Project Vision: EOSC Web of FAIR Data and Services for Science is an open, fair and reliable Research Community where every researcher will be accredited for their work and all research data will be equally accessible for processing without violating data protection regulations. In line with this vision, the mission of RAISE is to provide the infrastructure for a distributed crowdsourced data processing system, moving from open data to open access data for processing. RAISE will provide the mechanism for sending the algorithm to the dataset instead of sending the data to the algorithm.
Role:We will run a prediction use case with pre-existing data from AUTH. In this use case we will try to identify how the IPRs can be shared for algorithms development by a company with datasets that belong to another organization as well as the ethical issues that might arise from the secondary use of data. The goal of the use case will be the prediction of progression of Multiple Sclerosis.
Project Vision: RELEVIUM is an EU-funded project under the HORIZON-EUROPE framework. RELEVIUM’s main objective is the development of multi-modal interventions for the improved palliative care of cancer patients. RELEVIUM focuses on Pancreatic Ductal Adenocarcinoma (PDAC) patients for which different barriers to effective supportive and palliative care reduce their Quality of Life and overall survival. RELEVIUM aims to address these issues by proposing a multimodal supportive intervention based on continuous remote monitoring technologies for (i) pain estimation, (ii) estimation of muscle mass/sarcopenia, (iii) monitoring of nutrition, as well as (iiii) physical activity of advanced PDAC patients. The project will first conduct a feasibility and data collection study, and then a randomized clinical trial involving 5 clinical cancer centres in Germany, France, Belgium, Estonia and Israel. These studies will assess the effectiveness of the proposed interventions in improving the quality of life of advanced PDAC patients. Furthermore, several important parameters will be explored, such as (i) the cost-effectiveness of the proposed solution, (ii) its potential in increasing health equity across the populations of the participating countries, in terms of access to supportive and palliative care for advanced PDAC patients and (iii) the stress burden of the disease on the families and family caregivers of the patients. With these strategies and the proposed AI-based solutions, RELEVIUM aims to empower PDAC patients to self-manage their disease, reflect and optimize their individual QoL, and receive improved clinician guidance, independently of their home and country. AINIGMA has an active role in the project, through the research and development of a sophisticated algorithm for sarcopenia estimation and prediction.
iPROLEPSIS aspires to shed light upon the health-to-Psoriatic Arthritis (PsA) transition with a comprehensive multiscale/multifactorial PsA model employing novel trustworthy AI-based analysis of multisource and heterogenous (i.a., in-depth health, environmental, genetic, behavioural) data, digital phenotyping of inflammatory symptoms with emphasis on tracking of motor manifestations using smart devices and wearables, novel optoacoustic imaging-based markers of PsA in the skin and joints, and investigation of the role of mast cells in the PsA transition, to identify key drivers of the disease and support personalized models for PsA risk/progression prediction and monitoring as well as associated inflammation detection and severity assessment. AINIGMA will create dynamic, multiscale/multifactorial, and personalised xAI-driven models of the health-to-PsA inflammation transition that will enable prediction of risk and early diagnosis of PsA in people at risk, with emphasis on patients with psoriasis (PSO), as well as prognosis of disease progression targeting the prevention of PsA inflammation exacerbation.
ONCOSCREEN will develop a risk-based, population-level stratification methodology for Colorectal Cancer (CRC), to account for genetic prevalence, socio-economic status, and other factors. It complements this methodology by a) developing a set of novel, practical, and low-cost screening technologies with high sensitivity and specificity, b) leveraging AI to improve existing methodologies for CRC screening, allowing for the early detection of polyps and provision of a personalized risk status stratification, and c) providing a mobile app for self-monitoring and CRC awareness raising. Furthermore, ONCOSCREEN develops an Intelligent Analytics dashboard for policy makers facilitating effective policy making at regional and national levels. Through a multi-level campaign, the above-mentioned solutions are tested and validated. For the clinical solutions specifically, a clinical validation study has been planned with the participation of 4100 enrolled patients/citizens. AINIGMA will participate in the development of AI Algorithms for Polyp Detection & Classification from Real-time Colonoscopy images. Most cases of CRC start as a growth on the colorectal epithelium, called polyp. CRC can be prevented with the early detection and recognition of the type of polyps, and colonoscopy is the main diagnostic procedure. However, the visual detection and classification of polyps can be challenging due to several factors, including the illumination conditions of colonoscopy, the variability of texture and appearance, and the overlapping morphology between polyps. Moreover, the evaluation of polyp patterns by gastroenterologists is subjective leading to a relatively high interobserver variability. Even for experienced endoscopists, working on conventional colonoscopy for long hours leads to mental and physical fatigue and degraded analysis and diagnosis. Within the ONCOSCREEN project, AINIGMA will participate in the development of Deep Learning algorithms for the accurate detection and classification of colorectal polyps, with the aim to develop a computer-aided diagnostic tool for the improved screening of CRC.
