Skip to main content

Multimodal Imaging of Neurological Disorders

Final Report Summary - MINDVIEW (Multimodal Imaging of Neurological Disorders)

Executive Summary:
MINDVIEW project has successfully achieved its main objective: the development of a cost-effective working prototype of a high sensitivity and resolution scanner for simultaneous PET/MR imaging, dedicated to the brain examination.

The MINDVIEW approach to achieve simultaneous PET/MRI imaging is based on the development of a brain PET imager integrated into a purposely designed head dedicated transmitter/ receiver (TR) RF coil system. Therefore, MINDVIEW takes advantage of current MRI scanners and upgrades them with a brain dedicated PET/RF device. The dimensions of the full PET/RF system represent a balance between PET and RF performance as well as patient comfort, resulting into a portable and compact design.

The new head dedicated RF coil system has a birdcage TR design, with 16 coil rungs, providing a geometric aperture of 25 cm, ample for comfortable head allocation. The PET subsystem consists of 3 rings of 20 detector modules per ring, leaving a geometrical aperture of 33cm in diameter to allow enough vision to schizophrenia patients. Each detector module is made of a single LYSO crystal (5 cm 5 cm 2 cm), a SiPM (Silicon Photomultiplier) photo-sensor array and the related frontend electronics. The 3 rings cover an axial field of view (FOV) of 15,2cm. Coincidences of gamma rays between one PET detector module and any of the 9 opposite modules produces a transaxial FOV of about 24 cm, which ensures full coverage of the human brain.

The integration of the RF coil and the whole PET scanner was successfully carried out by late 2017 and installed at TUM in January 2018 (CSIC, ONCOVISION, NORAS, TUM). The performance of the PET system reached the expected values of around 1.5 mm spatial resolution at the centre (resolution of commercial clinical PET scanners is of the order of 4-6mm) and, due to its innovative depth of interaction information ability, did not degrade substantially towards the edges of the field of view. PET system sensitivity of about 7% almost duplicates the one observed in whole-body PET systems. Most importantly, the system performance did not suffer even when running at very demanding MRI imaging sequences. Derenzo-like and Hoffman phantoms were scanned both with the MindView PET and on standard whole-body PET systems. MindView exhibited an improved image resolution. Several image reconstruction algorithms approaches (ONCOVISION, U. BREST), including AC methods, have been studied and implemented in the course of this project. First volunteer human images were obtained at TUM and were very encouraging.

The clinical finding that reduced numbers or altered function of immune cells in brain appears in early stage schizophrenia is very relevant. However, the study also illustrates the problems in recruiting a large sample of first-episode medication free patients with schizophrenia. The use of 11C-carbon monoxide in new applications within a mini clean room and on new technology platforms has been demonstrated. The mini-clean room simplifies GMP production and will facilitate the access to various tracers. VAAT null mutant mice has proven to be an animal model with altered sensorimotor gating, a symptom associated to schizophrenia. Higher binding potential of 11C-raclopride has been measured with this model compared to wild type control animals.

Furthermore, the SMEs participating in MINDVIEW have further developed their core technologies during the project. This fact has contributed to the acquisition of at least a part of their businesses of two MINDVIEW industrial partners by large multinational companies. We believe this also represents a great success of the MINDVIEW project and a proof of well spent European tax-payers money.

MINDVIEW represents a real breakthrough in terms of new tools for imaging that will allow the definition of parameters allowing patient stratification in schizophrenia and depression diseases. This should also lead in the future to development of the next generation of drugs for treating those disorders. Furthermore, the new tools developed will allow for an extension of techniques that are currently only available to very few hospitals. We may conclude that the tools developed within the MINDVIEW project will facilitate and extend the technique of dynamic and simultaneous PET and MR imaging, opening new vistas for understanding mental disorders.
Project Context and Objectives:
MINDVIEW project has successfully achieved its main objective: the development of a cost-effective working prototype of a high sensitivity and resolution scanner for simultaneous PET/MR imaging, dedicated to the brain examination.

The MINDVIEW approach to achieve simultaneous PET/MRI imaging is based on the development of a brain PET imager integrated into a purposely designed head dedicated transmitter/ receiver (TR) RF coil system. Therefore, MINDVIEW takes advantage of current MRI scanners and upgrades them with a brain dedicated PET/RF device. The dimensions of the full PET/RF system represent a balance between PET and RF performance as well as patient comfort, resulting into a portable and compact design.

The new head dedicated RF coil system has a birdcage TR design, with 16 coil rungs, providing a geometric aperture of 25 cm, ample for comfortable head allocation. In order to minimize distortions in the MRI performance, the distance between the PET shielding and the RF rungs conductor is about 3 cm.

The PET subsystem consists of 3 rings of 20 detector modules per ring, leaving a geometrical aperture of 33cm in diameter to allow enough vision to schizophrenia patients. Each detector module is made of a single LYSO crystal (5 cm 5 cm 2 cm), a SiPM (Silicon Photomultiplier) photo- sensor array and the related frontend electronics. The 3 rings cover an axial field of view (FOV) of 15,2cm. Coincidences of gamma rays between one PET detector module and any of the 9 opposite modules produces a transaxial FOV of about 24 cm, which ensures full coverage of the human brain.

Achieving 1 mm spatial resolution in the central brain region with PET detectors close to the object, such as in MINDVIEW, is not trivial and requires depth of interaction (DOI) determination of the gamma rays inside the thick scintillating crystals, to avoid blurring produced by parallax error. Two major LYSO scintillator crystal configurations have been considered: a staggered pixelated crystal array and a monolithic approach. There are advantages and disadvantages for each configuration that have been described in the literature. We decided to build the prototype using monolithic blocks, as detailed elsewhere. For monolithic crystal designs, several DOI schemes have been proposed based on the correlation between the DOI and the measured light distribution.

All PET detectors required to build the prototype were produced (CSIC, OCV). Highly challenging mutual interferences between PET and MRI systems were successfully avoided. A high rate capability data acquisition system, including trigger and ADC boards for 60 detectors blocks, was developed and produced (OCV). Synchronization among all detectors was better than 200 ps. Row and column readout made it possible to characterize the scintillation light projections. An RF coil was successfully built (NOR) that is working within the new PET scanner (CSIC). The RF coil allows the usage of all needed MR sequences and shows a uniform image quality over the entire head.

The integration of the RF coil and the whole PET scanner was successfully carried out by late 2017 and installed at TUM in January 2018 (CSIC, ONCOVISION, NORAS, TUM). The performance of the PET system reached the expected values of around 1.5 mm spatial resolution at the centre (resolution of commercial clinical PET scanners is of the order of 4-6mm) and, due to its innovative depth of interaction information ability, did not degrade substantially towards the edges of the field of view. PET system sensitivity of about 7% almost duplicates the one observed in whole-body PET systems. Most importantly, the system performance did not suffer even when running at very demanding MRI imaging sequences. Derenzo-like and Hoffman phantoms were scanned both with the MindView PET and on standard whole-body PET systems. MindView exhibited an improved image resolution. Several image reconstruction algorithms approaches (ONCOVISION, U. BREST), including AC methods, have been studied and implemented in the course of this project. First volunteer human images were obtained at TUM and were very encouraging.

A mini clean room that simplifies GMP production and makes it more reliable and cheaper, aiming for an easy access to various tracers, has been developed. The mini clean room concept can be adjusted to the platform technology used and allows protocols to keep a GMP environment in control. In this project, the focus has then been to develop tracer technology on new platforms, mainly on the preparation of 11C-labelled tracers, which is adapted to the mini clean room concept in order to make the tracer production more applicable. In particular, the use of 11C-carbon monoxide in new applications and on new technology platforms has been developed. Both 11C-methyl iodide or 11C-carbon monoxide were used in the labelling of an established dopaminergic tracer like Raclopride, but now labelled in various positions to show the flexibility and potential of the labelling strategies explored. Furthermore, we have re-examined production methods for 11C-methyl iodide applied to these new technology platforms. In a monkey study, 11C-methyl iodide and 11C CO were used to produce 11C-raclopride labelled in 2 different positions, exploring its impact on the images due to differences in metabolism. The use of 11C-carbon monoxide is opening up new avenues to label many potential tracers.

