Inmatrixs could not be readily subjected to standard biomechanical testing protocols, and a novel protocol was developed (going beyond the original plan given in BIO-CT-EXPLOIT Annex I). The globules were subjected to quasi-static micro-splitting tests, which were accompanied by sequential microCT scans. In this way, thanks to recent equipment developments undertaken by SKYSCAN, the 3D fracturing behaviour of these biomaterials under mechanical load were, for the first time, investigated, and the corresponding test data provide a valuable basis for validation of any numerical simulations on hydroxyapatite globules. As a rule, fracturing processes were initialised at comparatively large pores inside the globules, as well as at microcracks stemming from the production process.
On the other hand, the rapid-prototyped, ceramic-polymer-based biomaterials were subjected to a multitude of different mechanical testing procedures (nanoindentation tests; quasi-static loading / unloading as well as poromechanical tests, with and without simultaneous microCT imaging; as well as ultrasonic tests at different frequencies), providing in this way a rarely seen completeness of mechanical characterisation. In particular, all methods were investigated concerning their consistency, and the unloading experiments were identified as those revealing elasticity (confirmed by acoustic testing), whereas loading tests are characterised by significant inelastic processes.
The key objective of WP7 'Demonstration' was the demonstration of the new software modules. These demonstration activities fell into two parts:
(i) showing the proper functioning of the newly developed software prototypes; and
(ii) numerical simulations based on these software prototypes, tackling real-life problems.
As regards proper functioning of BIO-CT-EXPLOIT software, the error clearing tool has fully met and partially even surpassed the original requirement catalogue of Annex I, including the following aspects: multipeak materials can be treated, high-speed C++ versions are available, clearing option is extended to polygon-based artefact removal. The code for principal direction identification works smoothly on images of strongly varying physiological cases. Attenuation-to-elasticity conversion was successfully demonstrated for a human mandible affected by a cyst, for a mouse femur, and for a ceramic biomaterial globule.
As regards demonstration of the new software for real-life applications, several large-scale simulations were performed:
(i) a fully inhomogeneous and anisotropic simulation of biting with a cyst-affected human mandible, delivering stress and strain profiles as function of simulation sophistication;
(ii) a high resolution microFE analysis of a mouse femur, showing the relevance of composition-driven inhomogeneity of elastic properties, for realistic estimation of stress profile in a physiological load case;
(iii) various high resolution simulations of ceramic, as well as ceramic-polymer biomaterial and tissue engineering scaffolds, again showing the relevance of composition-driven inhomogeneity of elastic properties, for realistic estimation of stress profiles under physiological strain.
In all these case, emphasis was also put on proper visualisation of the results.
The key objectives of WP8 'Dissemination and training' are:
(1) training of SME partners;
(2) identify and organise activities to promote the commercial exploitation of the results and the widest dissemination of knowledge possible;
(3) exploitation of the results in the health sector (clinics, implant manufacturing).
These objectives were tackled in a variety of activities:
- Three workshops were organised, each of them focussing on the use of the newly developed software in one of the three SME partners Skyscan, Simpleware, and Cadfem. In order to also maximise the audience from the side of the SMEs' customers, all three workshops were held in the framework of the SMEs' annual users meetings.
- The SME partners were provided with manuals describing the functionality of the new software prototypes, and how they can be used within their commercial software environment. In addition, the CT-scanning experience gained during the project was summarised in terms of guidelines for effective micro- / nanoCT scanning.
- Wide dissemination of knowledge gain was achieved by completion and publication of several papers for international scientific journals, and by presentations on BIO-CT-EXPLOIT topics at international conferences. The key challenge, namely not to present confidential details harming the market position of the SMEs, and to have results published within the relatively short project duration of two years, was successfully met.
- The wide public was informed, by press release, about the successful completion of the BIO-CT-EXPOIT project, and the message on the project has been well spread also during the project duration, as it has entered several brochures which are publicly available on the internet. In particular, BIO-CT-EXPLOIT was featured in the fit-for-health campaign (please see http://www.fitforhealth.eu/success-stories.aspx for details).
Potential impact:
Today the finite element method (FEM) is an integrated part of the development process in countless industrial activities, where its application helps the engineer to faster develop new designs, to increase the quality of new products, to reduce the development costs, and to bring goods faster to the market. A potential new market for FEM technologies has emerged: the health care market. There, FEM is used particularly in patient / individuum-specific simulations for diagnosis and therapy. However, FEM programs are still complex to use, and only trained people with a technical background can use the software. However, in order to enter the broad medical market, also the 'normal' doctor has to be directly addressed. Therefore, the FEM software has to be first of all easy-to-use, possibly automated, robust, and fast to apply. To meet this goal, CADFEM has established, in 2007, a new business unit: CADFEM Medical. Its aim is to identify surgeries where FEM can bring a benefit. For these surgeries, vertical applications based on standard and established FE packages like ANSYS are developed. These vertical applications are intended to fulfil the aforementioned requirements, such as easy use, high automation, robustness, and speed. However, CADFEM is facing new problems when addressing medical applications, problems which are not encountered in the 'classical' engineering field. The biggest problem is for sure the determination of reliable patient / individuum-specific material properties. And this is exactly where the project results of BIO-CT-EXPLOIT, namely the determination of realistic material properties out of CT data, turn out as the critical key to successful and significant penetration of the biomedical markets; and finally to establishment of the FEM in these markets - with CADFEM at the very forefront.
