The traditional physical solution connecting the last mile is copper twisted-pair wire, a 100 year old technology buried underground or suspended from telegraph poles, but new technologies have built on this physical channel to deliver new capabilities. First came dial-up modems, allowing point-to-point connection between computers and to bulletin boards, then IP connectivity revolutionised the network, followed by broadband modems and now the latest fibre-to-the-cabinet technology which brings tens of megabits of bidirectional capacity, enabling the information revolution to reach the home.
But a similar challenge is now to be found in what we call ‘the last inch’, the connection of devices with users over short distances. This is especially relevant in the medical space, where wireless devices are used to sense and monitor physiological parameters to help manage a medical condition, or perhaps to display and share information about a medical treatment via a mobile platform (such as a smartphone), and then to exchange this information with cloud-based software applications.
Just as with the twisted-pair cable in the landline, the basic building blocks enabling ‘last inch’ connection have been in place for some time, and telemedicine applications have been used for decades. Recent advances and other factors however, are enabling a burst of innovation in connected medical devices and apps, and by 2030 I believe that this innovation will both support patients and allow them to connect with medical professionals to enable better medical treatment.
Enablers to innovation
We can identify a number of enablers that, when taken together, allow innovation of connected medical applications that operate over the ‘last inch’. These include technological advances, regulatory change and emerging standards:
– Low power wireless connection to devices
Although the predominant Bluetooth short-range wireless connection standard has been around for many years, power consumption has long been recognised as a key limitation for wireless-connected measurement and monitoring devices. As a result, the Bluetooth standard has now been augmented to provide support for low power applications with Bluetooth Low Energy, also known as Bluetooth Smart, incorporated into the standard in 2010. As the chipsets mature, software manufacturers have extended Bluetooth Smart capability to iOS and Android devices, and by 2018 it is expected that 90% of the world’s smartphones will support Bluetooth.
In terms of improving power consumption, a two year life can now be provided by a 1000mAh battery (AAA cell) when used for a simple logging function, making sealed-for-life small connected devices now a practical proposition. For medical and healthcare applications, Bluetooth supports communications for functions such as:
- medical measurements such as temperature, blood glucose monitoring, blood pressure measurement, and heart rate measurement
- health and fitness measurements, running speed and cadence measurement
- location and navigation
- proximity alerting — when the distance between devices exceeds a set threshold.
In addition there are numerous low power sensing devices that underpin connected medical devices, providing the means of monitoring parameters such as temperature, blood pressure, motion, oxygen saturation and heart rate. There are also low-power actuators, pumps, valves and microelectromechanical sensors.
– The smartphone platform
With more than a billion smartphone units shipped in 2013, it seems a safe assumption that within a few years this technology will be readily available to most patients, although unsurprisingly, the uptake is strongest in the under-35s.
By the end of 2013, Android was the most popular smartphone operating system with an 81.9% market share, followed by iOS with 12.1%, Windows Phone with 3.6%, and BlackBerry with 1.8%.
The smartphone provides a familiar user interface, integrated sensors, wide area network connectivity, short-range wireless, ease of installation of new apps, and inherent extensible multifunction ability. It is small, compact and is generally close to the user, and so can provide the hub to connect a multiplicity of monitoring devices operating within the last inch of the patient.
– Electronic health monitoring standards
Recent announcements of commercially driven standards and platform developments in electronic health and fitness monitoring will enable the swifter development of connected healthcare applications, leading to new technologies which will eventually be used in regulated medical devices. Two of the most interesting are:
- Samsung Architecture Multimodal Interactions (SAMI), a data broker to enable wearable devices to upload information to the cloud, where developers can access the data and create new apps. SAMI is stated to be a secure, open, diverse data platform; its open application programming interfaces can collect data from many device types or online sources and deliver the data to any other device or app in real time. Samsung also recently announced the launch of Simband, an associated reference smart health device design to allow sophisticated data monitoring systems to be engineered quickly.
- Apple HealthKit, which offers capabilities to aggregate data from various health apps and commercial fitness devices to provide users with a clear and current overview in one place. Information from exercise apps, wearable technology, blood pressure and heart rate, weight and sleep patterns are aggregated to provide a single view, and each app can also take specific information from other apps to allow information integration. Apple has announced that it is working with an Electronic Patient Record (EPR) supplier on plans for transferring patient-held data from apps into the EPR and back again, so that patient record data could be available on the patient’s phone.
