How to Create Value From Health Data

Scott Plewes | October 10, 2019 | 5 Min Read

In this article, we’ll dive into each of these common ways of deriving value from your collected health data to give you a better idea of how you can apply them within your own company, benefiting both the business and your customers.

So you’ve done the hard work to enable your medical device to transfer the data it’s collecting from the device, into the cloud so that you can do “something” with that data. The next challenge to arise is often understanding what to do with all of the data you’re collecting. Typically, we see two primary functions for this data that sum up most use cases. The first is to leverage the data you’ve collected to create a new feature for your product, and the second is to use that data for sales and marketing to increase product sales.

Translating Data Into A Product Feature

Here we’re talking about translating the data your device is collecting (or could collect) into the creation of a tool. This is often where the term Artificial Intelligence or Machine Learning gets invoked. With a system like a hospital, it’s pragmatic to think of machine learning as something that will do a “thinking” task either more efficiently, faster, or with less error than a person. It’s basically the “thinking” equivalent of a robot on an assembly line who can do a physical task faster or better than a human. It won’t replace a role (not for many years, at least) unless the role comprises of a single task. For example, machine learning won’t replace radiologists any time soon, but it may be able to replace certain tasks they do, such as identifying a broken leg. Machine Learning will not eliminate the role of a radiologist, but it will allow them to be more effective in other aspects.

Leveraging The Data Your Medical Device Is Collecting

To take advantage of data science, your hardware product has to become a digital product as well. This needs to be designed just like the hardware – to a purpose. Part of that purpose could be to gather data from the “ecosystem” the product is in contact with, including the people (patients, clinicians, other roles) and the technology.

To leverage the data your device is collecting within this ecosystem and translate it into a usable feature for end-users, you need to establish potential use cases for where the data would be highly valued. The simplest way to do this is to interview people in the ecosystem and uncover where and how they might use this data collected by your device.

Typical examples of where value might be found from collected device data include:

There’s virtually an endless number of potential use cases, but they tend to fall into three categories:

  • Real-time data to help make decisions
  • Longitudinal data (data collected over weeks, months, even years)
  • Data that can be leveraged for an “AI/ML tool”

The third is really a special case of the first two. We’ve separated it because often it’s much more opaque as to what is happening with the model and/or analysis.

The first use case would be something like an alarm generated when data exceeds a certain threshold (I.e. an outlier is detected) for a patient’s vital signs. While this type of scenario has existed for a long time, data science can bring much more context-specific and detailed analysis to understand when certain thresholds might actually be an issue.

The second use case could be leveraging the information used by the device manufacturer (usage or performance) or data used by the hospital to track whether there is a relationship between medication administered and procedure, or timing in the procedure and patient outcomes.

The third scenario is applying analysis to large data sets or a large number of variables that are so complex that describing these relationships is essentially impenetrable to users or of no practical value. However, while understanding what the model does may be of no value, the result (ie. detecting that a nerve may be at risk of being damaged) could be of high value.

Leverage Data To Sell More Product

You’ve probably done this in some form in the past and have referred to it as business intelligence – using data (with math and computers) to describe, gain insight, or even predict what will or might happen in the business. What’s different today is the amount of data available, sophistication and availability of algorithms and computing power, as well as the vast array of domains it’s being applied to. There are simply many more possibilities of how to get data, mathematically model it, and make predictions.

If you are using math, computing and associated modeling to understand when a patient’s vitals are indicating possible issues, or when different practices lead to better outcomes, then you are in this space. What is different in data science is this is giving you insights that could not be done by a domain expert alone, and/or simple descriptive statistical techniques, and/or without large amounts of data and computers. This data can be leveraged by both the product design and development team to improve product performance, and by the sales and marketing team to demonstrate outcomes.

For example, sensory data collected from your ultrasound machine indicates that when sonographers hold the device at a certain angle, they produce clearer images. This ergonomic intel is used to redesign the ultrasound handle to naturally reproduce this angle and include more comfortable padding for the hand. The team then could patent the design of the handle and leverage that as a market differentiator.

With both of these approaches to creating value from your collected data, your team is engaging in a discovery process to specify the problem or opportunity at hand sufficiently enough to capitalize on it. The key concept that many tend to overlook is that data science is a core capability that can be added to your medical device or leveraged by your team to provide customers with additional value – value of which they’ll be willing to pay for. That is if you can execute your data science initiatives successfully and avoid common data science pitfalls product teams often encounter.

Download: Applications of Data Science in Medical Devices

In this eBook we demystify the domain of data science, articulate what’s involved, dive into the two main approaches to applying data science, and review the most common pitfalls experienced by teams that lead to failed data science initiatives.

Suggested Stories

Health Information System Integration

In this webinar, we discuss interoperability in healthcare and answer attendee questions on Health Information System integration. Download the webinar Now.

Read More

Achieving Interoperability in your Healthcare Organization

The challenge with integrating healthcare organizations is that they are just that – separate organizations. Each organization provides a healthcare service, but they provide that service using their own process.

Read More

Using Contextual Data in Connected Device Design

With more physical things being connected to the internet, the amount of data collected by these devices continues to grow. When designing an experience for an IoT device, considering context becomes increasingly important to avoid agitating users.

Read More