While the eminent strength of CADFEM lies in finite element solutions, our second simulation tool producer, Simpleware, is world-leading expert in conversion image data into geometric information and meshes. Simpleware software (ScanIP and ScanFE) provides state-of-the-art tools required for processing 3D imaging data as obtained by imaging modalities such as MRI and CT, for segmentation and identification of volumes of interest, as well as for automated generation of finite element meshes to be used in a wide range of third party solvers such as ANSYS (CADFEM). On the other hand, as for the traditional product line of the Simpleware, material property determination is largely left to the user. With BIO-CT-EXPLOIT having targeted at the determination of realistic elasticity properties from CT-data based micromechanics, Simpleware's portfolio in realistic simulation tools for complex problems related to BIO-structures has been significantly extended, opening new commercialisation routes not only in the biomedical, but also in the materials science market. In fact, very recently the latter market has gained significant importance for Simpleware.
BIO-CT-EXPLOIT also fits well into the market strategy of our CT instrument producer SKYSCAN, being quite distinct from those of the simulation tool providers. As a micro and nano-CT manufacturer Skyscan does not sell its software as a self-contained product, but as complement to CT instruments. Nevertheless, software can become one of the most important topics in a sale. Desktop microtomography systems always use polychromatic X-ray sources. The latter provide the possibility to use a micro-CT equipment in a laboratory environment, as an interesting alternative to a huge synchrotron facility. Unfortunately, polychromatic sources compromise the attenuation coefficients, making quantification difficult. BIO-CT-EXPLOIT has opened up new avenues for better interpretation of polychromatic X-ray values, based on the correction scheme for clearing CT data from instrumental effects including reconstruction algorithms. This is expected to already attract new customers deciding for desktop systems, while it will also be of great help for existing users. BIO-CT-EXPLOIT has evidenced the beneficial effects of new developments, such as special stages for performing in-situ measurements on objects being deformed in compression and/or tension. These in-situ measurements will also open up an important new market. In addition, the BIO-CT-EXPLOIT project results will enable SKYSCAN to offer new services to existing and new users within a short time frame.
Again a different market strategy is followed by our biomaterial producer, Inmatrixs, a highly innovative, young company which focuses on the processing of calcium phosphate scaffolds for stem cell research and tissue engineering research. The structure-property relations revealed through BIO-CT-EXPLOIT allow for optimisation of the mechanical performance and biological behaviour of these biomaterials. Use of entirely new, unparalleled design tools will allow for a significantly accelerated development cycle. This, in turn, will reduce both overall costs and time to market of Inmatrixs products, which will be directly used in the production (at least on pilot scale) and also in a clinical context.
Taking a larger view, BIO-CT-EXPLOIT has helped to fulfil a number of societal objectives in the European Community. Improved surgery planning or tailoring of tissue implants and scaffolds - through improved CT-data exploitation for simulation purposes - improves health and life quality, by significant reduction or repair of patients' disabilities or handicaps, together with decreased medical complications due to e.g. lower probability for further traumatical interventions. This affects particularly the largely growing population of elderly people in Europe. It improves the safety and mobility of this part of the population, enabling them to take a more active part in attacking Europe's challenges. Development of improved biomaterials for more economic and safer performance in living systems creates new industrial branches, helping to increase employment rates. Optimising materials leads to longer service life of permanent implants or of biological tissue grown on resorbable scaffolds, which both alleviates adverse effects on the environment. Finally, engineering within a living system requires its utmost comprehension: The highly relevant new knowledge obtained through BIO-CT-EXPLOIT will improve Europe's education and training standards. This is supported by scientific publications in international journals, guaranteeing the spreading of BIO-CT-EXPLOIT knowledge throughout the world-wide research and higher education scene.
World-wide dissemination of the project results is also eased through a world-wide network of companies, called TechNet Alliance (see http://www.technet-alliance.com for details), built up by CADFEM and including also our SME partner Simpleware. The members are usually companies offering in their local market a portfolio similar to those of CADFEM and Simpleware. If one TechNet partner feels to have a promising product the TechNet members can be used as a world-wide distribution channel, comprising countries such as Austria, Belgium, Brazil, China, Czech Republic, France, Greece, India, Ireland, Italy, Japan, Luxemburg, the Netherlands, Poland, Spain, Switzerland, Turkey, UK, or USA. All in all TechNet has more than 50 members in more than 20 countries. Linking all these partners to the biomedical market will guarantee an optimum world-wide spreading of the results obtained in the framework of BIO-CT-EXPLOIT.
Project website: http://bio-ct-exploit.imws.tuwien.ac.at