General 1 April 2019
Better use of health data could head off diabetes complications, tailor personalised cancer treatments and deliver smarter prescriptions, according to a new white paper from the Digital Health Cooperative Research Centre (DHCRC). The key to these and more life-saving possibilities is to connect existing health records with new kinds of data from sensor-driven mobile devices and omics related information, DHCRC researchers say.
“Modern medicine focusses on treatment rather than prevention and it’s the same for health data,” say Dr Uma Srinivasan and Joanna Khoo, co-authors of the white paper, Harnessing the full spectrum of digital data to support the delivery of personalised services across the health care continuum. “It’s time for that to change.”
The white paper outlines the different ways the public engages with health and the different kinds of health data that is captured by both clinicians and consumers.
The scope of this data is growing all the time. In addition to data stored in electronic health records, the rise of medical and non-medical devices connected to the Internet of Things (IoT), and the sequencing of genomic, phenomic and metabolomic data (collectively known as omics data) also have the potential to offer new insights in understanding health risks and personalising healthcare services.
“Most health data analytics techniques so far have focussed on structured health data, but more than 70% of digital data generated by the healthcare sector is unstructured data such as X-ray, MRI and ultrasound images, audio and video recordings, and we have not yet tapped these resources,” says Dr Srinivasan.
New approaches using artificial intelligence and machine learning technologies and omics sequencing tools will play an increasing role in the health cycle in coming years, the authors say.
With the growth in volume and variety of data, infrastructure too needs to keep pace to enable information discovery across diverse data sources. Robust infrastructure that can to accommodate differing data sets and types will be required, as will appropriate governance. The white paper presents a conceptual architecture with appropriate tools, and a framework for research governance based around risk and preserving privacy when managing diverse data sets.
Because of the sensitive nature of health data new systems must build in robust data security measures, says Dr Srinivasan, who is also a co-author of the DHCRC’s first two Flying Blind health reports which look at how consumers and researchers access digital health information.
“New systems should also clearly define what data was collected, how, when and by whom from the beginning,” adds Jo Khoo. “The more consistent the collection process is, the more useable the data will be for patients, health professionals and researchers.”
Dr Srinivasan works at the DHCRC, which was established in July 2018, and Jo Khoo recently submitted her PhD.
“We have ambitious plans to improve the health of all Australians,” says Dr Victor Pantano, CEO of the DHCRC. “We’ll do this through researching information capture, storage and flow, identifying and managing health risks, improving value, quality and safety through intelligent support and understanding clinical practice to support transparency and improve performance.”
The white paper has already been used as a reference by the Western Australian Department of Health, which is presently developing a health and medical research and innovation strategy.
The full white paper can be found here.
Sign up for the DHCRC newsletter.