Collect Data? Utilize Data! – The Blessings of Big Data
Collect Data? Utilize Data! – The Blessings of Big Data
Genome data, MRI images, and blood test results – data collected in the medical sector is not only very heterogeneous but also extremely extensive. However, it is important to not only collect this data but to also utilize it. After all, processed, linked and analyzed data provides many opportunities in research, hospital management and ultimately also for the individual patient.
Cancer, heart attacks or neurodegenerative diseases are disorders that result from malfunctions in organs or cells. Responsible for this is a complex correlation between genetic aspects but also environmental influences and lifestyle habits. It is difficult to not only understand the connections but to also untangle them to where the causes of the disease can be identified. Systems medicine makes this possible.
Systems medicine: systematic interdisciplinary research
Collaborative teamwork of the different types of special fields is needed to permit an integrated view of the body: biology, medicine, biochemistry, mathematics, and computer science. "The constant interplay between modeling and laboratory work is the core of systems medicine. Based on what has been discovered in the laboratory, computer models for specific aspects are being created. These are then in turn reviewed in the laboratory. In doing so, you are gradually moving towards an accurate result," explains Dr. Silke Argo, Head of the Germany Research Network e:Med, which is funded by the BMBF (German Ministry of Education and Research). Argo sees the advantage of systems medicine in interdisciplinary collaboration. "Innovative solutions are created when researchers from different disciplines with different research approaches jointly consider a medical issue."
Big data plays a key role in systems medicine. "The reason we keep getting better results is because there are vast amounts of data available on an increasing number of patients to use for research purposes," says Argo. Whether it is the human genome or proteins acting in cells – the whole organism can be analyzed in its complexity with the help of so-called “Omics“ technologies (genomics, proteomics, and metabolomics). Systems medicine utilizes this vast data trove. But other data collected from blood, biopsies or imaging is also being used.
More than 33.5 million people worldwide are affected by atrial fibrillation and the number of sufferers continues to increase. Yet so far, there are only a few known risk factors, while the exact cause is still unknown. The e:Med symAtrial project intends to change this. "So far, the exact molecular causes of atrial fibrillation could not be precisely determined. This is why our interdisciplinary team of physicians, molecular biologists, bioinformatics scientists, epidemiologists, and statisticians wants to adopt a systems medicine approach to not only examine individual features of atrial fibrillation but to interconnect the entirety of all individual aspects and make a clearer statement on atrial fibrillation," explains Professor Tanja Zeller, spokesperson for the e:Med executive committee and Professor of Genomics and Systems Biology at the University Medical Center Hamburg-Eppendorf. She sees the goal of systems medicine in individualized medicine. "In the long term, our approach intends to improve patient treatment on a personal level or even to be able to state up front, who will be affected by atrial fibrillation."
Argo is confident about the systems medicine approach: "This is an approach that can yield many far-reaching results and answer many questions that have long remained unanswered. The intelligence of this approach to think beyond borders, to network and ‘think outside the box‘ is precisely the way to yield many surprising insights that can ultimately benefit the patient."
Analyzing, linking and interpreting patient data
Meanwhile, there are more and more large genome collections that systems medicine is able to access. Britain’s 100,000 Genomes Project aims to – as the name already indicates – sequence 100,000 genomes. The Netherlands have the GoNL (Genome of the Netherlands), Saudi Arabia the Saudi Human Genome Program and in the U.S., former President Obama announced the Precision Medicine Initiative in 2015, which aims to collect genetic and medical data from 1,000,000 people. Genome sequencing creates a data pool that makes new scientific discoveries and medical insights possible.
In contrast, the Clinical Data Intelligence (KDI) project does not intend to collect new data but wants to link already existing data. Whether it’s genomic data, data from ECG recordings or medication data, whether it’s structured or unstructured – all available patient data is meant to be consolidated from different types of sources and automatically evaluated.
Watson and others – computers for daily hospital routines
Meanwhile, data volumes have reached a magnitude that can no longer be utilized without artificial intelligence (AI). AI systems work with self-learning computer algorithms. Their strength is particularly evident in the analysis of large data sets. The chances of creating links are higher with larger accessible volumes of data.
AI systems can be used in many areas of daily hospital processes and are able to support medical professionals in their work: they can analyze CT scans in radiology and use other images to compare; they can read the ever-increasing research literature, analyze it and thus support the decision-making process or compare the genome of one patient with that of another and suggest an applicable therapy.
Perhaps the best known artificial intelligence system is IBM‘s Watson. It has access to millions of databases and is thus especially supportive in combating rare diseases. After all, each year approximately 3,000 new studies are being published. It’s impossible for one health professional to read them and remember them all. Recently, the power of Watson became evident: after physicians were unable to diagnose a Japanese patient, Watson compared her genetic data with data from 20 million studies. After just ten minutes, Watson was able to diagnose a rare type of blood cancer.
Big data analytics also plays a major role in hospital management in support of hospital operations. After all, efficient processes not only benefit patients but also pay off financially – for instance, in the operating room. That’s why the goal of the InnOPlan project is to turn operating room equipment into intelligent data sources. “The goal is to make clinical processes more efficient, with an emphasis on the OR and its environment,“ explains Dr. Lars Mündermann, Project Manager Applied Technology Research, KARL STORZ GmbH & Co. KG in an interview with MEDICA-tradefair.com.
Big data plays a crucial role in medicine. However, large volumes of data only provide an added benefit if data is not only being collected and stored but also intelligently processed. That is when big data turns into smart data and offers many possibilities for research, patients, and hospitals.