Deep neural networks (DNNs) have achieved outstanding performance and broad implementation in computer vision tasks such as classification, denoising, segmentation and image synthesis. However, DNN-based models and algorithms have seen limited adaptation and development within radiomics which aim to improve diagnosis or prognosis of cancer. Traditionally, medical practitioners have used expert-derived features such as intensity, shape, textual, and others. We hypothesize that, despite the potential of DNNs to improve oncological classification performances in radiomics, a lack of interpretability of such models prevents their broad utilization, performance, and generalizability. Therefore, the INFORM consortium proposes to investigate explainable artificial intelligence (XAI) with a dual aim of building high performance DNN-based classifiers and developing novel interpretability techniques for radiomics. First, in order to overcome the limited data typically available in radomic studies, we will investigate Monte Carlo methods and generative adversarial networks (GAN) for realistic simulation that can aid building and training DNN architectures. Second, we tackle the interpretability of DNN-based feature engineering and latent variable modeling with innovative developments of saliency maps and related visualization techniques. Both supervised and unsupervised learning will be used to generate features, which can be interpreted in terms of input pixels and expert-derived features. Third, we propose to build explainable AI models that incorporate both expert-derived and DNN-based features. By quantitatively understanding the interplay between expert-derived and DNN-based features, our models will be readily understood and translated into medical applications. Fourth, evaluation will be carried out by clinical collaborators with a focus on lung, cervical and rectal cancer. These proposed DNN models, specifically developed to reveal their innerworkings, will leverage the robustness and trustworthiness of expert-derived features that medical practitioners are familiar with, while providing quantitative and visual feedback. Overall, our methodological research will advance interpretability of feature engineering, generative models, and DNN classifiers with applications in radiomics and broad medical imaging. With this project we aim at maximizing the impact on the patient management of ML and DL techniques by developing novel methods to facilitate training of decision-aid systems for clinical treatment strategies optimization. The methodological approaches we propose in this specific area will play a major role in facilitating the acceptability of DL-based decision-aid systems relying on medical imaging for oncology. The proposed validated predictive models in various cancer types within the context of this project might subsequently be used to drive future prospective clinical studies in which patients could be offered alternative treatment strategies based on the results of these predictive models. Such a clinical and social potential is further enhanced by the public-private collaboration proposed in this project, where the developed methodologies will find their way in products. The multidisciplinarity of INFORM is key to meet the target challenges and achieve the proposed goals. All partners have their individual world-leading qualifications and additional scientific expertise providing all the prerequisites for the efficient implementation of INFORM’s approach. The successful implementation of this project will have a large and prolonged impact both in the Medical/Oncology and the Computing/ Artificial Intelligence field of predictive radiomics model, as well as the same methodology could be extended to other diagnostic and therapeutic medical applications.

The main objective of VIVOIMAG is to develop bone implants including a new contrast agent sensitive to enzymatic activity of metaloproteases, which will permit for the first time to follow the integration and cell differentiation activity in bone tissue bioreactors in vitro and in grafts in vivo using existing non invasive magnetic resonance imaging techniques. The proposal aims at integrating a magnetically functionalized extracellular matrix material into the bone scaffolds, seeding them with cells, implanting them in animal models and following the fate of the implants in vivo using MRI. The aim is to obtain similar results with the magnetically modified scaffolds as the ones obtained currently but having now endowed the grafts with enzymatic reporting activity that can be monitored noninvasively in the living animal. There are currently no methods to detect in situ and in vivo this enzymatic activity without previously sacrificing the transplanted animals, therefore the successful accomplishment of this project would have huge and prolonged impact in the medical field of tissue regeneration. The VIVOIMAG project brings together a multidisciplinary consortium of specialists in different areas of bone implant research, nanoparticles formulation and characterization, magnetic resonance and scintigraphic imaging, who will join forces in order to propose and assess a novel technique for the evaluation of the progress of bone implants in vivo, which can substitute existing invasive techniques based on biopsies. A well planned exchange program among academic and industrial partners will facilitate knowledge sharing, maximize collaborative work and finally achievement of project objectives. The consortium, being aware of the scientific and social importance of bone tissue engineering, has planned a series of dissemination and training activities, aiming at making project knowledge and outcomes available to the scientific community and society.

The incidence of Cardiovascular Disease (CD) claims worldwide 17.1 million lives a year, with an estimated 31% of all deaths globally and a EU cost of 139 billion euros. Up to 40% of all deaths occur among the elderly. In spite of all medical efforts, the 5-year mortality was reduced significantly less than that of malignant diseases. This highlights the urgent need to overcome the difficulties associated with present pharmacological therapies (i.e. drug instability, and unspecific targeting) by developing new ground-breaking therapeutic strategies that go far beyond any current regimens. New approaches for safe, efficient, and heart-specific delivery of therapeutics are strongly required. CUPIDO is envisioned to meet these critical needs by providing an unconventional and effective strategy based on nanoparticle-assisted delivery of clinically available and novel therapeutics to the diseased heart. In particular, CUPIDO will develop innovative bioinspired hybrid nanoparticles formulated as biologicals delivery, which are i) biocompatible and biodegradable, ii) designed for crossing biological barriers, and iii) guidable to the heart. A combination of multidisciplinary manufacturing and validation approaches will be employed, bringing the envisioned product beyond the currently available clinical and day-to-day management of CD individuals. Scale-up production, and respect of medical regulatory requirements will allow CUPIDO to deliver a final product for future late pre-clinical and clinical studies. Altogether, CUPIDO will foster the translation of nanomedical applications toward the cardiac field, which although still in its start, offers great potential to overcome the limitations associated to the currently pharmacological treatments.

The main objective of ERROR is to develop a new software tool, which will offer to the clinician the possibility to assess alternative imaging and therapeutic protocols, in real time, in silico, in order to minimize patient dose, while maintaining image quality of therapeutic effect. This tool will be designed, implemented and evaluated with specific focus on pediatric patients, since this is a rather sensitive target group, where dose considerations are high and no standard protocols and solutions exist. The project will exploit the new generation of computational anthropomorphic phantoms, in combination with well validated Monte Carlo simulations and Machine Learning Tools. In this way, it is envisaged that advanced, yet mature technologies will be integrated, to provide a novel tool, which can lead to a final product. The ERROR project brings together a multidisciplinary consortium of specialists in different areas of medical physics, biomedical engineering, physicians and computer engineers, who will join forces in order to design, implement and clinically assess a novel software tools, which initial focus in the optimization of diagnostic and therapeutic protocols for pediatric exams. Two new SMEs will provide their expertise, as well as investigate the ways to exploit project outcome. A well planned exchange program among academic and industrial partners will facilitate knowledge sharing, maximize collaborative work and finally achievement of project objectives. The consortium, being aware of the scientific and social importance of pediatric clinical applications, has planned a series of dissemination and training activities, aiming at making project knowledge and outcomes available to the scientific community and society.

Long-range research and innovation goal towards transformative biomaterials and technologies: The long-range goal of PEPSA-MATE is to develop and translate sustainable nanostructured functional materials, based on peptides and polysaccharides, using computer-based theoretical design approach and innovative fabrication technologies.PEPSA-MATE will generate a library of functional nanosaccharides and nanopeptides to be used as building blocks for the fabrication of advanced biomaterials. Key Research and Innovation Goals: The objective of PEPSA-MATE is to turn biopolymers into sustainable products using innovative technologies. This will result in alternative biomaterial to conventional, non-degradable synthetic polymers and inorganic materials. Scientific impact: Translation opportunities will be explored in three topic research areas; 1) Phytoglycogen nanoparticles to fabricate bioadhesives, bioplastics and drug carriers; 2) Ultrasonic fabrication of biofunctional nanopeptides; 3) Nanostructured saccharide microparticles as universal platform to deliver a wide range of therapeutic molecules. Impact on early-stage researchers: PEPSA-MATE involves key participation of junior researchers providing strong support for the development of their careers. Intersectoral dimension: PEPSA-MATE will develop intersectoral collaborations between a SME and academic partners to boost entrepreneurship creativity towards sustainable products and processes. Interdisciplinary dimension: PEPSA-MATE will bring together interdisciplinary and complementary expertise in the field of nanomaterials science, polysaccharides and peptides chemistry, acoustic technologies, cell biology, biophysics, and computational modelling. International dimension: PEPSA-MATE will enable long-term, transformative research collaborations between research groups and SMEs settled in European Countries (Italy, Czech Rep., Germany and Spain) and in four non-EU Countries (Australia, Brazil, Cuba and South Africa).