There are many types of heath data that can be raised, ranging from blood pressure levels right through to DNA sequencing. In order to be able to use artificial intelligence, large volumes of data need to be collected. This process has been going on for a long time now, and the amount of data gathered is growing tremendously. But in order to use this data, it needs to be both machine-readable and accessible. Many different players are already working on mastering these challenges. By seeking to improve standardisation and accessibility, their goal is to bring health data into the era of artificial intelligence so that it can be used for treatment and research.
Big data – How to use it well?
Just recently, Apple launched its Digital Health Platform in the USA. It opted to base the platform on the Fast Health Interoperability Resources (FHIR) standard – part of the Health Level 7 (HL7) set of international standards. Using FHIR makes it possible to process health data on smartphones and tablet computers and to incorporate it into existing systems. This is an important step in enabling machine-readable health data to be exchanged.
By contrast, in Switzerland, standard-setting and the collection of data is not left to the big technology companies. This is conducted instead by the non-for-profit cooperative midata.coop, which was set up in 2015, mainly to undertake research. Members of the public can decide whether they want to allow their data collected as part of scientific studies to be saved in the database or not.
Who manages the data?
The Swiss model is an attractive approach as allowing the control of health data to be completely decentralised has many risks. This is also the view taken by the Munich start-up Climedo, which has developed a smart digital assistant to help doctors undertake modern cancer treatment, tailored towards the patient, both safely and efficiently.
“It’s certainly an interesting thought that patients should decide themselves what happens with their data. If the data is not handled with the necessary care, however, this can lead to unforeseeable consequences for the patients,” says Climedo COO Veronika Schweighart. She believes that “a decentralised system that continues to be primarily managed by local doctors but that also sends relevant information to the healthcare stakeholders concerned could offer the same advantages but at a lower risk.”
The debate about who should ultimately manage health data is far from over. The Nuremberg Start-up IT-Labs is using its Alberta software to develop the first smart management platform that can be used by all parties involved in caring for patients with chronic conditions. “Obstacles mainly appear when we want to exchange data with other companies or institutions, such as clinics,” says Güven Karakuzu, CEO of IT-Labs, in an interview. When it comes to the management of health data, Mr Karakuzu believes we need to stop thinking in terms of extremes: “Discussion often revolves around solutions that are at two ends of the scale: should the data be managed by the state or a semi-governmental body, or should it be managed by the patients themselves? In the end, however, the reality is that there will be various sources of health data and various different owners”, says Mr Karakuzu, making a prediction about the future.
The main challenge, in the eyes of the CEO, is to ensure that all parties involved carry responsibility and liability, without this leading to a standstill in development. Many people find the public discourse unsettling and it is also leading to a further set of problems. Mrs Schweighart from Climedo points to the lack of acceptance for cloud storage in the healthcare market and the lack of affinity for IT among most decisionmakers, who perceive cloud storage as a danger rather than a major opportunity, despite the fact that it meets all data protection requirements. She opines that this greatly holds back the use of AI as big data requires the use of high computing capacity.
Artificial intelligence is a major driver of growth
The Munich start-up Climedo and the Nuremberg start-up IT-Labs – both supported by the ZOLLHOF Tech Incubator in Nuremberg and partners of Digital Health Hubs Nuremberg/Erlangen – are part of the effort to store health data in a system using a uniform format and to enable it to be retrieved in an efficient manner. Artificial intelligence and its applications are key future trends for both digital health companies.
Climedo particularly sees major potential when it comes to developing cross-hospital systems, built upon a structured data basis, that can assist with decision-making. In the future, such systems might not only make it possible to provide effective treatment, but to also see where preventative measures ought to be taken.
For IT-Labs, machine-based learning – built around the use of image recognition and data analysis techniques – is to help make processes simpler and more efficient for those using their software and to ultimately improve patients’ quality of life. In conclusion, it can be said that the progress made in AI is unlikely to be stopped. It is now up to decision-makers in the healthcare sector, in government and in business as to how fast this progress takes place or to what extent it will be hampered.