To validate and extend on the use of radiotracers, we established a mouse model that affects the aminergic systems of the brain. Specifically, we used mice that had the allele encoding the vesicular aminergic associated transporter VAAT, or Slc10A4, deleted. VAAT adds functionality to aminergic neurons through a mechanism involving presynaptic vesicles. VAAT null mutant mice have been shown to develop normally without obvious abnormalities and normal motor function in spite of defective cholinergic synaptic transmission at the neuromuscular junction. However, VAAT null mutant mice show clear alterations in aminergic homeostasis, have increased susceptibility to a cholinergic form of status epilepticus and are hypersensitive to dopaminergic stimulants such as amphetamine. Moreover, decreased dopamine uptake has been observed in vesicle preparations from VAAT null mutant mice, while over expression of VAAT resulted in increased acidification of synaptic vesicles. These alterations led us to hypothesize that VAAT null mutants were affected in systems where aminergic functionality is a prerequisite, such as in behaviors related to the symptoms of schizophrenia. For this purpose, we here investigated whether the deletion of the VAAT gene in mice would alter sensorimotor gating, similar to the deficient sensorimotor gating seen in the dopamine overactivity rodent model of schizophrenia.

Mice were analyzed in different pre-pulse inhibition regimens; high amplitude (20 db above back ground noise) and short vs long inter-stimuli intervals. Mice that were either wild-type (WT, n = 17), VAAT knock-outs (KO, n = 21) or overexpressing the VAAT gene (NSE, n=13) displayed either normal or enhanced sensorimotor gating responses in the pre-pulse inhibition (PPI) test at 30ms inter-stimuli intervals (short IEI) at 85dB (high amp) pre-stimulus amplitudes. At 100 ms inter-stimuli intervals (long IEI) VAAT null mutant mice show increased sensorimotor gating. Mice overexpressing VAAT (Slc10a4 NSE mice) showed responses similar to wild type (WT) littermate controls. VAAT null mutant animals had a higher ability to filter, process and link the pre-pulse and startle pulse compared to control animals. In contrary, mice overexpressing Slc10a4 did not show any differences in the PPI test. Such mice were constructed to overexpress Slc10a4 under the neuronal specific enolase promoter, Slc10a4-NSE. Our results suggest that VAAT null mutant mice may be used as a small animal schizophrenia model based on their response to the pre-pulse inhibition (PPI) test.

We next evaluated uptake and distribution of radiopharmaceuticals in VAAT null mutant mice using established and novel radioligands in PET. We used the dopamine D2 antagonist 11C-raclopride, which is a reliable and robust radioligand with well characterized properties in small animal models. Importantly, 11C-raclopride has since long been used in clinical studies to evaluate patients with schizophrenia. In addition to dopamine, serotonin has also been associated to schizophrenia. Thus, we also used the novel serotonin 2A (5-HT2A) receptor agonist 11C-Cimbi. On the same experimental day, two PET measurements with 11C-raclopride and 11C-Cimbi, respectively, were conducted in each animal. We imaged four wild type mice and five VAAT null mutant mice with 11C-raclopride. Three wild type mice and five VAAT null mutant mice were imaged with 11C-Cimbi. In a preliminary analysis, we found that Slc10a4 null mutant mice have higher binding potential of 11C-raclopride compared to wild type control animals whereas no such evident effect could be observed for 11C-Cimbi binding. From the study on the binding potential with WT and VAAT mice it could be concluded that the alterations found in the monoamine homeostasis are more readily detectable with Raclopride than with Cimbi. Whether this is due to a larger effect on the dopaminergic system in Slc10a4 mutant mice or to the properties of the radiotracers used in this study remains to be established.

At the Karolinska Institutet there have recently been examined 16 drug-naïve, first episode psychosis patients and 16 healthy controls using PET and the TSPO radioligand 11C-PBR28 [41]. There was a significant reduction of 11C-PBR28 VT in patients compared to healthy controls in gray matter (GM) as well as in secondary regions of interest. In this limited sample, there was no correlation between GM binding and clinical or cognitive measures after correction for multiple comparisons. The observed decrease in TSPO binding suggests reduced number or altered function of immune cells in brain in early stage schizophrenia.
The clinical finding that reduced numbers or altered function of immune cells in brain appears in early stage schizophrenia is very relevant.

ECONOMIC AND SOCIAL IMPACT

Increasing the sensitivity of PET scanners is mandatory for routine imaging diagnosis and follow-up of patients with mental disorders. One way of increasing the sensitivity of current PET scanners is by augmenting significantly the axial FOV, which implies a very high associated cost. The approach followed by the MINDVIEW project, i.e. integrating the PET system with the RF coil, has the advantage of increasing significantly the sensitivity of the scanner for the brain examination with a reduced number of detector modules. This alternative method of increasing the PET sensitivity has the potential of further extending the PET/MRI technique through upgrading the large MRI installed base with a cost-effective and portable brain PET/RF scanner.

Current PET/MRI systems are very expensive, only affordable by well-financed health centres. The market price of commercial PET/MRI systems is extremely high (between 4 and 7 million euros). Therefore, in spite of the medical usefulness of the technique and the commercial efforts made by the large multinational companies (SIEMENS, GENERAL ELECTRIC) since almost a decade, only less than hundred systems altogether have been installed at hospitals of excellence.

Cost is the main advantage of the new system. Our new design aims at upgrading the current MRI installed base with a brain dedicated PET, integrated in the Radiofrequency (RF) coil for the head MRI examination. Since more than 40% of the MRI examinations are related to the brain, the new device will represent a very useful and cost-effective device for the diagnosis of mental disorders and neurodegenerative illnesses. This will trigger an expansion of PET/MRI technology around the World.

There are more than 40.000 MRI systems installed around the World. Therefore, these systems might be upgraded with a brain dedicated PET, integrated with the RF coil for the head examination. Moreover, as already stated above, the new PET design is not only considerably less expensive but also presents better performance with respect to resolution and sensitivity.

This significant cost reduction will finally open the road for a widespread expansion of the very useful PET/MRI technique by making it much more affordable. Therefore, not only hospitals which are centers of excellence but also smaller hospitals that can not afford a new whole-body PET/MRI system might be able to acquire this powerful technology.

Furthermore, the SMEs participating in MINDVIEW have further developed their core technologies during the project. This fact has contributed to the acquisition of at least a part of their businesses of two MINDVIEW industrial partners by large multinational companies: SenSl, an Irish company that develops SiPM sensors, has been acquired by ON Semiconductor; the small animal PET business of ONCOVISION, a Spanish company that develops molecular imaging systems, has been acquired by BRUKER, a German multinational company. BRUKER is currently producing and distributing worldwide PET/MRI scanners for small animals which contain SenSl sensors and ONCOVISION PET technology. Stanford University has acquired one of those systems recently. We believe this also represents a great success of the MINDVIEW project and a proof of well spent European tax-payers money.

The significance of this project is substantial. Neuropsychiatric disorders affect a large part of the population, and cause a great deal of suffering both to the patients and to those close to them. Mental disorders represent a major social and economic burden in our developed societies. When commercialized, the new scanner will definitely benefit the patients with those illnesses by providing an earlier and precise diagnosis of the disease. Visualizing with high resolution neurotransmitter pathways and simultaneously the activated areas of the brain will open new venues for the understanding of the human mind and its illnesses.

CONCLUSIONS

Several important tools have been developed during the MINDVIEW project for the diagnosis and follow up of schizophrenia disease. In particular, an innovative and cost-effective 1,6 mm spatial resolution, over the entire FOV, PET scanner has been developed, which has been integrated with a head dedicated RF coil. It is estimated that a PET spatial resolution of about 1mm is required to answer specific questions relating to brain function in small brain regions and improve the diagnostic potential of the images. Therefore, we have achieved the required resolution, and at the same time increased the sensitivity, by designing a method to provide good DOI information, which is critical due to the close proximity to the brain. The high sensitivity of the PET scanner of 2,7% (state-of-the-art whole body PET systems provide sensitivities close to 1%) should allow for longitudinal examinations on the same patient.

The use of 11C-carbon monoxide in new applications within a mini clean room and on new technology platforms has been demonstrated. The mini-clean room simplifies GMP production and will facilitate the access to various tracers.

VAAT null mutant mice has proven to be an animal model with altered sensorimotor gating, a symptom associated to schizophrenia. Higher binding potential of 11C-raclopride has been measured with this model compared to wild type control animals.

