Precision Medicine Requires Data Storage, Infrastructure Upgrades

data storage

– Precision medicine will require significant investments in data storage, infrastructure, and security systems in the coming years to achieve its full potential, a recent report by the California Precision Medicine Advisory Committee concluded.

Data storage and infrastructure must keep pace with the demand to store, manage, and share patient data, which is expected to increase for each patient from several gigabytes to terabytes of generated data soon, stressed the report. The commission was set up by outgoing Gov. Jerry Brown in 2017.

The report acknowledged that there have been advances in data storage and infrastructure since the early 2000, most importantly, cloud computing.

“With the expansion of cloud-based and hybrid networks, distributed computing is now available providing better support for collaborative productivity tools. Trends show that health systems are increasingly migrating to software and cloud-based solutions to store and protect data,” the commission related.

Governments, healthcare organizations, and the scientific community will need to address several challenges involving management and sharing of increasing volumes of medical data.

First, the storage and computer costs of biomedical data initiatives should be factored in from the start. Cloud computing requires a substantial investment for large amounts of data and analysis.

Second, data quality needs to be assessed because reliable data is essential for meaningful analysis. At the same time, assessing data quality is not easy given the diversity, breadth, and depth of medical data.

Third, health data needs to be managed and stored in a secure environment. Precision medicine projects will be required to develop data security programs that incorporate HIPAA Privacy and Security Rules and NIST information security standards.

“Overall, there will continue to be a balance to promote broad biomedical discovery while factoring in the costs, quality, and security needed when assessing these data,” the report observed.

“Health systems are increasingly migrating to software and cloud-based solutions to store and protect data.”

In addition, the report stressed that improvements in data sciences and analysis and training for the next-generation of data scientists in life sciences will be needed to assess the volume and diversity of emerging biomedical data.

According to a 2016 survey cited in the report, data scientists currently spend 80 percent of their time collecting existing datasets and organizing data. Less than 20 percent of their time is available for data mining for patterns that generate breakthrough discoveries.

“As tools continue to improve for data collection and organization, the aim is to further allow data scientists to contribute more robustly to the next generation of precision medicine efforts,” the report noted.

New systems are available for the collection and organization of data, as well as artificial intelligence, machine learning, and deep learning strategies to accelerate precision medicine.

Analytic tools developed by a range of other disciplines will have applications to biomedical research and precision medicine, the report predicted.

Data sharing is a key component of precision medicine. Data needs to be shared to enable the sort of analysis and insight that produces scientific advancement.

There have been a few initiatives to spur data sharing, yet these efforts continue to be fragmented, creating a challenge to realizing precision medicine’s potential.

The report cited the need for public-private, cross-sector coordination, and innovation in the area of health data sharing.

California is currently cooperating with the World Economic Forum on an international data sharing effort. Data sharing models will be transformed over time. New models, where participants are paid for sharing their data, are starting to emerge.

“Data are the fundamental inputs for precision health and medicine analysis and understanding. Analyses will ultimately feed back into the learning healthcare system to improve clinical care. When data from genes, proteins, environment, lifestyle, and behavior are evaluated at the same time, the findings will yield unparalleled insights,” the report concluded.

Source link


Please enter your comment!
Please enter your name here