FEATURE
https://www.dzd-ev.de/en
Connecting the Dots Dr Alexander Jarasch and Prof. Dr. Dr. h.c. mult. Martin Hrabě de Angelis emphasise the
potential that huge amounts of data hold in helping to improve medicine and the care sector. However, they stress that researchers need to embrace new technologies in order to do so.
As part of its Grand Challenges approach, the UK Government is looking to utilise big data and artificial intelligence (AI) to tackle a number of areas including early diagnosis of diseases such as heart disease, diabetes and Alzheimer’s, leading to better prevention and treatment coupled with enhanced care of an ageing population.
Within 15 years, better use of AI and big data could lead to over 50,000 more people per year having their cancers diagnosed earlier, for example. This could mean around 20,000 fewer people dying within five years of diagnosis, compared to today.
The UK isn’t alone. Other countries, including our homeland of Germany, are looking to address these pressing issues. But the simple truth is, we need to take a much deeper dive into these chronic conditions and diseases to find not only better ways of preventing them – but also advancing treatments. The problem, however, is that, due to the vast amounts of data produced, classic analysis methods may have reached their limits. What we need is a dynamic way to leverage and connect big data to provide new valuable insights.
COMPLEX DATA POINTS
Research today, especially in the field of Life Sciences, is no longer built around one technology or discipline. We carry out research, for example, at The German Centre for Diabetes Research, the country’s national centre for studying diabetes. Here, you will find a multi-faceted organisation covering studies being carried out in all corners of the country. As you can imagine, we have gargantuan amounts of data from clinical trials and patient information covering a host of disciplines.
But how do we exploit all this big data? If we need to answer biomedical questions about diabetes, we know that we have to join the dots in this data to form new patterns and ultimately new understanding by linking in as much data as possible to our research.
This is the next leap forward for biomedicine and the healthcare sector in general. We are already seeing a move away from general, all-encompassing drugs to more targeted, individual treatments. For this to truly advance, we need to network more, tap into and connect more data. This is exactly why DZD and other researchers are looking to graph databases to help see pictures in data that can help in the prevention, early diagnosis and treatment of major illnesses across the globe.
Our research teams were using well-worn and proven technologies such as relational databases, spreadsheets and
- 40 -
document files. But, as we have started to see the value in connections in big data to our research we have fast realised we need a far more powerful tool. One that can take data out of silos and actually make the join-ups for us, turning it into actionable intelligence. We have found the answer in graph technology.
Take diabetes, for example. It is a metabolic disease, but researchers need to dig much deeper than simply siſting through metabolic data. They need to look at other disciplines such as genomics. The human body is, aſter all, connected via metabolic pathways, so this makes logical sense. Connecting these data sets to show correlations enables us to finally visualise this data.
REALISING THE LINKS
The ability soſtware gives us to join these connections and manipulate massive amounts of data in real time is invaluable in our research.
To understand these complex data relationships, we have been using a graph technology called Neo4j. It provides us with a visual interface, which we can utilise for queries and experimentation, whilst at the same time dramatically speeding up data analysis. We are using graph database soſtware to mine deeper into our diabetes ‘map’ to seek out hidden relationships, allowing us to examine new avenues of research.
At the same time, we are looking to build new data models and compare areas, such as that of animal and human data we have collected. In a graph representation, abnormalities, patterns or connections can be easily picked out and questioned. In the future, data from diabetes research could be integrated with that of Alzheimer’s research, for example, to discover possible connections. The possibilities are endless.
We are also looking to exploit the power of Machine Learning (ML) coupled with graph technology to identify new subtypes of diabetes. It will eventually be possible to create predictive models that will track the stages of a disease to a certain degree of probability.
This data management and analysis approach will undoubtedly be the way forward in precision medicine, prevention and treatment of diabetes – and undoubtedly other diseases. Graph technology’s innate ability to discover relationships between data points and allow us to understand them have an enormous role to play in medicine and healthcare in the future. Making these data connections is only just the beginning.
www.dzd-ev.de/en www.tomorrowscare.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46
