SETTING THE BENCHMARK

Selecting a Connectivity Architecture for Medical Devices

by Ralph Peterson / July 11, 2022

As medical device companies set their sights on staying competitive in a post-COVID world, developing connected devices stands out as a leading strategy. A recent Deloitte Insights survey identified technological advances as a top challenge, even when compared to regulatory activity and changing consumer attitudes. Acquiring the talent and supply chain necessary to add capabilities to devices can be difficult for MedTech companies.

While significant advances in medical technology subsectors like in-vitro diagnostics and robotics have great potential, simplifying the collection, transfer, and utilization of data from existing devices is just as promising for moving the needle on patient outcomes. More efficient data use can also reduce administrative costs and improve business decisions. The same Deloitte insights report notes that digital investments by medical organizations primarily aim to reduce costs and gain insights into the execution of business strategies.

Connected medical devices are at the forefront of better use of data in medicine. Selecting the best connectivity architecture for a device is critical to achieving success in the eyes of patients, care providers, and shareholders. This blog will explore the primary types of connectivity architectures and selection considerations for specific use cases like patient compliance, point of care enablement, and clinical trial execution.

Device-to-Gateway Options

The device to gateway connectivity model can be one of the most cost-effective solutions for adding connectivity to an existing or new medical device. A device can connect to a gateway via wireless technology that enables patient monitoring data to be transferred a relatively short distance from the device to a type of computer called a gateway. The "gateway" refers to a centralized data aggregation device like a router or phone that connects over a high bandwidth connection to send data to the cloud.

Device-to-Gateway: Bluetooth Low Energy

One of the most popular types of connectivity technology to connect the device to the gateway is a subset of Bluetooth radio known as Bluetooth Low Energy (BLE). BLE offers low data transmission rates (up to two megabits per second) and low power usage. Overall, Bluetooth radio can struggle to maintain reliability depending on the application: has your phone ever failed to connect to your car, or has your smartwatch ever failed to connect to a faraway phone? If so, you can imagine why Bluetooth connectivity may not be suitable for life-critical data.

Bluetooth Low Energy devices usually have much better battery life than devices connected by other means, thus ideal for battery-constrained devices that require mobility. For this reason, Bluetooth connectivity is usually recommended for activities like long-term biomonitoring that will not be hindered by intermediate lapses in connection.

For example, several home blood pressure monitoring devices leverage Bluetooth technology for the intermediate connection between the blood pressure monitoring system and the gateway, transferring patient data over a secure and easy-to-use cellular connection for provider review and action. Collecting early-warning signs of a heart attack is another excellent example.

Device-to-Gateway: Wi-Fi

Another connectivity technology that fits in the paradigm of device-to-gateway technologies would be Wi-Fi. Wi-Fi does away with some of the concerns surrounding Bluetooth as it can transmit large amounts of data with minimal variable costs. Wi-Fi connectivity has much greater reliability than a Bluetooth connection, but not more than a wired or cellular connection. Many patient-critical Wi-Fi connections powered by ethernet use cellular as a backhaul option.

There are downsides to choosing Wi-Fi connectivity for some medical device use cases. First, high battery consumption makes Wi-Fi connectivity unsuitable for mobile medical devices that cannot be recharged frequently.

Secondly, Wi-Fi connectivity suffers from usability concerns. Maybe you've suffered through the arduous process of exchanging Wi-Fi passwords while trying to connect to your in-laws' wireless network. In that case, you can imagine how uncomfortable the experience might be for a patient or caregiver attempting to communicate with an over-stretched hospital IT organization.

Wi-Fi networks may also be considered less secure depending on the quality of the network architecture and surrounding security apparatus.

Device-to-Gateway: Ultra-wideband

While Bluetooth and Wi-Fi may be ubiquitous in consumer electronics, the benefits of Ultra-wideband (UWB) connectivity are less widely known. UWB technology offers many advantages in medical applications, including obstacle penetration, high precision ranging at the centimeter level, low electromagnetic radiation, and low processing energy consumption. Certain medical applications can benefit from this technology, particularly those that require an automated, hassle-free user onboarding experience where users will remain in a given building or campus.

A hypothetical connected dialysis machine may be one such example; UWB allows transmission of data while also being able to identify the precise location of the machine within a large hospital complex. The verified location makes the device easier to locate for use and maintenance. By using UWB in collaboration with other IoT technologies like Bluetooth, a diverse set of features emerges, including verified user access control. Patients or providers can control the use of the device using UWB connectivity to validate the user's intent by confirming the medical device's location.

Device-to-Cloud Options

Device to cloud technology skips the intermediate-step data transmission from device to gateway and transfers data directly from the end-user device to the cloud. As mentioned previously, the main reasons for gateway use in connectivity architecture design are usually battery life or connectivity cost concerns. When these two issues are not as important, a device-to-cloud design can offer a more user-friendly experience in exchange for an increased bill of materials and service costs.

Device-to-Cloud: Real-time cellular

With the advent of 5G cellular technology, real-time cellular offers the most hypothetically robust and reliable connectivity option. The low latency of this medium makes such a technology appropriate for high-speed, high data information transfer like that which might happen during your remote surgery operation. Such technology could allow some of the world's best doctors to operate on the patients who need it most without leaving the comfort of their own offices. In exchange for ultrafast connectivity, battery life is a critical concern. Accommodating sufficient battery storage may necessitate a large device form factor, wiring the device to a power source, or frequent recharging.

One additional cost of operation of these networks is the setup and maintenance of cellular service subscriptions. Just like your personal phone must be connected to one of the networks (Verizon, Vodafone, etc.), so must your cellular-connected medical devices. If you've ever been in a car and noticed your friend on another cellular network has service while you do not, you can imagine what havoc that could play on mission-critical mobile medical devices. Various workarounds can be used, including working with mobile virtual network operators (MVNOs) that offer one connectivity plan with access to a global collection of networks through a single card-type or embedded-type SIM.

Device-to-Cloud: Low-power, wide-area technology

Cellular technology can still be leveraged if user-friendliness and reliability of data transfer are critical, but the speed of data transfer is not. The advent of low power wide area technologies like narrowband IoT cellular networks have promising implications for "store and forward" applications of medical device technology. In this architecture, patient monitoring data can be collected every few hours or months and then transferred to an analytics platform or provider when appropriate. This model is not useful for up-to-the-minute data information but can be used to monitor less time-critical patient data like the prevalence of heart murmurs. Such low-cost, low-power connectivity use cases have given rise to 'flat-rate' prepaid SIMs that only cost a few dollars and should have enough data allowance to last the device's entire life.

Ensure an appropriate and adaptive connectivity strategy

No matter which of the many connectivity mediums you choose, one thing can be sure: the selection process is critical, and nobody can predict the future. Establishing a third party as an expert in connectivity technology can break product development deadlock.

Organizations must also leverage some form of adaptability into their connected device model to respond to future changes in consumer preferences, regulatory compliance, or differences in their own internal needs. These factors influence the demand for thoughtful connected device design and strategic mitigation of manufacturing cost and setup risk.

Benchmark helps medical device companies select and implement the optimal connectivity type for their device. Our team has expertise in IoT security, firmware, battery life optimization, FCC certifications, and a wide range of additional skills to give connected medical devices market staying power. Benchmark's medical device engineering, manufacturing, and lifecycle management services combine an unparalleled 40+ year legacy of experience with a global footprint built to propel disruptive medical companies and their devices to success. Learn more here.

Medical Technologies Connected Devices Engineering

about the author

Ralph Peterson

Ralph Peterson is a Lead Engineer at Benchmark focusing on connected medical devices. He led the team that developed Benchmark's biomedical monitoring patch platform, working with customers to optimize the performance of all aspects of a connected device. Prior to Benchmark, Ralph was a Quality Engineer at Boston Scientific and a Postdoctoral Researcher at InCube Labs. He earned his Doctorate and Bachelor's degrees in Biomedical Engineering from The University of Texas at San Antonio. Ralph served eight years as a Qualified Nuclear Reactor Operator for the U.S. Navy.