Connected health tools (a.k.a. IoMT devices) are quickly becoming the norm throughout the provider space. From hospitals and specialty practices to walk-in clinics and home health services, such tools are already generating massive amounts of data each and every day. Data that must be integrated with other data. Data that must be stored. Data that must be shared. All in a single, powerful IT infrastructure. Getting it right is critical for a number of ?right now? reasons (compliance, brand confidence, etc.). But there?s another reason to get it right: Structuring and storing your big data properly now will make it easier for your organization to embrace ?what?s next,? specifically machine learning.

According to PointClear Solutions? Vice President of Technical Services Michael Atkins, healthcare organizations must make sure their networks are secure enough to absorb and manage multiple connected devices, including their attendant bandwidth and storage requirements.

?Your infrastructure must also be able to scale horizontally to accommodate changes in the volume, velocity, variety, and veracity of your IoMT data,? he adds.

What other steps should healthcare providers take today to help ensure their data is ready for a machine learning-centric tomorrow? Atkins suggests these five things:

1. If you haven?t done so already, adopt an IT infrastructure based on modular and open architecture principles that make it easy to add or update components and integrate machine-learning solutions as plug-in functions to your EHR.

2. Develop internal policies and mechanisms to transfer health data into and out of legacy systems securely, so that you can build, test, and deploy machine-learning algorithms at a later date. This approach enables you to take advantage of industry innovations while reducing the risk of machine learning-system obsolescence and avoiding the cost of custom integrations.

3. Understand that EHRs often don?t capture outcomes of care or treatment goals (i.e. context) in a standardized format ? and solve for it. Without standardized endpoints, it?s impossible to ?train? machine-learning algorithms to sufficiently explain the variability in outcomes that can be translated into better tailoring of diagnostic or treatment processes.