The clinical finding that reduced numbers or altered function of immune cells in brain appears in early stage schizophrenia is very relevant. In addition, schizophrenia research would benefit from the potential to examine each patient with two or more radioligands targeting different neurotransmission systems or functional markers such as TSPO. Ideally these studies should be repeated during different states of the disorder. The sensitivity of the next generation PET-systems will be critical to allow for such extended protocols within as low as reasonably acceptable limits of radiation exposure.

MINDVIEW represents a real breakthrough in terms of new tools for imaging that will allow the definition of parameters allowing patient stratification in schizophrenia and depression diseases. This should also lead in the future to development of the next generation of drugs for treating those disorders. Furthermore, the new tools developed will allow for an extension of techniques that are currently only available to very few hospitals.

We may conclude that the tools developed within the MINDVIEW project will facilitate and extend the technique of dynamic and simultaneous PET and MR imaging, opening new vistas for understanding mental disorders. It will enable the full breath of activation paradigms to be used. These include dynamic imaging using activation paradigms as the perturbation method for studying the brain from several aspects: morphological, functional, metabolic, neurotransmitter and neuro-receptor effects. The ability to perform serial studies, i.e. examinations within the same subject, paves the way for personalized medicine such as early diagnosis of disease onset and therapy monitoring.
Project Results:
In the following there are summaries of the main scientific and technical results divided by the different work packages of this project.
3.1 WP2: Sub-mm resolution MRI-compatible PET detector system design and PET module production
Progress towards objectives WP2:
During execution of this work package, the PET system, to be integrated in the brain dedicated RF coil, has been designed and constructed, as well as successfully validated both with phantoms and patients. The primary technical objective of this work package has been the design and development of a PET module in a format necessary for assembling a complete PET/MR detection ring. This has been accomplished. The PET module contains the crystal, SiPM photodetector array and associated front-end electronics developed in WP3. As described in the first reporting period, we defined the first prototype to be built with 20 mm thick monolithic crystals, 12x12 SiPM arrays and a row and column readout providing scintillation light projections. The main tasks and achievements are mentioned in the following.
Task 1. PET system design: OCV and CSIC, together with NOR, have defined the geometry of the first PET system, integrating the RF coil.
Task 2. PET module detector configuration: CSIC and OCV have tested the two main configurations namely the three staggered layers of crystal arrays and the monolithic 20 mm thick scintillator. See attached Figure 1 and Figure 2.
Task 3. Photodetector Element Optimisation: SEN has developed and accomplished the design of SiPM which are MR compatible, significantly reduce the dark noise level, but keeping good gain and temperature stability, together with fast timing response.
Task 4. Production of SiPM pixel test devices: SEN has developed 12x12 SiPM arrays for test purposes used at CSIC and OCV. They have been successfully used with 20 mm thick monolithic crystals.
Task 5. Photodetector array optimization: The 12x12 array has been mounted on a PCB using conventional technics, which means without paying attention to MR compatibility.
Task 6. PET modules, scintillation blocks have already being acquired and SiPM parts already mounted. Characterization was carried out as follows. Figure 3 shows examples of the final scintillation crystal and SiPM array and PCB.

Highlighted significant results: During this reporting period the most significant achievement in this WP is the accomplishment of the final detector design, both at the crystal treatment level, the SiPM arrays that are MR compatible and the readout electronics.
All detectors were finally manufactured, and successfully integrated in the prototype. Their working performance was very good as it will be shown later.
3.2 WP3 PET Electronics
Progress towards objectives WP3:
The main objectives of this work package were the development and construction of electronic hardware including the PET electronics for the SiPMs and the PET data acquisition system (DAQ) hardware and firmware. The readout has to be MR-compatible and compact. Here, the PET data acquisition system (DAQ) hardware and firmware, should deal with a high number of channels and a Trigger supporting several detectors. All these tasks were successfully accomplished.
This work package also includes the development of hardware and software for tests and validation of the DAQ and the SiPMs and to optimize the components for the simultaneous operation of PET and RF coil. The combined components were delivered to OCV for system assembly and validation. There are not tasks involving the current reporting period, however given that further tests and studies were carried out with this electronics, we are providing here the latest findings. These are related to tasks 5 and 6 of WP3. The photographs of figure 4 show the electronic cabinet with ADCs and Trigger (on the left) and the cabinet inside the MR room (on the right) at the left side of the rear part.
The block diagram of figure 5 summarizes the acquisition system architecture. The main components are the ADC and Trigger cards. In order to make the whole system flexible and to speed up the development time, all cards have been designed in a modular way.
The whole system has been divided into three independent crates, consisting of 1 Trigger and 10 ADC cards. Each one of these crates collects information of a ring. There are 60 modules distributed in 3 rings with 20 modules each. In total there are 30 ADC cards and 3 Triggers.
Each PET detector module sends information with 32 analog channels organized in 16 rows and 16 columns. The acquisition card has been designed with 64 channels so that it can acquire information from 2 PET modules. The trigger block is designed on a single control card that sends the conversion start signals to each one of the connected cards.
Three front panels has been used in order to distribute all signals coming from the pet detector modules to each one of the acquisition cards in such a way that it is possible to add up to 60 pet modules arranged in 20 modules per crate. In the same way, three rear panels carry the power supply voltages and the conversion start signals coming from the trigger cards. Each ADC card has a 1G Ethernet port to send the information of each event. All Ethernet ports converge on a single port router that combines all the data packets and transmits to the computer the set of information through a 10G Ethernet port.

Highlighted significant results: The entire electronic cabinet works well, for as more as 60 detectors blocks. It works well also inside the MR room. The objectives of this WP were successfully reached, demonstrated by the PET-RF coil system performance.


3.3 WP4 RF coil development
The MindView approach is to achieve simultaneous PET/MRI imaging of the human head. The development thus aimed to design and build an integrated brain PET imager that is nesting a purposely designed head dedicated transmitter/receiver (TR) RF coil system. In the final period of the MindView Project the goal was to finalize the RF coil that is an integral part of the MindView PET-MRI device. As statet in the former deliverable a transmit-receive birdcage RF coil capable of sending and receiving the magnetic resonance signal (MRI) signal was built, see next figure. The RF coil was built and tested with the PET system. The final RF coil was made compatible with a 3 Tesla Siemens MRI system. In this case the Siemens MRI system is a so called PET/MRI system that has a built in PET system. The system was chosen for reasons that would allow a comparision between the MindView device and a commerically available system like the Siemens PET/MRI system. The components of the MindView project (the RF coil) and the PET system were combined and installed in January 2018 in Munich at the MindView partner TUM.
Final MindView RF coil design
The final dedicated RF coil system is made from a birdcage TR design with 16 rungs providing a geometric aperture of 25 cm. The system is equipped with a hybrid that allows splitting the inbound RF energy from the RF subsystem of the MRI system while simulatanously adding a 90 degree phase shift. The RF signal is then fed through two custom designed TR switches into the head end of the head end ring of the RF coil. The subcomponents are described below. The single subcomponents of the coil are then mounted on the inner housing (and the RF screen is added. The RF coil is then tuned und matched to a Larmor frequency of 123 MHz while the RF coil is loaded with a tissue-mimicking phantom.
Mechanical Design
The coil was designed to fit into the twenty corner polygon shape of the PET insert. The PET insert is shown in figure 8.
Final RF coil tests
Phantom tests. The RF coil was tested using a phantom mimicking a human head load. A uniform phantom was used to obtain signal-to-noise-ratio (SNR) data from the RF coil. The coil was combined with the PET system and installed at the parter site of the TUM Munic. The coil was loaded with a phantom bottle (see figure 9 attached) and SNR data was obtained.
The data shows a uniform circular polarization thoughout the phantom as it should be for a coil working in a quadrature birdcage configuration.
Human volunteer tests. First tests with the RF coil and the PET dummy system were conducted on a 3 Tesla Siemens MRI system. A human volunteer, male of 27 years was placed in the coil and common sequences for neurlogical imaging were tested, see figure 10.

Highlights: Currently the system is finished and successfully working at the TUM in Munich. See examples in the following images.

