Our main objective is to identify and address the bottlenecks in precision medicine. For us, this means developing and demonstrating implementation tools that create value for patients and clinicians alike. These can take the form of clinical decision support (CDS) implemented in electronic health records and other currently used applications, but can also be tools to automate data collection, such as large-scale phenotype extraction from medical records based on natural language processing. Each of the clinical focus areas in BigMed has different needs that will be addressed by the development of different tools. As we work with this development, we will describe the barriers and develop possible solutions in collaboration with a wide network of internal and external partners.
Understanding the underlying genetic code causing disease is key to precision medicine, be it personal genomes, cancer genomes or microbial genomes. With improved availability and affordability of genomic sequencing in the last decade, the challenge has shifted towards efficient and effective bioinformatic analysis of the output of genomic sequencing. The bioinformatic pipelines currently in use vary greatly from site to site. There is urgent need for standardization of reporting and use of pipeline components, to ensure equal quality of input to genetic clinical decision support systems. To this end, BigMed will develop and define high-quality bioinformatic pipelines which will contribute to define, design and evaluate elements of clinical decision support systems.
Optimal implementation of precision medicine is reliant on sharing data on genes, variants and phenotypes. Access to large data sets of comparable genome information is fundamental to the ability to analyze each patient’s genetic fingerprint, and to understand the implications of individual variants. Differences in variant frequencies between populations further underline the importance of large and ethnically diverse data sets. BigMed will make clinical genomics data more available and actionable through the construction of genomic databases and decision support tools.
The sensitive nature and volume of data needed for Precision Medicine practice poses new challenges for IT infrastructure. Storage capacity, and processing and transmission speed requirements, are higher than what today’s standard environments can deliver. An architecture is also needed that can deliver real-time information and tools for advanced analytics such as machine learning and related techniques. Finally, the solution will need to comply fully with local, national and regional regulations for privacy and security, with cost-effectiveness. Bigmed will approach this in two phases, first by using already established high performance infrastructure provided by the University of Oslo. Later by establishing a platform to simulate the real hospital ICT environment and demonstrate how the BigMed solutions can operate in the hospital environment in a near future.
The learning healthcare system of tomorrow has the potential to continuously integrate data and implement new knowledge – meaning immediate benefit to the patients. Practically, this means using experiences from all patients – which until now has been considered to be private and personal health data - in order to individualize and tailor diagnosis and therapy for another patient. This challenges our thinking about primary and secondary use of clinical data. In BigMed, we want to address this and other issues related to privacy. Which issues can be resolved within today’s legal boundaries, and what policy and legislative changes are needed in order to realize the full potential of precision medicine? Which ethical considerations do we need to take into account in order to ensure the future that we as a society wish for? And how do we ensure that patients’ rights and integrity are being ensured as technology advances? BigMed will establish a group of lawyers dedicated to these questions and will work on both considerations and recommendations relevant to both policy makers and the public .
Colorectal cancer is the second most frequent cancer and represents 13.2% and 12.7% of all cancer cases in men and women, respectively. Approximately 50 % of all patients will develop liver metastasis and surgical resection is currently the only curative treatment for these patients. Novel molecular diagnostics of the tumors together with an expanding array of other diagnostic tools is challenging the decision processes for choice of treatment and patient pathways. There is a need for improved clinical decision tools to support this development.
Monogenic diseases result from inborn errors in a single gene. As of today about 5,500 of the 20,000 human genes have been associated with disease, and it is estimated that 5-8% of the population suffers from a genetic disorder. Monogenic diseases are notoriously difficult and very costly to diagnose from clinical findings. Novel high throughput sequencing has lately proven to be superior to the traditional one-gene-at-a-time testing and increases the yield of correct diagnosis. The method does, however, require standardized patient phenotyping, high-performance computing and data sharing for optimal output.
Sudden Cardiac Death (SCD) accounts for approximately 5,000 deaths yearly in Norway. High-risk individuals when identified can be offered life-saving therapy. Selection of patients for this therapy and prediction of SCD is one of the greatest challenges in current cardiology. Use of patient-specific genetic data combined with clinical data is instrumental in risk stratification for SCD.
Frostbites BIGMED activities will support and benefit from current initiatives in Norwegian military medicine to investigate the use of big data methodologies to better understand medical challenges to cold weather and arctic operations. Cold and frost injuries in the Norwegian Armed Forces are estimated to affect as much as 2% of the conscripted soldiers on an annual basis. A ten years cohort based on the military medical records is planned to be investigated based on methods made available through the BIGMED project, illuminating aspects of big data series.