Community Research and Development Information Service - CORDIS


BERTI Report Summary

Project ID: 605162
Funded under: FP7-PEOPLE
Country: Germany

Periodic Report Summary 1 - BERTI (Biomedical Imaging & Informatics – European Research and Training Initiative)

Mission, Vision and Objective

The innovative doctoral programme BERTI: Biomedical Imaging & Informatics – European Research and Training Initiative aims at setting new standards to train Early Stage Researchers (ESRs/ESRs) within the triangle of the three research domains natural sciences, informatics & engineering and medicine, in order to meet new requirements related to biomedical questions in industrial or clinical practice. With this interdisciplinary education, the transfer of knowledge in areas adjacent to the core fields of study as well as with the acquisition of transferable business competences by our ESRs, we aim to contribute to successful, innovative research and excellent science. To achieve this ambitious goal, 15 partners combined their core competences to offer fourteen ESRs a unique and outstanding initial training network. TUM – Technische Universität München (Technical University of Munich) is the sole beneficiary, the associated partners are: GE Global Research Europe, UnternehmerTUM GmbH, Klinikum rechts der Isar der Technischen Universität München, Kinderspital Zürich – Eleonorenstiftung, Université de Lorraine, Max-Planck-Gesellschaft zur Förderung der Wissenschaften, Erasmus Medisch Centrum, Johns Hopkins University, Cardiff University, Helmholtz Zentrum München - Deutsches Forschungszentrum für Gesundheit und Umwelt GmbH, Columbia University in the City of New York, McKinsey & Company, Bayrische Patentallianz, Vossius & Partner and Scholz & Friends.


Milestone 1 ‘Successful Completion of Preparatory Phase’ (completion of literature research, setup of measurement systems) as well as milestone 2 ‘Successful completion of Implementation Phase’ (first prototype, first proof-of-concept) has been completed for all ESRs, except for ESR 1.4 who started in August 2015. Over half of the ESRs have already completed Milestone 3 (Successful completion of first dissemination phase), some are already in milestone 6 -their internalisation phase at their training partner.

Within less than 18 months, the BERTI ESRs have been able to deliver solutions, which are getting high academic and industrial recognition. As an example, future neuro and cardiac MR imaging methods or MR image guided therapies we will see in normal clinical use will be based on technologies currently explored, developed and evaluated by the team of early stage researchers in BERTI.

Work Package 1

MR Neuro employs advanced MR imaging methods, accelerated by compressed sensing, to study degenerative changes in the brain.
The first three Early Stage Researchers of Work Package 1 show good progress in accordance with the work plan. They familiarized themselves with magnetic resonance physics and identified the key research areas of their thesis project. First experiments – in vivo and in silico – are on their way. All three ESRs presented results at international conferences. ESR 1.4, who just recently joined the project is on good way for familiarizing herself with her thesis topic.

Work Package 2

Work Package 2 includes the whole Cardiac MR work process, from the direct programming of the MR sequences on the scanner device, over the compensation of possible motion artefacts during the acquisition time, to reconstruction of spatial and temporal models of a heart from a sequence of acquisitions. The information flow is bidirectional in the project: On the one hand, each consecutive module builds upon the results of the previous one. On the other hand the necessary compensation steps in the higher modules help to accelerate and steer the processing in the modules below. The ESRs finished their phantom studies on the scanner systems from GE and at the TUM and they are now validating their algorithms on patient studies during their international phase, together with our partner university in Nancy. The results from the Work Package 2 were already presented in an oral presentation at the leading conference in the field of MR ISMRM (International Society for Magnetic Resonance in Medicine) and the motion compensation work is considered to be submitted as a patent. The ESRs were already invited to participate in winter schools on the topic with external funding and are invited for additional stays in international labs.

Work Package 3

Work Package 3 is spilt into motion flow measurements and thermometry.
ESR 3.1 started his doctoral project in July 2014 on cardiovascular flow measurements using phase contrast MRI. Focus of the project is the development of a 4D flow MRI technique, which implies phase contrast MRI in 3 velocity directions in a 3D volume. The goal is to measure and visualize the blood flow pattern in and around the heart. His work includes the development of 4D flow MRI with respect to acquisition time and accuracy, phantom testing, and practical implementation of the technique in the clinical environment of Kinderspital Zürich.
ESR 3.2 works on thermometry. After learning about the current research in real-time interventional MR temperature mapping balanced steady-state free precession (bSSFP) was found to be capable of mapping PRFS and T1 using different inversion time delays after inversion preparation. This allows for temperature quantification in aqueous and adipose tissue types. This technique is investigated with means of Bloch simulations. Promising simulation results and initial temperature dependent T1 measurements in a piece of pork lard are successfully submitted to the conference of ISMRM and presented in Toronto in June 2015. Post processing routines are written in MATLAB to calculate T1 by a curve-fitting algorithm based on the trust region method.

Work Package 4

In Work Package 4, X-Ray Tensor Tomography, both ESRs have successfully completed the preparatory phase. In particular, literature research was performed and the setup of the experimental hardware and software was completed. On the hardware side, the X-ray phase-contrast tensor imaging setup is now fully working and first data sets have been acquired. On the software side, simulation of data acquisition is possible, and the entire image-processing pipeline is fully functional.
Both ESRs are currently working on the implementation phase. A common, preparatory part for both ESRs was the adaption of a component-based tensor reconstruction algorithm (Malecki et al., Eur. Phys. Lett. 2014) to a massively parallelized X-ray forward and backward projector, which has been completed and now allows fast, component-based tensor reconstructions.
In summary, both ESRs are progressing well and according to schedule. Currently, no delays on the deliverables D.4.1 and D.4.2 for month 24 are expected.