3.4 WP5 MindView PET/RF system integration
Progress towards objectives WP5:
The main objective of this WP is the integration of all the components of the brain dedicated MINDView PET/RF embedded system and the software development of the system. The first task was to design the portable system from the PET modules developed in WP2 and the head-dedicated RF coil developed in WP3.
The system consisted of the following elements: Mechanics to hold both the PET modules and the RF coil; RF system for brain studies; PET dedicated system for brain studies; Software for PET and image fusion of PET/MR.
OCV has coordinated the development of the system in partnership with CSIC (PET system development) and NOR (RF development).
The insert was finished in Autumn 2017. A calibration process was carried out at the bunker of the Polithecnical University of Valencia, and then shipped to TUM (Munich) in early January 2018. In the following we describe the different tasks.
Task 1: Innovative Mechanics for PET/MR compatibility. Hardware Interface Validation. The mechanics of the support for the dedicated RF coil and PET modules have been constructed. For these prototypes we have used a 3D printer combined with PLA material (plastic like).
Hardware Interface Validation. The mechanics of the support for the dedicated RF coil and PET modules have been constructed. For these prototypes we have used a 3D printer combined with PLA material (plastic like, see figure 11).

Task 2: PET Control and Calibration Software Development (OCV). This task includes the development of "slow"-control and initialization methods, user-level calibration and performance verification procedures and "online" displays. The picture in figure 12 shows the calibration tools and an example of a flood map during one of the detector calibration processes.
Task 3: PET Module Impact Position Estimation Software (UR). An innovative multi-parametric position algorithm has been developed permitting to fully recover the information released into the scintillation light. It is based on an iterative multiparametric estimator of the light spread before the position calculation. It takes into account distribution width, light threshold and shape, which is demonstrated to achieve a complete reconstruction of the position of photon interaction within the crystal position and deep of interaction.
Task 4: PET Image Software Development (UB, CSIC). Off-line reconstruction tools have been developed. There is software suite that allows acquiring and reconstructing data, applying different algorithms and corrections to the image such as scatter, random or attenuation.
An innovative multi-parametric position algorithm has been developed permitting to fully recover the information released into the scintillation light. It is based on an iterative multiparametric estimator of the light spread before the position calculation. It takes into account distribution width, light threshold and shape, which is demonstrated to achieve a complete reconstruction of the position of photon interaction within the crystal position and deep of interaction (figure 13).

Task 5: Software Development of MR-PET Image Fusion (UB).
This task has been driven by two partners UB and OCV. Both have made great progresses on image reconstruction and software developments. In the following there is a description of tasks carried out by both. UB has centred on the development of a new projector as described below, OCV has made it use of their experience in integrating the PET system, and software applications.
Taking advantage of the mechanically fixed MR coil of the MindView system, a registration has been implemented through a rigid transformation based on trilinear interpolation. This registration allows the use of MRI anatomical data. Additionally, when a MRI based attenuation estimate is provided (as it is the case of the Siemens mMR), these anatomical data are used to compute subject-specific attenuation correction. Specifically, the developed software takes into account three attenuation values: air, soft tissue and bone, for the corresponding emission energy. However it can be easily extended to an arbitrary number of different tissues. A CT scan of the MR coil has been registered and included in the attenuation correction computation, in order to account for the metallic and plastic parts of the coil. The coil has a mechanical guide that ensures its relative orientation with respect to the MindView PET detectors.
For PET image reconstruction, two optimizers have been used: Ordered Subsets Expectation Maximization (OSEM) and One-Step-Late (OSL). An Intrinsic Detector Response Function (IDRF) for the MindView system has been implemented, relating the output of the detector with the photon interaction coordinates. This function takes into account the scintillation process as well as other factors of the MindView detectors such as crystal geometry, coating and finish, or photosensor’s layout. An IDRF allowed the implementation of an Iterative Random IDRF Sampling (IRIS) projector within the reconstruction process. This projector defines a spatially variant volume-of-response through multiray casting relaying on the implemented IDRF.
Results obtained with IRIS projector show a SNR increase compared to standard projectors such as Siddon. Line profiles through patient images show the recovery of similar activities between the mMR built-in PET and MindView system images reconstructed using IRIS projector. However, peak to valley ratios are lower for MindView images. A reason for this is that there is no scatter correction implemented currently for the Mindview system as it is the case of the mMR. This influences the peak to valley ratio between the white and gray matter.
The IRIS projector and the regularized OSL optimizer, contribute to a SNR increase in images reconstructed with the MindView system. This SNR increase will contribute to the image quality of dynamical brain studies, or in general, studies in which the preservation of the SNR is a key feature. All the software produced have been included into the CASToR platform, an open-source multi-platform project for 4D emission (PET and SPECT) and transmission (CT) tomographic reconstruction, and will be released as part of a future version. The platform is under active development, including relevant tools for dynamical brain studies such as temporal regularization or an image-base linear deformation model to reduce involuntary patient motion artifacts (figure 14).
PET Reconstruction software for MindView has been optimized for GPU-based reconstruction. Anatomical information from MR can be incorporated to get a better attenuation correction. With appropriate MR image segmentation, anatomical priors could be incorporated in the PET reconstruction matrix.
Reconstruction process. An initial reconstruction needs to be performed from initial PET data. This reconstruction will generate a PET image with no attenuation correction. Once this PET image is available, it can be used as the reference image in the co-registration process (described below). Based on the co-registered anatomical MR image, an attenuation map is generated and a second reconstruction is performed, which generates the final reconstructed attenuation-corrected PET image.
Image coregistration. The PET-MR co-registration process:
1. The image orientations in the PET images describe the relative position of patients with respect to the scanner. Due to the inconsistencies of scanning protocols, the images were re-oriented to standard orientation with the FSL package before any further processing
2. Now, an image segmentation that extracts the brain from both images is performed, and those extracted images are fed into the co-registration algorithm that calculates the linear (FSL-FLIRT) and non-linear (FSL-FNIRT) transformations to be applied. These transformations are saved at the transformation matrix.
In the particular case of Mindview a rigid registration with a correlation ratio cost function is enough, due to the both images PET and MRI come from the same scanner, so the orientation will barely change. Six degrees of freedom to register both images is enough for the current FOV size.
3. Finally, the calculated transformations saved in the matrix are applied to original MR image, and image is cropped and/or filled to have the same dimensions as the PET image.
MINDView images, patient images. The following images show a reconstructed MindView PET Image with attenuation, co-registered with a MR image that was acquired simultaneously on a Siemens mMR system. The upper image has Partial Volume correction applied to reconstructed image, as shown in figure 16.
Task 7: Production of final prototypes. This report specially concerns about the construction, installation at TUM and validation of the system, both with phantoms and patients.
The distance from opposite crystal-to-crystal is about 330 mm. This, together with the allowed map of coincidences, defines a system FOV of 240 mm in diameter (transaxial) and 154 mm in the axial direction. The total volume of LYSO material is 3000 cm3, and the number of digitized signals as high as 1440 (total insert PET weight about 45 kg). These compare to about 9175 cm3 of LYSO and 4032 channels, for the PET within the Siemens mMR. The total scanner diameter including radio-frequency shielding is 42 cm, with an axial length of roughly 80 cm.
The detector blocks are temperature stabilized using temperature controlled air flow, resulting after controlling the input pressure to five vortex tubes. The temperature sensors at the read-out electronics (near the SiPM arrays) are read and a PID controller manages the output air temperature. All PET measurements inside the MR were carried out at a stable average temperature (60 blocks) of 27.5 ºC. As it can be seen in next figure, the temperature spread during 7 hours while continuously running MR sequences, was below 0.05 ºC (sigma). A detail of one set of various sequences is also shown in figure 18. The temperature regulation is visible, however even when running aggressive-pulsing sequences such as ultra-short time echo (UTE) or echo-planar imaging (EPI), no effect on the average temperature was observed. This target temperature is a compromise between PET system performance and demanding air flow.
The PET performance was evaluated using a 1 mm Na-22 source moved across the axial FOV, in steps of 0.5 mm, in order to measure the system sensitivity. The figure below shows the measured sensitivity curve for an energy range of 350-650 keV. We observed almost 7% sensitivity at the CFOV, agreeing well with previous measurements with one ring and also with expected values. Noise Equivalent Count curves have not yet been analyzed but initial results show system capabilities to manage above 150 MBq within the FOV, without apparent image deterioration.
The 0.25 mm in diameter Na-22 source was moved along the radial axis, at three axial positions namely the center of the FOV, 1/4 and 3/8 of this axis. The data were reconstructed using LMOS with 0.7 mm voxels (cubic dimensions) and 1.4 mm virtual size pixels. One iteration and 15 subsets were selected. As depicted in the figure above, the FWHM of the radial, transversal and axial components at the CFOV are about 1.7 mm, degrading to 3 mm in the case of the radial and axial components, and to 2.2 mm in the case of the transversal at 100 mm off-radial distance. No significant deterioration is observed at other axial positions. The results show good uniformity across the FOV, especially for the 120 mm in diameter center, as it was expected, due to the accurate DOI correction. We have calculated the volumetric FWHM for the case of sources placed at the axial center and compared with data obtained without DOI correction. We observed a high degradation towards the FOV edges (figure 19).
In addition to iterative algorithms, and according to the NEMA protocol, sources at center of the axial system center, and at two radial offsets, have also been reconstructed using single slice rebining and filtered backprojection with a butterworth filter. The image matrix was 240 mm × 240 mm × 154 mm, and different cubic voxel sizes tested.

PET performance vs. MR sequences
The effect of different MR sequences was evaluated with different FDG filled phantoms. We have measured the count rates for all phantoms as a function of the different MR sequences (and time). Figure 20, at the top, shows the number of acquired counts (prompt) as a function of time and for different MR sequences, for three phantoms. The data points (black squares) have been fitted to an exponential decay curve fixing the 18F half-life (red line in the top panels). Initial activities were 12 MBq, 40 MBq, and 32 MBq, for the named mini-Derenzo, Uniform and NEMA phantoms, respectively. We have estimated the percentage of randoms events, and they are in the range of 2-4%. The panels below depict the residue (open circles) in absolute counts (left axis) and in percentage (right axis). No significant count losses are observed, all differences are below 1%.
In addition to count rate studies, we have also studied the energy profiles for some detectors for all sequences. We did not observe significant differences among all the measurements. As an example, the figure above (right) shows the energy profile for one detector when running experiments with the NEMA phantom. A slightly 2-3% worsening energy resolution, but not count rate losses, was observed during UTE.
We reconstructed acquisitions of the small animal NEMA phantom using LMOS with 1 mm voxel size and 2 mm virtual pixel. We analysed a profile across the smallest rods for all cases, as plotted in the bottom-left panel. We have determined the Gaussian widths and plotted them against the sequence type without observing any significant deterioration for any particular sequence. In a further step, LMOS reconstructions of acquisitions with the PET insert but using cubic voxels of 0.7 mm and virtual pixels of 1.4 mm were also possible, showing an improved image quality (FWHM) of about 10%.
(Figure 22) The image quality phantom for small animals was also used to provide information on the SNR and recovery coefficients, as a function of the MR sequence. The data were reconstructed using OSEM (CASTOR) since it allowed varying the number of iterations and subsets finding the optimum for these tests. Indeed, best performance has been achieved for 1 or 2 iterations and 10 subsets. Moreover, the 2 mm rod was unfortunately in the exact CFOV and, using CASTOR it was possible to minimize and underestimate the mean values by computing the sensitivity matrix directly from the normalization data. The SNR was calculated as the ratio of the difference of the mean values obtained for the rods and background, to the standard deviation of the background. The VOI of the rods had diameters matching the true rod diameters and a height of 12 mm. The background was taken from a cylindrical VOI of 4 mm in diameter and 12 mm height in the center volume of the 5 rods. The RC were calculated as the maximum of the VOI over the mean value of the uniform area (25 mm diameter times 10 mm height). In general, both the SNR and the RC show no dependencies with the MR sequence. SNR improves with the rod diameter, as expected. Also the recovery coefficients exhibit this behaviour.
(Figure 23) Additional tests to explore the effect of the MRI sequences on the PET insert performance were carried out using the mini-Derenzo phantom. The phantom was placed parallel to the patient bed. The images in figure 23, that are in absence of an MR sequence, show the reconstructed images for data sequentially obtained with mMR PET and PET insert, left and right, respectively. The mMR PET makes use of an OSEM algorithm with voxel sizes of 1 mm × 1 mm × 2 mm (axial). A few rods of 2.5 mm were not well filled. The images are displayed without applying any filter. The PET insert image is obtained using MLEM with 63 iterations, whereas the mMR image is again an OSEM with 3 iterations and 21 subsets and, therefore, equivalent effective iterations. The PET insert, in contrast to the mMR image, shows the capability to resolve the 2.5 mm rods. We have plotted below the profiles of a row of 2.5 mm capillaries for different MR sequences, without observing any degradation.
Hoffman phantom performance
The system was also test before patients with human brain phantoms, the so-called Hoffman phantom. In figure 24 we show a comparative study of the data obtained with the whole-body PET system from the mMR and the MINDView PET insert (two reconstruction types: OCV and UB). The comparative results was successful.


Highlighted significant results: This summary shows the well working performance of the system both outside and inside the MR, and with multiple sequences typically used for brain imaging, some of them being very challenging from the point of view of possible interferences.
The system has shown to work well and provide very high resolution images. The images are significantly improved with respect to the whole-body PET in terms of spatial resolution. When running phantom and patients, the images show very good performance as well. The system works easily inside a 3T MRI as it was envisaged in the project proposal.
3.5 WP6 Radiopharmaceutical development
Progress towards objectives WP6:
One of the main objectives of this work-package was to develop and use an innovative Radiopharmaceutical Laboratory Platform (BLox), which is a Mini-clean room equipped with microfluidics for CGMP production of PET tracers.
Bencar has created and built a platform for the production of 11C-carbon monoxide in collaboration which has been placed and tested at Karolinska Hospital integrated with microfluidics technology under development. Part of the work is described in a publication from 2015. This platform use liquid nitrogen as cooling medium, something we are working to replace.
To demonstrate the value of 11C-carbon monoxide as the building block for more complex labelling syntheses we have also demonstrated how useful it is to produce some PET ligands. This has been done in collaboration within Mindview. We have shown i.e the production of 11C-raclopride partly using 11C-carbon monoxide (Method 1) and 11C-methyl iodide with the existing method (Method 2). It emphasize the value of 11C-carbon monoxide is an interesting option (ref 2). We have even developed a method of use alkyl iodides and (11C) CO in a nickel-mediated cross-coupling reaction and thereby demonstrated a successful use of different electro files containing beta-hydrogen atoms in different 11C-carbonylations (ref 3).
To be able to produce the building blocks 11C carbon monoxide and methyl iodide in a simple and rational way is therefore quite useful if it can be integrated into new platforms specially applied with microfluidics technology.

