The BD2Decide project realises and validates an Integrated Decision Support System linking together population-specific epidemiology, behavioral and environmental data, patient-specific multiscale data from genomics, pathology, clinical and imaging data with available multiscale prognostic models and Graphical Visualization tools that allow to:

  1. Improve the clinical decision process for patients diagnosed HNC, by implementing a model-based prognostic system which increases the accuracy of current TNM staging system at least by 10%.
  2. Validate the existing and newly developed models in different populations to either adjust model or develop general applicable models.
  3. Uncover patient-specific and population-related patterns that can improve care and pave the way for better tailored treatment guidelines, by assessing the importance (scoring) of new prognostic markers as compared to TNM staging for outcome prediction.
  4. Demonstrate in a proof-of-concept clinical study that the project contributes to improve the patient’s quality of life of at least 15% of patients.
  5. Reinforce the multidisciplinary decision-making process and patient's co-decision through advanced data visualization and presentation, measured by the usability evaluation of the system by the multidisciplinary team involved in HNC management (at least 80% positive usability scores).
  6. Create a virtuous circle of learning between research and clinical practice, through the mutual feeding of external population data and patient-specific clinical data, and the creation of a large and shared data repository for HNC (at least 2.000 cases from different populations).

The above overall objectives translate into the following technical objectives:

  • Deploy Big Data techniques to exploit the value potential of prognostic prediction. The main goal is to explore the effectiveness of cloud computing applied to healthcare decision making process and the usefulness of big data for the discovery and validation of personalized prognostic patterns for HNC that outperform current TNM system, through:
    • Setup of Big data and cloud infrastructure to collect and homogenize data, in compliance to state-of-art standards.
    • Application of Big data analytics and reasoning techniques, used to categorize each individual patient, each HC sub-type, find person-specific prognostic patterns as compared to the reference population of patients and to automatically apply the most suited prognostic models.
    • Integration and assessment for clinical use through a Knowledge Management System.
  • Enrichment and refinement of multiscale personalised prognostic models for improved prognosis prediction compared to TNM system, through:
    • Extension of prognostic models enriched with the increased Big Data input, to precisely stratify each patients at baseline and to produce personalized prognostic signatures.
    • Prognostic factors scoring, to provide physicians an insight of he added value brought by each prognostic factor.
  • Refine and validate advanced imaging and radiomics tools aimed at the discovery of new prognostic signatures, namely:
    • a functional imaging analysis and features extraction tool, that will extract imaging features and calculate tumor and lymph-nodes volumes from CT/MRI through segmentation algorithms.
    • a radiomics software applied to CT, MRI and DWI MRIs to capture phenotypic heterogeneity in tumors.
  • Development of a highly interactive visualization and presentation suite that acts as the key in HNC understanding and treatment. More specifically, BD2Decide will provide:
    • Realization of Digital Patient data exploration tools.
    • Realization of the Patient co-decision environment, aiming at the active engagement of patients in the personalized therapeutic process in line with the “no decision about me without me” initiative.
    • Development of assistive visualization and data presentation aimed at research purposes, focusing on the exploitation, representation and visualization of different types of information retrieved from large-scale and heterogeneous sources.
  • Clinical validation of the prognostic models and of the overall decision support in different EU populations, to entrust physicians, allow use in clinical settings and possible qualification as medical device.