Periodic Reporting for period 2 - TRABIT (Translational Brain Imaging Training Network)
Période du rapport: 2019-10-01 au 2022-09-30
In recent decades, medical imaging techniques such as computed tomography (CT), ultrasound (US), and especially magnetic resonance imaging (MRI) have gained a central role in the clinical management of disorders of the brain. Since brain imaging often visualizes disease effects with much greater sensitivity than clinical observation, it holds great promise to help diagnose patients at the earliest stages of their disease, when treatment is most effective; and personalize their treatment by evaluating their response to a specific intervention.
A fundamental bottleneck in translating the wealth of information contained in medical images into optimized patient care is the lack of patient-specific computational tools to help analyze and quantify the torrent of acquired imaging data. The last two decades of medical image computing research have matured to allow robust and automatic assessment of carefully homogenized scientific studies of mostly healthy brain scans. Yet analyzing the "wild" type of neuroimaging data arising in the standard clinical treatment of brain disorders has remained a hard and unsolved problem.
The overall goal of the TRABIT project was to enlarge the scope of quantitative brain image analysis from scientific group studies of the human brain into improved healthcare delivery in clinical applications, benefitting individual patients suffering from brain disease. Towards this end, TRABIT aimed to significantly advance the state of the art in the field of computational neuroimaging to achieve the following specific research objectives:
- RO1: To develop fundamentally new computational models to reliably analyze clinical brain scans in several major brain disease categories;
- RO2: To implement the developed methods in software tools that can be robustly applied across imaging sites;
- RO3: To test and validate the proposed methods and tools on large collections of (multi-site) clinical neuroimaging data;
- RO4: To share the accomplished methodological advances across the network and beyond;
- RO5: To facilitate true clinical adoption of the developed methods and tools by strongly engaging with the medical imaging industry.
In order to achieve these ambitious research objectives, TRABIT has trained a new generation of young European researchers capable of seamlessly combining a deep understanding of computational neuroimaging; the clinical needs and constraints arising in the treatment of brain disorders; and the commercialization processes needed to bring new research developments into actual clinical practice. In particular, we successfully trained 15 early-stage researchers, each with an individual research plan centered around one of four major brain disease categories - Multiple Sclerosis (WP1), Fetal Brain Disorders (WP2), Brain Tumors (WP3), and Stroke/Neurovascular Disease (WP4) - and each supervised by an academic, a clinical, and an industrial mentor.
A fundamental bottleneck in translating the wealth of information contained in medical images into optimized patient care is the lack of patient-specific computational tools to help analyze and quantify the torrent of acquired imaging data. The last two decades of medical image computing research have matured to allow robust and automatic assessment of carefully homogenized scientific studies of mostly healthy brain scans. Yet analyzing the "wild" type of neuroimaging data arising in the standard clinical treatment of brain disorders has remained a hard and unsolved problem.
The overall goal of the TRABIT project was to enlarge the scope of quantitative brain image analysis from scientific group studies of the human brain into improved healthcare delivery in clinical applications, benefitting individual patients suffering from brain disease. Towards this end, TRABIT aimed to significantly advance the state of the art in the field of computational neuroimaging to achieve the following specific research objectives:
- RO1: To develop fundamentally new computational models to reliably analyze clinical brain scans in several major brain disease categories;
- RO2: To implement the developed methods in software tools that can be robustly applied across imaging sites;
- RO3: To test and validate the proposed methods and tools on large collections of (multi-site) clinical neuroimaging data;
- RO4: To share the accomplished methodological advances across the network and beyond;
- RO5: To facilitate true clinical adoption of the developed methods and tools by strongly engaging with the medical imaging industry.
In order to achieve these ambitious research objectives, TRABIT has trained a new generation of young European researchers capable of seamlessly combining a deep understanding of computational neuroimaging; the clinical needs and constraints arising in the treatment of brain disorders; and the commercialization processes needed to bring new research developments into actual clinical practice. In particular, we successfully trained 15 early-stage researchers, each with an individual research plan centered around one of four major brain disease categories - Multiple Sclerosis (WP1), Fetal Brain Disorders (WP2), Brain Tumors (WP3), and Stroke/Neurovascular Disease (WP4) - and each supervised by an academic, a clinical, and an industrial mentor.
The work performed in the project can be summarized as follows:
1. After a broad international call attracting 401 applicants from 67 different countries, we carefully selected 15 excellent early-stage researchers to be trained by the consortium. Each of these early-stage researchers were subsequently enrolled in an individual PhD program co-supervised by an academic, a clinical, and an industrial mentor. The researchers followed an individual research plan centered around one of the four major brain disease categories we aimed to tackle.
2. We organized five week-long training events: three TRABIT training schools (attended by all TRABIT early-stage researchers as well as 28 external PhD students) and two internal TRABIT workshops (focused on the TRABIT students' research work). These events took place in Munich, Copenhagen, Lausanne and Eindhoven, and -- during the pandemic -- virtually, and offered 10 scientific courses as well as 7 transferable skills courses in total. Dedicated attention was given to hands-on training in a commonly adopted rapid-prototyping software environment to facilitate testing, demonstrating the developed tools in hospital environments, and disseminating the obtained models and methods.
3. As a final event, we also organized an international conference entitled "Translational Brain Image Analysis" (see https://trabit-network.github.io/conference). This online event, which focused on one specific scientific work package of TRABIT on each of the four subsequent days, was widely advertised internationally and well attended (171 participants excluding the speakers). It included internationally leading keynote speakers; round table discussions with panel members from academia, hospitals and the European medical imaging industry; 2-minute pitches by the TRABIT ESRs; and several networking sessions.
4. Over the project period, many TRABIT early-stage researchers developed open-source software; some contributed to patents and/or commercial software products & services; while yet others focused on enabling future research activities outside the action. Overall, around 90 full-length, peer-reviewed scientific articles were published by the TRABIT ESRs - around half of these appeared as journal papers, whereas the remaining ones were published in conference/workshop proceedings. One ESR filed a patent application.
5. We have maintained an active website (https://www.trabit.eu/) that features, among many other things, 1-1.5min video clips explaining each ESR's work and its impact in a form that is accessible to the general audience. In addition a number of blog entries and newsletters reporting on important project-related events, the website also has an automatic twitter feed displaying the messages posted on the (very active) TRABIT twitter account (https://twitter.com/TRABIT_ITN). Overall the TRABIT website has served as a central dissemination platform for the network, and contains all the information concerning the consortium, its ESRs, the project objectives, TRABIT training events, and our publications.
1. After a broad international call attracting 401 applicants from 67 different countries, we carefully selected 15 excellent early-stage researchers to be trained by the consortium. Each of these early-stage researchers were subsequently enrolled in an individual PhD program co-supervised by an academic, a clinical, and an industrial mentor. The researchers followed an individual research plan centered around one of the four major brain disease categories we aimed to tackle.
2. We organized five week-long training events: three TRABIT training schools (attended by all TRABIT early-stage researchers as well as 28 external PhD students) and two internal TRABIT workshops (focused on the TRABIT students' research work). These events took place in Munich, Copenhagen, Lausanne and Eindhoven, and -- during the pandemic -- virtually, and offered 10 scientific courses as well as 7 transferable skills courses in total. Dedicated attention was given to hands-on training in a commonly adopted rapid-prototyping software environment to facilitate testing, demonstrating the developed tools in hospital environments, and disseminating the obtained models and methods.
3. As a final event, we also organized an international conference entitled "Translational Brain Image Analysis" (see https://trabit-network.github.io/conference). This online event, which focused on one specific scientific work package of TRABIT on each of the four subsequent days, was widely advertised internationally and well attended (171 participants excluding the speakers). It included internationally leading keynote speakers; round table discussions with panel members from academia, hospitals and the European medical imaging industry; 2-minute pitches by the TRABIT ESRs; and several networking sessions.
4. Over the project period, many TRABIT early-stage researchers developed open-source software; some contributed to patents and/or commercial software products & services; while yet others focused on enabling future research activities outside the action. Overall, around 90 full-length, peer-reviewed scientific articles were published by the TRABIT ESRs - around half of these appeared as journal papers, whereas the remaining ones were published in conference/workshop proceedings. One ESR filed a patent application.
5. We have maintained an active website (https://www.trabit.eu/) that features, among many other things, 1-1.5min video clips explaining each ESR's work and its impact in a form that is accessible to the general audience. In addition a number of blog entries and newsletters reporting on important project-related events, the website also has an automatic twitter feed displaying the messages posted on the (very active) TRABIT twitter account (https://twitter.com/TRABIT_ITN). Overall the TRABIT website has served as a central dissemination platform for the network, and contains all the information concerning the consortium, its ESRs, the project objectives, TRABIT training events, and our publications.
The methods and tools developed by the TRABIT consortium have significantly advanced the state of the art in clinical neuroimaging, enabling improved detection and progress monitoring in a wide array of brain disorders, and optimizing healthcare delivery specifically tailored towards the needs of individual patients. Our research and training program has further generated 15 high-skilled scientists trained to take advantage of computational, clinical, and commercial opportunities in a key sector in Europe.