Work Package 5

Work Package 5 deals with optoacoustic imaging.
ESR 5.1 works on the development of combined ultrasound-optoacoustic endoscope and external probe for clinical applications. One of the fundamental problems to be addressed is developing accurate spectral processing tools for three-dimensional multispectral optoacoustic imaging. Two directions have been identified to develop methods for accurate spectral processing. Firstly, metrics would be developed to understand spectral quality. Later, methods would be deployed that can effectively correct spectral measurements by utilizing “standard measurements” (acquired as reference from the field of view in MSOT (Multispectral Optoacoustic Tomography) using known optical propertied) for normalizing and standardizing the quality of spectral measurements offered by MSOT. These corrections would be independent of other parameters like stability of the laser employed or parameters that could affect MSOT tomography / inversion methods. Developed methods would be analysed and the output of this phase will be reported as a journal publication.
ESR 5.2 focuses on the investigation of hybrid ultrasound - optoacoustic imaging. Currently, the research is focused on characterizing the frequency-domain OA response of different tissue-mimicking phantoms. Phantoms are prepared using mixtures of water, ink and intralipid with concentrations giving precise optical properties (absorption coefficient and scattering coefficient) mimicking human tissues. Initial measurements of the frequency response of the ultrasound (US)-generating volumes were done using an oscilloscope, which was not robust. Consequently, acquisition is repeated using in-phase/quadrature (IQ) demodulation technique to give more robust time-independent results.
The acquired measurement should be repeated with several illumination scenarios and transducers with different frequency ranges to reveal more information about the frequency response of tissue-mimicking phantoms using frequency modulated continuous-wave illumination. The output of this phase will be reported as a journal publication.

Work Package 6

Work Package 6 is about developing a snake like robot for minimal invasive surgery (MIS) for health care robotics. MIS is an intervention technique, which uses natural orifices or small incisions to access the surgical sites. This allows patients to recover in a shorter time period with less pain. As the surgeons usually don’t have direct access to the surgical site as in open surgery, they have to rely on devices. At this point robotic assisted surgery becomes an important tool to assist surgeons in terms of control and visualization in general. Clinically applied examples can be found in urologic and gynecologic surgeries. However, the existing robotic technology is challenged by the endonasal approaches in ear, nose, and throat (ENT) surgery that provide various anatomic constraints. As a result of that, the prospective surgical robots for the endonasal applications should possess some key features such as being small in size, flexible, steerable and robust. Within the lab of WP6 medical components are manufactured with Selective Laser Sintering (SLS). For this purpose a biocompatible polyamide is used to manufacture the parts. The ultimate objective of this study is to develop a snake-like robot for minimally invasive surgery using biocompatible metals based on flexure hinges.

Training Activities

In Work Package 7 ‘Scientific Training’ the offered courses were adapted to the ESRs´ needs.
We introduced additional training for Work Packages 1 to 3 in MR imaging, trained only ESRs using EPIC software in an EPIC course, streamlined to colloquia series and combined it with numerous TUM Bioengineering Lecture Series and TUM Bioengineering Workshops Series.
Transferable skills training from Work Package 8 began in October 2014 after all ESRs started with their project. Additionally we organised a German class, that finished with evel A1 in German. This course teamed the group up even more.
Early on all workshops, lectures and trainings carried out in Work Package 7 and 8 have been scheduled and will be held in dedicated training months: November 2014, February 2015, July 2015, July 2016, October 2016 and April 2017 in order to enable the ESRs clear time slots for their secondment. More details on the very successful training can be found below.
An overview of all planned and accomplished events is available online on:
Events and Dissemination
The official BERTI Kick-Off Symposium took place on the research campus in Garching from January 18th to January 20th, 2014. The whole consortium attended this event and the program especially the guest speakers (Prof Carsten Welsch from the University in Liverpool and Prof. Peus from the Chair of Research and Science Management at TUM was very well appreciated. The speeches were followed by a vivid discussion among all participants.
In July 2014, eight ESRs -who had already started at TUM- participated in a two-day Kick-Off event, where all details of the programme were presented and explained to them. Apart from the finances, organisation of the programme and their personal expectations we also included a time slot for team building activities. The Kick-Off was a huge success and really did pay in on the BERTI identity of the ESRs.
The next section will concentrate further on the dissemination and branding strategy.
As first measure, a press release was formulated in July 2013. Secondly the BERTI management initiated the ITN’s website which illustrates and informs about the ESRs´ projects, provides information on planned training events, provides access to relevant documents for the ESRs as well as an up-to-date news section.
To further strengthen the brand building and development of BERTI and make the programme truly visible, the next initiative started was creating a logo for our brand ‘BERTI’. Being recognized and remembered is key to successful communication. Icons created for each of the work packages lead to a feeling of belonging, make it easier for external people to recognize and separate the various work packages and give the project the professional appearance it deserves. . A printed flyer has been developed in order to enable the ESRs and the consortium to become more visible within the participation and other interested institutions, on events, and conferences. One publication in TUM Campus (10.000 printed editions, high online visibility) described a very successful podium discussion that was held in BERTI on the ‘Digitalization in Health’. Another early printed publication was realised through the annual report of GSISH, now GSB. BERTI associated at the Graduate school of Bioengineering at TUM. Not only their research projects were published, but also a quick overview of the structure of BERTI was given.
Additionally BERTI regularly exchanges information and best practices with other European Research actions within TUM, and especially with ITNs beyond the TUM-wide network, mainly MacSeNet and the University at Liverpool (Prof. Carsten Welsch).

More information can also be found on


Alois Knoll, (Chair of Robotics / Director of TUM-GSISH)
Tel.: +49 8928918142
Fax: +498928910805


Life Sciences
Record Number: 184032 / Last updated on: 2016-06-09
Information source: SESAM