Now Bencar focuses on developing a new technology for the new 11C-methyl iodide system in a mini cleanroom concept. In Figure 25 schematically described the system currently under development for utilizing a cold plasma technology, based on technology that we have evaluated in other systems.
Producing 11C-methyl iodide from 11C-carbon dioxide needs that the 11C-methane can be produced using proven methods then has to be integrated in our system. We need to be able to control the temperature in the range of 400oC and also have access to iodine in the plasma we intend to generate.
Task 2. Validation of best radiopharmaceutical combination for schizophrenia animal models
Work in collaboration with Uppsala University. Transgenic mice were developed at Uppsala University and transported to Karolinska Institutet for imaging using micro-PET. For details, see report from Klas Kullengren.
Task 3 Radiopharmaceutical imaging the dopamine system
The selective dopamine D2 receptor antagonist raclopride is usually labeled with 11C using [11C]methyl iodide for quantification of dopamine D2 receptors in the human brain. The aim of this work was to label raclopride instead at the carbonyl position using [11C]carbon monoxide radiochemistry. Methods: Palladium-mediated carbonylation using [11C]carbon monoxide, 4,6-dichloro-2-iodo-3-methoxyphenol and (S)-(-)-2-aminomethyl-1-ethylpyrrolidine was applied in the synthesis of ([11C]carbonyl)raclopride. The reaction was performed at atmospheric pressure using xantphos as supporting phosphine ligand and palladium (π-cinnamyl) chloride dimer as the palladium source.
The selective dopamine D2 receptor antagonist raclopride is usually labeled with 11C using [11C]methyl iodide for quantification of dopamine D2 receptors in the human brain. The aim of this work was to label raclopride instead at the carbonyl position using [11C]carbon monoxide radiochemistry and to compare ([11C]carbonyl)raclopride with ([11C]methyl)raclopride in non-human primate (NHP) using PET with regard to quantitative outcome measurement, radiometabolism and protein binding. Methods: Palladium-mediated carbonylation using [11C]carbon monoxide, 4,6-dichloro-2-iodo-3-methoxyphenol and (S)-(-)-2-aminomethyl-1-ethylpyrrolidine was applied in the synthesis of ([11C]carbonyl)raclopride. The reaction was performed at atmospheric pressure using xantphos as supporting phosphine ligand and palladium (π-cinnamyl) chloride dimer as the palladium source. ([11C]Methyl)raclopride was prepared by a previously published method.
In an applied PET study, two female cynomolgus monkeys were examined under gas anesthesia of sevoflurane. A dynamic PET measurement was performed for 63min with the HRRT PET camera after intravenous injection of ([11C]carbonyl)raclopride and ([11C]methyl)raclopride, respectively, during the same day. Binding potential (BPND) of the putamen and caudate was calculated with SRTM using the cerebellum as a reference region. Conclusion: Raclopride was successfully labeled at the carbonyl position using a palladium-mediated [11C]carbonylation reactionThe monkey PET study with ([11C]carbonyl)raclopride showed similar results as for ([11C]methyl)raclopride.
Task 5. Radiopharmaceuticals imaging the serotonergic system
Following the work within Task 3 we have also extended a similar paradigm into other radioligands with affinity to the serotonergic system such as [11C]WAY100.635. Within this work a novel method have been developed using a metal mediated reaction for [11C]carbon monoxide even in cases where one of the reactants is containing beta hydrogens. Shortly, transition metal-mediated cross-coupling reactions of non-activated alkyl electrophiles suffer from competing beta-hydride elimination and therefore have limited applications in synthetic organic chemistry. Hereby the first successful use of nickelmediated carbonylative cross-coupling of non-activated alkyl iodides using [11C] carbon monoxide has been developed.
Tests and validation of mouse model. Within the MindView project, we have carried out development of in-vivo models for schizophrenia, or more specifically, mouse models with deficient responses in the pre-pulse inhibition test. The idea is to investigate such models with PET to evaluate the homeostasis of endogenous neurotransmitters like dopamine and serotonin. The monoamines and acetylcholine are well-known neurotransmitters with profound effects on multiple neurons, and which are essential for normal behavior and mental health. We recently reported that the orphan transporter SLC10A4, located in aminergic presynaptic vesicles, exerts a regulatory role in monoamine homeostasis. Mice lacking SLC10A4 were hypersensitive to psychostimulants including amphetamine, while indifferent to direct inhibitors of presynaptic and vesicular dopamine transporters.
Turning to in-vivo recordings, we found that endogenous and exogenous dopamine was less efficiently cleared from the extra synaptic space in mice that lacked SLC10A4. Furthermore, monoamine levels were drastically reduced in the striatum of SLC10A4 knock out mice. Impaired monoamine homeostasis and neurotransmitter clearance could have consequences for behavior, similar to provocation with psychoactive drugs. These findings suggest a role for SLC10A4 in monoaminergic signaling and reveal a novel mechanism for neuromodulation. Further, we showed that SLC104 is expressed in the human brain and thus represents a novel target for treatment of mental disorders. Mental and neurological diseases are often associated with an imbalance between monoaminergic and cholinergic transmitters, and drugs available for treatment of mental disorders, which meet a high demand, are predominantly targeting the monoaminergic systems.
We have established the potential use of Slc10A4 mice as a small animal schizophrenia model based on their response to the PPI test (Figure 1). We performed a PPI evaluation of Slc10a4-knock out mice, Slc10a4 over expressing mice and wild type mice. Mice were analyzed in different pre-pulse inhibition regimens; high amplitude (20db above back ground noise) and short IEI (4 mSec) vs long IEI (20 mSec). High amp and short IEI gives a significant difference for both gene variants, with wild type control animals displaying a PPI ability equivalent with results in the literature. However, no significant difference was when comparing for pre-pulse inhibition with high amp and long IEI. Slc10a4-KO animals appear to have a higher ability to filter, process and link the pre pulse and startle pulse compared to control animals. In contrary, mice overexpressing Slc10A4 show a decreased ability to link the pre pulse to the startle pulse indicating an inability to habituate to the pre-pulse and startle pulse. Data is shown as mean±SEM, nWT = 17, nKO = 21, nNSE = 13 animals. Statistic analyses by one-way ANOVA followed by Tukey´s MCT (GraphPad Prism 5).
Animals are breed from heterozygote parents. Animals were analyzed at the age of 10-15 wks. Habituation to the PPI apparatus was done over two weeks and directly prior to the performance of the experiment. Every animal was evaluated twice per day for 3 days before data collection. The PPI program consisted habituation for startle pulse (16 times), followed by 8 times randomized pre pulse-short IEI, high high-pre pulse-long IEI, high pre pulse-alone, startle alone, low pre pulse-short IEI, low pre pulse-long IEI and low pre-pulse alone, figure 26.
Tests and validation of radioligands. From our test, we conclude that SLC10A4 ko mice show defective prepulse inhibition. In a collaboration between Mindview partners (UU, Bencar and KI) we next evaluated uptake and distribution of radiopharmaceuticals in VAAT-KO mice using established radioligands in PET.
We used the dopamine D2 radiotracer, 11C-raclopride, which is a a reliable and robust radioligand with well characterized properties in small animal models. PET studies have shown that changes in the binding of the 11C-raclopride can reflect increases in synaptic dopamine secondary to behaviors including reward, cognitive learning (Badgaiyan et al., 2007), the smell and presentation of reinforcing foods (Volkow et al., 2003). In addition, 11C-raclopride has since long been used in several studies to evaluate schizophrenic patients (Farde et al, 1990, for review see Howes 2015).
In addition to dopamine, serotonin has also been associated to schizophrenia (Bleich et al., 1988). Thus we also used a serotonin 2A (5-HT2A) receptor agonist PET radioligand called 11C-Cimbi.(Finnema et al) Agonist radioligands may target specifically the G protein-coupled state of the receptors and thereby provide a more meaningful assessment of available receptors than antagonist radioligands.
In the current study we characterized 11C-raclopride and 11C-Cimbi receptor binding in the mouse brain. On the same experimental day, two PET measurements were conducted in each animal. We have imaged in total 4 wild type mice and 5 Slc10A4-/- mice with 11C-raclopride and in total 3 wild type mice and 5 Slc10A4-/- mice were imaged with 11C-Cimbi.
A 63-minute (11C-raclopride) or 93-minute (11C-Cimbi) PET measurement was initiated immediately upon intravenous injection of the radioligand. All PET measurements were perfomred using the nanoScan PET/MRI® and the nanoScan PET/CT® (Mediso ltd, Hungary) system. The two systems have identical PET performance and were calibrated to provide consistent results. Genotypes were divided equally betweent the two systems. The reconstructed dynamic PET images were co-registered to an in-built mouse MRI template available in PMOD, which also incorporates volumes of interest (VOI’s) sets (PMOD Technologies Ltd., Zurich, Switzerland). With the help of these VOI sets, decay corrected time activity curves (TAC) were generated The regional brain uptake values were expressed as percent standard uptake value (%SUV), which normalizes for injected radioactivity and body weight. The binding potential (BPND) of 11C-raclopride was calculated in PMOD with the Simplified Reference Tissue Model (SRTM) using cerebellum as a reference region.
There was no difference in average %SUV of 11C-raclopride in the cerebellum comparing wild type mice with Slc10A4-/- mice. There was increased binding of raclopride in Slc10A4-/- mice compared with wild type with a significantly increased BPND in Slc10A4-/- mice compared with wild type controls measured in the striatum (see figure 27).
The preliminary results show that Slc10a4 KO mice show a higher binding potential of 11C-raclopride compared to WT animals. There is no difference in 11C-Cimbi uptake between Slc10a4 KO and WT mice. We now continue to further examine possilbe differences in the dopmaminergic system in Slc10a4 KO mice compared to WT mice. We will investigate the responsiveness of the dopamine system by adminster amphetamine before [11C]raclopride imaging. The hypothesis is that the reduction in BPND of 11C-raclopride after amphetamine will be different between WT and Slc10a4 KO mice.
3.6 WP7 Clinical and Preclinical Validation and Trials
The primary objective of this work-package was to test the integrated PET – RF coil system into preliminary applications in clinical investigations. We plan to study the impact of the new tools in two major diseases: schizophrenia and severe depression. Protocol and data for clinical measurements with PET and MR have been developed.
The PET insert and RF coil were successfully delivered to TUM in January 2018. As it will be described below, it was validated with multiple phantoms after the installation.
Partners have focus onto neurodegenerative disorders such as Alzheimer’s disease (AD). Age-associated neurodegenerative disorders are accompanied by brain atrophy and vascular lesions, which substantially complicates accurate PET measurements and thus challenge the MindView system under the construction. Dr. Igor Yakushev, a co-investigator of the MINDVIEW project, spent several months at the Karolinska Institute at Stockholm to generate cooperative projects between the two clinical sites involved in the MINDVIEW project. As primary study at the TUM, we have established a tau PET tracer automation, CMC and imaged tau proteins, a pathological feature of AD, in transgenic mice and patients with mild cognitive impairment (MCI) and dementia due to AD.
Preparatory activities - radiochemistry
Following the successful development of novel radiolabeling of radiopharmaceuticals (see tasks 6.3 and 6.5) the Head of Production at the KI PET Center optimized the radiochemistry procedures for application in tasks 7.5 and 7.6.
This core of the radiochemistry work was the development and testing of platforms at KI for labeling of radiotracers with 11CO and 11CH3I. Thereby continuing the work of Rahman et al (1). The work included supporting and monitoring 4. BENCAR in technology transferred to the labs at KI and represents a continuation of previous work at 4. BENCAR, developing the mini-clean room concept BLox a new concept to produce PET tracers using microfluidic technology and to reach cGMP-requirements.
Preparatory activities - clinical
While awaiting the prototype, the clinical methodology and recruitement of first-episode neuroleptic naïve patients was fine-tuned and validated in a clinical PET-study on TSPO, a neuroinflammatory marker primarily representing microglia (2). As outlined in the original proposal an examination by a research physician included psychiatric interview (SCID), medical and psychiatric history, physical examination, blood and urine sampling and MRI examination of the brain. Patients age 18-45 years, satisfied criteria for Schizophrenia according to DSM-5. Psychopathology was rated using PANSS. Patients previously treated with antipsychotic drugs, with drug abuse or significant neurological or other somatic disorder were excluded.
We examined 16 drug-native, first episode psychosis patient and 16 healthy controls using PET and the TSPO radioligand 11C-PBR28 [Collste et al). There was a significant reduction of 11C-PBR28 binding in patients compared to healthy controls in gray matter (GM) as well as in secondary regions. In this limited sample, there was no correlation between GM binding and clinical or cognitive measures after correction for multiple comparisons. The observed decrease in TSPO binding suggests reduced number or altered function of immune cells in brain in early stage schizophrenia.
Tests with the PET insert.
The Mindview system is integrated by a PET insert enclosing a two-channel radiofrequency head coil. The integrated system lays on the bed of the Siemens Biograph mMR at the height of where the patient head is usually located. The Mindview system is connected to a cabinet at the back of the room, containing the acquisition electronics, through a shielded cable. The cabinet is located beyond the 5 GAUSS line not to disturb the electromagnetic field of the mMR scanner. The cabinet is connected to an isolated power socket in the Siemens mMR room, connecting the ground of the Faraday cage, the Siemens mMR scanner and the Mindview system, thus avoiding differences of potential between the systems.
The PET system consists of 20 PET detectors with temperature sensors and a cooling system, based on circulating air extracted from the Siemens mMR room that keeps the detectors at a stable temperature of 27.5°. Scans as long as 7 hours have been acquired while the temperature was monitored, not observing any visible variation in the temperature of the detectors. The performance of the PET insert has been investigated while running typical MRI sequences used in clinical routine, including some of the most aggressive sequences (strong gradients) like EPI and dual-UTE, not observing any variations in the PET performance.
The PET insert is covered in PMMA and is in direct contact with the RF coil on its inner surface. The RF coil has an inner diameter of 250 mm and outer diameter of 330 mm, matching the inner diameter of the PET insert. The material of the cover is fiber glass, providing mechanical strength and electrical safety. The RF coil contains watchdogs that limit the maximum power of the amplifiers, avoiding possible heatup in the electronics. Figure 1 shows the Mindview PET insert with the cabinet and the radiofrequency coil that used for reception.
Electrical safety tests following the IEC 60601-1, developed for medical electrical equipment, have been performed in the PET insert and the RF coil, passing both systems all the tests.
Imaging procedures
According to an established clinical protocol, subjects are to be injected with a 5 mCi 18F-FDG PET at rest, after fasting for at least 6 h before scanning. Image acquisition is started 30 min post-injection. To avoid a negative bias towards the MindVew system, patients are to be examined alternately: a half of them undergo imaging in the MindView, another half in the hybrid PET/MR scanner first. This first scan immediately follows the second scan, in the hybrid PET/MR or MindView, respectively. The routine clinical scan in the PET/MR systems takes 35 min, in the MindView 15 min. Study subjects are not exposed to any additional radiation due to the imaging with the MindView system.
The MindView system is described above. Technical specifications of the PET/MR scanner Siemens Biograph mMR (referred to as mMR thereafter) are summarized elsewhere (Delso et al., 2011). The construction allows a simultaneous acquisition of PET and MR signal from the same body volume, thus enabling to simplify logistics and save time. Unlike CT (in a PET/CT system), MRI does not use ionizing radiation. The camera consists of a high-end 3T MRI scanner (technically corresponding to the Siemens Verio system) that harbors a fully functional state-of-the-art avalanche photodiode-based PET system within its gantry. The PET scanner has a spatial resolution of 4.3 at 1 cm and of 5.0 mm at 10 cm from the transverse FOV and a sensitivity of 15.0 kcps/MBq at the center of the FOV.
PET insert evaluation
TUM has evaluated the image quality provided with the MindView system and compare it with a clinically established whole body hybrid PET/MR system Siemens Biograph mMR (referred to as mMR system hereafter). It has been successfully functioning in our Department since 2010. Our work can be arbitarily subdivided into three consecutive parts: phantom measurements, first human measurement, and a prospective study in a patient population.

