IMAGica: An Integrative personalized Medical Approach for Genetic diseases, Inherited Cardiac Arrhythmias as a model

The chance a person is diagnosed with a disease and its prognosis are highly dependent on multiple factors including clinical-physiological (anamnesis, observations and measures), genetic (‘mutant genes’), and, less clear and to be studied in depth, psycho-social (behaviour, emotional problems, stress, insufficient social support, etc.) and environmental factors (exposure to smoking, alcohol, drugs, malnutrition, etc. – pre and post-natal). Today, these factors are mainly been studied separately and due to their typical epistemological and methodological traditions, genuine cooperation among these domains is scarce. However, potentially important and statistically significant associations and synergistic interactions may exist between factors from these different dimensions. Thereby, what is missing is a thorough understanding of the complex interplay between all these factors, as they need to incorporate knowledge from different disciplines like medicine, biomedicine, engineering, humanities and psychology. Therefore, it is crucial to study the interlink between various dimensions and try to find out answers for fundamental questions, like why certain individuals with a ‘mutant genes’ do not develop the disease or have a better prognosis.

Since the decryption of the human genome in early 2000’s, there has been a lot of emphasis on analysing the genomic data and mine out knowledge for early identification of diseases and providing better diagnostics. Computational developments like databases and other big data storage methods have helped in storage of this avalanche of structured and unstructured data, followed by techniques of data mining and machine learning, which have played a crucial part in revealing the hidden knowledge. In the scope of this project, beside the genomic data, clinical-physiological, environmental and psycho-social  data will be collected and analysed with the aid of interpretable machine learning techniques. This will enable to perform multidimensional  associative analyses of the data. The interplay of all these factors is very specific to the individual patient and hence a more patients-specific and integrative approach can be acquired, including preventive measures, follow-up and/or therapy tailored to the specific person, in other words: personalized integrative medicine.

Consortium:

  • Artificial Intelligence Lab, Vrije Universiteit Brussel
  • Centre for Medical Genetics, Universitair Ziekenhuis Brussel
  • Heart Rhythm Management Centre, Universitair Ziekenhuis Brussel
  • Mental Health & Wellbeing Research Group, Center for Neuroscience, Vrije Universiteit Brussel

Project Info

Start: 01/01/2016

End: 01/01/2021

Funding: IRP VUB

Involved Members: