The Future of Connected Medical Devices: From All-In-One Devices to Distributed Sensor Ecosystems
TL;DR
- Connected medical devices are moving beyond single-device limitations.
- Distributed sensor ecosystems enable richer, patient-specific insights.
- IoT protocols (MQTT, LoRaWAN) make multi-device data sharing possible.
- Secure development and offline safety are critical for adoption.
- Scalable ecosystems reduce re-spins and create new data value streams.
Many of the medical devices created at Starfish take advantage of sensors to convert the real world into digital data that can be understood by computers. AI data analysis continues to grow, the amount of data that a computer can understand has increased astronomically; however, the data collection capacity of any one device has remained more constant.
For example, a CPAP machine may have a set of sensors but is only used at night while one is sleeping. A smart watch can only capture so much data on its sensor even if it’s worn all day. A bathroom scale for example, only tracks weight when someone stands on it.
In a world of connectivity, one could imagine an eco-system of connected sensors that all send data to one centralized location, such as a cloud service on the Internet. Once there, the multi-device data could be reviewed by algorithms that analyse the information to provide alerts and even perform remote diagnosis, providing data-driven care using more data than could be collected by a single device.
From Single-Device Thinking to Connected Ecosystems
Medical devices are becoming increasingly connected but are still limited by “single device” thinking. A medical device might have a large set of sensors, but they are all physically connected to the device that collects the data. This implies the device must be with the patient, or the patient must be with the device.
Much like the seamless vision of the “smart home”, the future of health monitoring, analysis and diagnosis can quickly become decentralized from a single medical device moving to a disperse data set of physically sparse, edge-based micro-devices, housing a variety of sensors. Using specific and disperse targeted sensors also personalizes care and empowers patients to track data specific to them ultimately reducing patient burdens and enhancing quality of life.
Using a common Internet of Things protocol such as MQTT (Message Queuing Telemetry Transport) or LoRaWAN (Long Range Wide Area Network) one could conceive of a network of sensors that connect directly to the cloud, or through a gateway device like a hardware hub or a smart phone to send the data. In some cases, you could see the need for sensors to react to other sensor data. This implies a common protocol between sensors, along with super scalable publisher/subscriber design patterns to empower the entire eco-system and enable seamless integration with 1st or 3rd party sensors. Further to this, it allows the cloud infrastructure to collect and store all this data, as well as publish commands, analysis or diagnosis for display or action.
Case Study: Solving Sleep with Connected Sensors
Sleep is a good example of how connected sensors can work together. Understanding the patient is only one part of the challenge, understanding the environment is an entirely separate problem. An example of a fully integrated constellation of medical sensors to tackle sleep problems could consist of a smart watch monitoring heart rate and blood oxygen levels, along with a smart sleeping pad that measures patient movement. This would arguably indicate if a patient were sleeping or having trouble sleeping, but this could also be combined with a light sensor for ambient light, a microphone for ambient noise, and a CO2 sensor for air quality in the room. Should the sleeping pad register movement, the smart watch could increase the sampling rate of the heart and blood oxygen while the patient is presumably tossing and turning, sensors could react to inputs from other sensors, thus increasing the resolution of the data to be analyzed. As the light and noise increase in the bedroom, all the interconnected sensors can prepare to capture the wake-up moment in high data resolution. The more sensors added to the eco-system, the more data-driven care the patient will get. Combined with all the data in the cloud from other patients, once the product becomes commonplace, any AI engines can ultimately provide evidence-based insights to the patient about their sleep.
Since the sensors are not connected to any one device, they can each use different methods to connect to the Internet, ranging from common cellular communications for devices users take with them, or take advantage of wi-fi in the home if a device is meant to stay in one location. Furthermore, the lack of wires to a master device reduces patient risk and minimizes complications while tossing and turning. Light-weight sensors reduce power consumption of the device, providing extended continuous monitoring.
Overcoming Connectivity and Cybersecurity Challenges
Building connectivity into medical devices rightly brings extra scrutiny from the FDA to ensure those devices are secure from cyber-attack as well as managing offline risk. However, the technology exists to create secure environments for data to flow over the Internet and back again. Starfish employes a Secure Product Development Framework that guides our development team on secure coding practices; follows up with Cyber Security analysis through the creation of Data Flow Diagrams; conducts STRIDE (Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, and Elevation of Privilege) analysis; creates and monitors Software Bills of Materials (SBOMs); and works with penetration vendors to ensure secure cloud solutions.
In addition to online cyber security risks, offline risk analysis is done to ensure patient safety during connectivity outages. Analysing, and planning for the impact of severed connectivity should be designed from the start with a focus on safety of the patient and functionality of the devices. No one would want the decision to deploy airbags in their car to be made by some remote server interpreting data collected from the vehicle; that system would break down if connectivity was lost. For an at-home device, connectivity could be broken by a power outage to a wi-fi router, or a cellular tower. Mission critical services that keep the patient safe need to be handled without the cloud service.
Scalability, Safety, and Value for Manufacturers
The extra effort required to develop highly connected devices pays off in scalability. Once an ecosystem is in place, it can grow with minimal re-spin of hardware, firmware and cloud software. If you already have your smart watch and sleep pad, adding a temperature sensor to the patient would be minimal work and not require touching the other hardware and firmware on the other sensors. Adding a new edge-sensor and then revalidating the cloud software to take advantage of this new sensor would reduce time to market and enable customers to upgrade by simply buying a new sensor instead of replacing the entire system.
Now What? The Road to Fully Connected Medical Devices
Beyond connectivity itself, manufacturers gain another layer of value through the sheer volume of data that can be collected. The medical device cloud component can aggregate this data to be data mined for either improving the sensors, comparing sleeping issues to other patients, or the data can be sold for additional data streams, such as how many people sleep well versus not, or what light level is perfect for the best sleep, and so on. This enables multiple data streams for the manufacturer of the sensor.
Taken together, these insights point toward the next stage of innovation — a world where connected medical sensors evolve into full ecosystems powered by cloud computing. This evolution represents the natural next step beyond single-device thinking, moving toward networks of devices that share data, adapt in real time, and deliver richer clinical insight. As more medical devices become part of the Internet of Things, the takeaway is clear: building truly connected ecosystems requires investment in secure, scalable infrastructure, but the payoff is profound for patients, clinicians, and manufacturers. The frontier ahead is not just smarter devices, but entire constellations of devices that learn, adapt, and act together.
Sean Daniel is the Starfish Software Manager. He entered the medical device field in 2024, bringing with him over 12 years of experience managing software teams—10 of those in electromechanical devices. Over his 25-year career, Sean has led software development efforts ranging from Fortune 50 companies to start-ups. He brings a sharp focus on safety-critical processes, combined with the agility to move quickly while mitigating risks in product development roadmaps.
Image: ChatGPT
Related Resources
Nick Allan and Nigel Syrotuck share their end-of-summer reading list, featuring FDA regulatory books and PCR memoirs. From navigating regulatory hurdles to celebrating groundbreaking discoveries, their choices show how science reading can be both educational and entertaining.
Project managers are on the front lines of rising complexity in medical device development. They sit at the intersection of vendor timelines, regulatory constraints, and engineering realities.
While medical devices often dominate development conversations, the way drugs are delivered across regions can dramatically change how treatments succeed — or fail.
As a software engineer with experience in both web development and medical system software engineering, I’ve worked on projects ranging from consumer-facing web applications to medical device graphical user interfaces (GUIs).