First human subject
To obtain a preliminary estimatation of the image quality in human, we examined a 72 years old patient with a suspected Alzheimer’s disease. The subject was studied twice, using the mMR and, right afterwards, the MindView system. To this end, the PET insert was temporarily integrated in the mMR. During the second scan the PET part of the mMR was disabled. The subject was injected with 185 MBq of F18-FDG. Image acquisition was perfomed at 30 to 45 min p.i. and at x to y min p.i. in the mMR and MindView system, respectively. Figure 28 shows reconstructed PET images with a typical 18F-FDG distribution.
An effect of the partial volume correction based on simultaneously acquired T1 MRI images was estimated. Figure 29 shows a fusion between (MindView) PET and T1 images after attenuation correction. Motion correction and isochronous acquisition is yet to be implemented. With the appropriate MR information, motion correction techniques would improve PET image quality.
Prospective study in a patient population
As soon as the preliminary data as reported above were available, we initiated a clinical study in patients with a suspected neurodegenerative disoder. Due to the lack of time we focused on this clinical population insted of patients with schizophrenia, who are much more difficult to recruite. Still we are convinced that the patients with neurodegenerative disoders represent an meaningul target population for evaluation of the MindView system. Specifically, our analyses will focus on structures of the basal forebrain that include small to very small regions such as nucleus accumbens, nucleus basalis, substantia innominata, and the medial septal nucleus. Degeneration of the basal forebrain cholinergic nuclei is known to be associated with cognitive decline in Alzheimer’s disease. So far, due to its small size these structures could not be reliably measured with PET. The study protocol (in English) as well as patient information (in German) are included in the corresponding deliverable. The study was preliminary approved by the local ethics committee. A revised version of the study proposal is currently under review by the ethics committee. The final approval should be available next week. Afterwards, we will start scanning.
Potential Impact:
Regarding strategic impacts, MINDView will be a real breakthrough in terms of new tools for imaging that will allow the definition of parameters allowing patient stratification in schizophrenia and depression diseases. This should also lead in the future to development of the next generation of drugs for treating those disorders. Furthermore, the new tools developed will allow for an extension of techniques that are currently only available to very few hospitals.
The “European Pact for Mental Health and Well Being” in its second article, states: “We agree that ...- There is a need to improve the knowledge base on mental health: by collecting data on the state of mental health in the population and by commissioning research into the epidemiology, causes, determinants and implications of mental health and ill-health.”
Although still an SME, OCV is the leading company in the World in the emerging intra-operative molecular imaging market, but has also developed a PET/CT/SPECT system for small animals, currently distributed by Bruker (German Multinational World leader in preclinical imaging). NOR is a leader in the development of RF coils for dedicated organs. Both companies will benefit directly from this project. The project represents an opportunity for BEN and SSL to enter the promising medical market and will thus represent a direct opportunity of development for them.
Concerning clinical impacts, the significance of this project is substantial. Neuropsychiatric disorders affect a large part of the population, and cause a great deal of suffering both to the patients and to those close to them. The cost to society is substantial. A three-year study covering 30 European countries - the 27 European Union member states plus Switzerland, Iceland and Norway - and a population of 514 million people, looking at about 100 illnesses covering all major brain disorders (including depression and schizophrenia, concluded that Europeans are plagued by mental and neurological illnesses, with almost 165 million people or 38 percent of the population suffering each year from a brain disorder.
Mental disorders are on the rise in the EU. Depression is already the most prevalent health problem in many EU-Member States. More specifically, the prevalence of schizophrenia and depression in year 2005 were 0,8% (3,7 million) and 6,9% (18,4 million), whereas in year 2011 were 1,2% (5 million) and 6,9% (30,3 million), respectively.
With only about a third of cases receiving the therapy or medication needed, mental illnesses cause a huge economic and social burden -- measured in the hundreds of billions of euros -- as sufferers become too unwell to work and personal relationships break down. Those few receiving treatment do so with considerable delays of an average of several years and rarely with the appropriate, state-of-the-art therapies.
Mental disorders have become Europe's largest health challenge of the 21st century. Mental illnesses are a major cause of death, disability, and economic burden worldwide and the World Health Organization predicts that by 2020, depression will be the second leading contributor to the global burden of disease across all ages. Suicide remains a major cause of death. In the EU, there are about 58,000 suicides per year of which ¾ are committed by men. Eight Member States are amongst the 15 countries with the highest male suicide rates in the world. Mental disorders and suicide cause immense suffering for individuals, families and communities, and mental disorders are major cause of disability. They put pressure on health, educational, economic, labor market and social welfare systems across the EU.
MINDView will allow a better understanding of neurobiology, including the role of specific neurotransmitters within the brain circuitry and their interrelations. The abovedescribed technology will enable simultaneous dynamic PET and MR acquisition following specific activation paradigms using specific tracer combinations. We can envisage several completely new and never before performed applications, which represent generic areas that can be investigated if high resolution PET and MR data can be acquired simultaneously.
The new technology to be developed during the project will open new venues. By the novel concept of simultaneous, high resolution PET-MR imaging, we are able to integrate both anatomical alterations (MR), connectivity (Diffusion Tensor Imaging), neuronal signalling (fMRI), neurochemistry (receptor ligand PET and Magnetic resonance spectroscopy), and neurotransmitter release (dynamical PET imaging in perturbation paradigms), in the same patient. Collectively, the combination between PET and fMRI with an increased resolution will profoundly change our view on how brain circuitry underlies behavior.
The long-term vision is that this PET/MR will have a major impact also on the clinical routine in mental health problems. The gains and advantages of combining the noninvasive imaging modalities PET and MR to allow for the simultaneous measurement of both imaging signals, is expected to have significant impact on prediction, diagnosis, monitoring and prognosis of schizophrenia disease.
Regarding economic impact, the MINDView project is triggering an expansion of the needed PET/MRI technology by making it useful for the diagnosis, monitoring and management of mental disorders and by making it accessible to most hospitals. Current PET/MRI technology is only affordable to very few hospitals in the World (about 50 units). MINDView can extend the PET/MR technology into practical clinical use by making it accessible to most hospitals via upgrading the existing installed base of whole body MRI systems.
The MINDView Project has involved four SMEs working in different domains such as Molecular Imaging, including imaging equipment and radiochemistry modules, Radiofrecuency Coils for organ dedicated imaging in MRI, and Silicon photosensors. OCV is currently entering the organ dedicated imaging market through a PET mammography system that is currently being validated clinically. NOR is a leader in the development of RF coils for dedicated organs. Therefore, both companies will join efforts and develop a dedicated system for the brain examination that includes both imaging modalities (PET & MRI) together. The MINDView system has an enormous commercial potential since the installed based of MRI systems is around 36.000. The current strategy can also be implemented for the examination of other organ such as breast (for instance the recently granted EU HYPMED Project), prostate or heart. Finally, for SensL will be a unique opportunity to enter the medical market segment, which has high revenue potential, and to take lead positions through the combination of their technologies. Indeed, SensL has been acquired recently by the multinational company ON SEMI. Furthermore, the small animal PET business of ONCOVISION, a Spanish company that develops molecular imaging systems, has been acquired by BRUKER, a German multinational company.
Beyond the direct recruitment of more than 30 researchers in the course of this project and the economical impact of the 4 MINDView SMEs partners, as well as their suppliers, this project has also indirectly contributed to employment through the development of high level technology. In fact, OCV, NOR, BEN and SensL already employed around 100 highly qualified people and are committed to the expansion of their businesses through large R&D investments.
If we see the European efforts, there has not been research teams in Europe, capable of conducting a research project including such a wide range of techniques, knowledge and facilities. Furthermore, such an undertaking would be impossible in a single European country and with only National funding, since it requires the best European experts to accomplish the objectives of the project within the execution timeframe stipulated by the FP7 Health call. This is also the only way to produce an impact in the European health care system for the benefits of the schizophrenia and depression patients. MINDView capitalises on existing scientific and technical expertise, skills and initiatives, through past and current European-funded projects or international collaborations.
The construction and first validation of the MINDView system can be considered as a good example of translational research in medecine – which is the main aim of the Health thematic in FP7 – with transfer of basics research (chemistry, physics, optics, electronics) to clinics with the intermediary steps to integrate the technologies and components, to calibrate the probe (through preclinical investigations on pigs) and a first clinical testing of the prototype (through pilot clinical test on patients with pancreatic tumors). All these steps, as part of WP6, have been achieved thanks to a perfect cooperation between end-users (defined by clinical doctors able to define patients’ requirements) and engineering developers.