BioStreams aims on supporting the optimal use (and re-use) of health data (e.g. biological, demographic, epigenetic, etc.) to generate metadata/knowledge and provide new evidences, methodologies & tools for: a) creating and deploying the Bio-Streams Biobank that will act as a scientific platform for research in obesity and better diagnostic and therapeutic approaches; b) understanding the transition from metabolically healthy to unhealthy, preventing under age obesity; c) designing better strategies to educate & empower young citizens for weight self-management; d) coordinating authorities & policy makers to develop cross-sectoral solutions for health promotion & under age obesity prevention. Specifically, Bio-Streams delivers a dedicated underage obesity biobank providing real-world health data (including biospecimens, anthropometrics, behavioural and cost data) from retrospective and prospective sources, while taking into account efficient data harmonisation and standardisation principles, transforming data valorisation towards underage obesity prevention and future research. Moreover, the Bio-Streams framework for data handling, provides robust and transparent methodologies for operational procedures (data infrastructures), analysis and reporting (via meta-reviews and AI-based Apps/components, evaluated via 12 multi-site pilots (7 clinical and 5 schools), citizen awareness and lifestyle alteration, delivering obesity prevention guidelines, knowledge generation while fostering considerable opportunities for regional and national health authorities/policymakers. Within BIO-STREAMS, AINIGMA is responsible for the development of an AI-powered Risk Stratification Tool and Recommendation Engine, aiming to improve management of obesity in childhood and adolescence.
Following a trustworthy and inclusive approach to AI development, multidisciplinary expertise and broad stakeholder engagement, AI-PROGNOSIS aims to advance PD diagnosis and care by: i) developing novel, predictive AI models for personalized PD risk assessment and prognosis (regarding time to higher disability transition and response to medication) based on multi-source patient records and databases, including in-depth health, phenotypic and genetic data ii) implementing a system of digital biomarkers informing the AI models by tracking key risk/progression markers in daily living, and ultimately and iii) translating the models and digital biomarkers into a validated, privacy-aware AI-driven toolkit, supporting healthcare professionals (HCPs) in disease screening, monitoring and treatment optimization, via quantitative, explainable evidence, and empowering individuals with/without PD with tailored insights for informed health management
Our role: AINIGMA will develop the PD progression predictive models for the AI-Prognosis.
The objective of DIOPTRA is to tackle certain challenges in Colorectal Cancer (CRC) detection, screening, and progression by introducing advanced technologies. DIOPTRA aims to develop a primary screening tool for identifying individuals at high risk for colorectal cancer (CRC) based on risk factors and protein biomarkers. Blood and tissue samples will be analyzed to identify a specific set of prognostic proteins that can be detected through standard blood tests, indicating the need for further evaluation, such as colonoscopy. In addition to medical data, behavioral data will also be considered as a potential risk factor. Artificial intelligence (AI) will be utilized to evaluate the prognostic power of the biomarkers, while personalized behavioral change strategies will be implemented to address modifiable risk factors. The project will also explore the effectiveness of network modeling and AI-based decision support systems for early CRC detection. AINIGMA has an active role in the part of AI utilization in the project, through the implementation of ML models and multi-modal data fusion for the development of a robust and explainable AI-powered Decision Support System for CRC screening.
PREVENT aims to enhance the implementation of primary interventions for managing weight control during childhood and adolescence, with the objective of reducing cancer risks in adulthood. This approach is based on current evidence linking excess body weight with an increased risk of cancer. To achieve this goal, PREVENT adopts several implementation research strategies. First, it identifies the obstacles that prevent current interventions and policies from being scaled up to various geographical, socio-economic, and cultural settings. Then, it introduces new interventions that are context-sensitive and multi-stakeholder in nature, along with innovative user engagement tactics to overcome existing scaling-up difficulties. Within PREVENT, AINIGMA is responsible for the development of trustworthy AI-powered toolkits for the improved data processing and analysis.
NERO is an advanced Cybersecurity Ecosystem consisting of five interrelated frameworks provisioned to offer a Cybersecurity Awareness program, as recommended by ENISA to be the best way to educate and develop a security-first culture amongst employees on how to mitigate the impact of cyber threats and incorporate activities, resources, and training to foster a cyber security culture. NERO primarily provides SMEs with a Cyber Immunity Toolkit Repository, Cyber Resilience Program, and Cyber Awareness Training via Gamification and provides its modules through a user-friendly Marketplace. The effectiveness and the performance of the concept will be validated in three use case demonstrations on different domains: Enhancing Patient Data Security in Healthcare through Cybersecurity Tools; Strengthening Supply Chain Resilience through Cybersecurity Awareness in the Transportation and Logistics Industry; and Boosting Financial Security through Enhanced Cybersecurity Awareness and Tools.
PHASE IV will advance the current state-of-the-art data synthesis methods towards a more generalized approach of synthetic data generation. We will also develop metrics for testing and validation, as well as protocols that enable synthetic data generation without access to real-world data (through multi-party computation). We aim to provide: 1) Improved methods and technical pipelines for privacy-preserving data synthesis including different data formats such as EHRs and medical images, 2) Easy to use and configurable data services to enable AI developers’ access to larger pools of decentralized de-identified data through multi-party computing, 3) Provide anonymous data on demand or from a (temporary) repository, 4) Establish a Data Market – facilitating data sharing and monetization incl. incentives-based provision of data to the services, 5) Integrate the data market and the data service ecosystem as a X-European health data hub in the European Health Data Space, and 6) Validate the results with real-world use-cases focusing on high impact diseases, cancer types in particular.
Partners and Collaborators:
