From defibrillators and ventilators to infusion pumps and dialysis equipment, Internet of Things (IoT) medical devices significantly enhance outpatient care. Patients can record and send measurements remotely, eliminating the need for follow-up visits. At the same time, these devices help understaffed outpatient facilities that might not have the capacity to send caregivers for home visits. However, to ensure the optimal functionality of these IoT outpatient devices, original equipment manufacturers (OEMs) must address several considerations.

Data aggregation

IoT devices in outpatient settings may capture massive amounts of data; nevertheless, it would be inefficient to send all of this data over a cellular modem due to bandwidth and power restraints, e.g., in battery-powered devices. Likewise, data plans often are not cheap, making sending huge amounts of data economically unviable. As such, it is important to send meaningful data to save power and increase the runtime of battery-powered devices and minimize the cost of connectivity.

New artificial intelligence (AI) and machine learning (ML) technologies can help with data interpretation and aggregation and help IoT devices save power and minimize the cost of connectivity. These AI and ML models can analyze patient data to discover patterns, detect anomalies, and filter out errors to send only the meaningful data to the health care provider or medical OEMs.

Edge processing

There is also the challenge of maintaining service amidst connectivity issues. When sending patient data over the network to a cloud service for processing, there is the risk of the network going down, preventing the device from performing its necessary functions or even interrupting a treatment. In medical settings, it is critical to have a device’s core functionalities detached from a network to safeguard against such disruptions, which is a key reason why many OEMs enable their devices to process data at the edge.

Processing at the edge enables the IoT device to continue operating even if the cellular network connection is disrupted or the patient’s home Wi-Fi is unavailable, allowing it to continue to provide feedback on the treatment or send alerts. This ability to uphold data processing even if a connection is temporarily severed is critical in medical settings where the patient’s well-being is on the line. Not only does edge processing permit a device to maintain its core functions amid connectivity issues, but it also helps it operate faster and more reliably because it does not have to send data over the network, process it in the cloud, and wait for a response.

While edge processing has its advantages, one should not completely dismiss the power of the cloud. A mix of both cloud and edge processing in medical IoT device use cases is valuable. For example, the cloud can perform deeper analytics of outpatient data than the edge; likewise, the cloud can effectively handle multiple devices in a large area for outpatient facilities or provide greater analytics of a bigger dataset collected by many devices in the field.

Device security

IoT devices in outpatient settings capture sensitive patient information and must securely store, process, and transmit this data. There are many regulations OEMs must abide by, with new legislation covering cybersecurity requirements getting passed regularly. One of the key requirements is to provide the capability to address new upcoming security vulnerabilities in connected devices. OEMs need to provide a process to detect new security vulnerabilities affecting their devices, integrate fixes for these, and perform regular remote firmware updates; in other words, OEMs must continuously and remotely update the security of their devices in the field to protect the patient data, keep them operating, and performing as expected while protecting the patients’ data.

Unfortunately, many OEMs have not focused on the security of devices until recently, when cybersecurity legislation was put in place that mandates the implementation of certain security features to protect devices. These OEMs now often struggle to achieve such capabilities because they do not have the resources or expertise. Partnering with a provider with security tools and features built into their products can reduce the risk, and medical OEMs can secure their devices through remote, automatic security and firmware updates. An ideal partner will also maintain security over the entire lifecycle of their products due to the ever-evolving nature of the threat landscape.

Reliability and longevity

Reliability is key for medical IoT devices, and predictive maintenance can ensure maximum reliability. When a device is connected, OEMs have the ability to monitor it to identify issues or anomalies like vibration, high temperatures, moisture, etc. With this insight, an OEM can be proactive rather than reactive and send a service technician to fix the device or preemptively deliver a replacement before the device breaks down and becomes unusable.

Device longevity is equally important as OEMs deal with long development and certification cycles before getting their devices to market. It is not an exception that a device takes more than five years to complete the development, testing, and approval process before it can be sold and used by patients. The embedded systems OEMs integrate into their medical products need to be designed for long lifecycles. Otherwise, a medical OEM cannot sell and support their product for long enough to turn a profit. To that end, OEMs should partner with embedded systems providers offering devices and modules supported for 10-plus year lifetimes, that have a track record in providing longevity to their customers. Even better — embedded solutions that are supported by developer tools and also integrate remote monitoring and management capabilities can give OEMs an enormous boost in their go-to-market strategy and ability to provide ongoing management services.

Complete vs. piecemeal solution

These considerations are vital to ensure the optimal functionality of IoT devices in outpatient use cases. Nevertheless, it is recommended that medical OEMs avoid cobbling these various security and connectivity features and components piecemeal from different vendors to avoid interoperability challenges, integration issues, and project delays. Given the challenges medical device manufacturers face in a connected world, it is better to prioritize IoT providers that offer complete solutions. While this might appear more expensive initially, the faster time to market and the lower total cost of ownership throughout the lifecycle of a product typically makes this the more cost-effective approach.

Andreas Burghart is a health care executive.