EMERALD takes a unique, holistic approach to patient-specific predictive modeling and MDSS development by extracting and integrating knowledge from new research, clinical tests and EHR using advanced analytic techniques.

ICT technologies (such as Data Mining, Deep Learning (DL) and Advanced Fuzzy Cognitive Tools) will play a key role in EMERALD enabling the analysis simplification of large patient data collections, explainability of decisions made and thus allowing the development of personalized predictive MDSSs. Additionally, dynamic FCMs as a soft computing technology, will play a major role in EMERALD concerning the model-driven data analytics to support decision-making focusing on data interpretability and modeling complexity (i.e. personalized treatment and health technology assessment). Knowledge obtained from EHR and FCM-based models will be combined with experts’ knowledge concerning other risk factors (e.g., social, environmental, occupational and economic factors) to build high-level FCM-based MDSSs. EMERALD will further introduce the new concept of DeepFCMs as the innovative and structural component of an XAI-MDSS. DeepFCMs can fuse the multitude of medical data spanning from text to images and they can be executed as typical deep neural networks but they can be presented visually as FCMs to provide visual explanations and reasoning. Moreover, DeepFCMs are primarily FCMs thus, they can be endowed with self-explanatory capability (as other fuzzy models) providing non-expert users with linguistic descriptions and explanations in Natural Language that facilitate the comprehension of the given visual explanations. Finally, EMERALD aims to create a medical science ecosystem that will optimize physician decision making through NM and AI. To this direction, a multimodal repository of heterogeneous EHR will be orchestrated via a unique, holistic approach to patient-centered predictive modeling that incorporates dynamic risk assessment capabilities at every step of DL. The proposed models will provide decision making based on:

  • The data analytics focusing on data interpretability and modeling complexity.
  • The ability to prioritize and predict the influence of risks and outcomes of complex and critical systems.
  • The provision of high-value healthcare, concentrating on quality (patient outcomes, safety, service).

The explainability of medical decisions in diverse medical domains such as CAD and NSCLC and the dissemination of well-defined protocols to promote optimal care for patients with chronic conditions.

For more details, you may visit the official EMERALD website