4.1 Dissemination
During the this project, the Consortium has created a first version of the Exploitation Plan (deliverable D8.2). This deliverable describes the future planned actions to be done by the Consortium related to the exploitation of the project results.
Other important milestones achieved during this period are:
• Update of the website of the project
• Publication of scientific papers
• Participation in different congresses to disseminate information about the project results
We have worked on:
- A website to be used to facilitate the internal exchanges and will contain a free access section for information of the public/industry/medical and scientific community.
- A kick-off meeting, where major medical imaging and biomarker industries were invited, in addition to all parthers.
- Regular meetings have been carried out.
- Publication of the MINDView results achieved their rapid dissemination in high profile refereed journals. To this end, the results have been divulgated to the following, as appropriate:
o to other Partners in the network;
o to the Partners’ own administrations;
o to the European Commission;
o to the scientific and medical communities;
o to the medical Imaging and biomarker industry;
o to the concerned patients’ associations;
o by submission of collaborative research articles;
- by publication,
- Participation appropriate international conferences, workshops and in association with other European projects: conferences (Annual meeting of the SNM, Annual meeting of the EANM, MICCAI, IEEE NSS/MIC, SPIE Medical Imaging), forums and workshops, for specialists but also those open to the general public.
The partners identified key target groups for dissemination. It was ensured that the Dissemination and Awareness campaign of the project reached the five following groups:
1. The General Public through patients organizations;
2. The International Scientific community as for example the European Society of Molecular Imaging, European Society for Mental Health and Deafness, the European Network for Mental Health Promotion and Mental Disorder Prevention;
3. The Clinicians, through different clinical networks of the clinical partners of the project;
4. The Health and Education Professionals;
5. The industrial stakeholders working in the filed of medical imaging and biomarker development.

Moreover some members of the consortium are already involved in large scale dissemination actions. As an example: M. Schwaiger (P3) is member of several European medical societies and participates to a number of international events, where he tirelessly promotes the importance of translational research in medicine.
The MINDView consortium has carried out many dissemination activities, including the participation in conferences/exhibitions, publications and web/social media activities.
The whole consortium has made it possible to publish a high number of works, as a result of the projecrt efforts. There are about 25-30 peer-reviewed publications in high impact journals such as IEEE TNS, IEEE TRPMS or NIM-A, to name but a few. Many, about 20 conference proceedings have been published.
Regarding oral contributions, there has been a very high number of presentations given to a variety of different public, namely scientific or divulgative. All partners have carried out these tasks, alwasys publicitating the MINDView project.
List of Websites:
http://www.mindview.i3m.upv.es