These recent announcements show the intent of the manufacturers — to use smartphone hardware, connectivity and interfaces to support the development of data discovery, data aggregation and usage for more sophisticated healthcare apps using all data sources available. The emergence of standards will allow an ecosystem of solutions, components and expertise to develop in parallel.
– Algorithmic extraction of information from signals and sensor fusion
An important requirement for any wireless medical sensing device is an ability to separate the information of interest from the signals measured using a medical signal processing algorithm. Often the measured signals are a large data set and the information extraction process involves filtering out noise and estimation of a small number of parameters of interest. There may be real-time requirements to accomplish this quickly and at a high update rate.
Algorithms could be hosted on a small connected device which can sense data over ‘the last inch’, potentially reducing the amount of data to be transferred from the connected device, in turn reducing both power and time demands. On the other hand, processing could be hosted on a smartphone or connected computing resource after the raw signals are transferred, giving access to more computational throughput and other resources. So there are more architectural choices about how to structure the system. As there may be many signals available, the tasks of discovering them, combining them and drawing valid conclusions (for example, to produce a useful prognostic) requires data fusion techniques. As an example, OBS Medical have developed their Visensia technology to automatically analyse and interpret the combined patterns of up to five vital signs — heart rate, temperature, blood pressure, respiration rate and blood oxygen — regardless of whether the data is collected continuously or periodically, and the results are then fused to generate a predictive, numerical index representing the patient’s ‘wellness’. In a clinical setting, this technology is reported as showing quicker responses to patient deterioration, and a reduction in unexpected deaths, overall leading to improved efficiency through better use of the information acquired.
– Regulatory relaxation
Regulators are increasing the freedom to innovate in this space, with recent FDA guidance stating that for certain classes of mobile app, discretion will be exercised when policing the development of new apps with a medical/healthcare function. For eligible apps, this will decrease the regulatory burden faced when turning a new idea into a marketable product, and will therefore encourage innovation. The FDA guidance includes apps that help patients/users self-manage their disease or condition without providing specific treatment suggestions. This includes devices which provide, for example: simple tools to organise and track health information; easy access to information related to health conditions or treatments; help in documenting, showing or communicating potential medical conditions to health care providers; a way to automate simple tasks for health care providers; or which enable patients or providers to interact with Personal Health Records (PHR) or Electronic Patient Record (EPR) systems.
Regulation will still be needed if patient safety could be affected by the app or device, so as a result, any mobile app which performs a medical device function (for the diagnosis, cure, mitigation, treatment, or prevention of a disease or condition) will be considered a medical device, and will be regulated as such throughout its development and marketing.
What could emerge?
If the enablers are in place to make connected medical devices and apps operating over ‘the last inch’ a fertile ground for innovation, we naturally start to wonder what might be over the horizon. So here are some thoughts on what we might expect to see emerging in the connected medical device space in some key areas:
– Drug delivery devices
By 2030, a predicted increase of 51% is expected in the number of people that have diabetes compared with 2011, according to the International Diabetes Federation Diabetes Atlas. By that time the artificial pancreas — a closed loop insulin delivery system with continuous glucose monitoring, such as that proposed by Medtronic — is expected to be accepted by regulators. Connected systems, exemplified by Cellnovo, will continue to be developed, and will provide an integrated system for management of the disease. Connected patch pumps will be increasingly used for diabetes and other conditions.
Inhaler devices and pen injectors will become more routinely equipped with sensors to record and communicate time and location of usage as well as effectiveness of usage, in order to support users in managing their condition.
– Smart device accessories
There are many devices on the market which have received approval for sale and for which there is an understandable reluctance to make modifications necessitating resubmission to the regulators. Adding an accessory to an existing device that doesn’t change the medical function can add new benefits which could help the user manage their treatment and provide choices about how they do so. For instance, a connected smart accessory for a medical device could work with a smartphone to provide the means to help the user locate the device, to prompt the user to use the device, to assist in usage, to log usage, and to provide instructions. The accessory can also be a gateway for the device manufacturer to keep in touch with, and better understand, its users.
– Self-management monitoring and feedback devices
In Team’s survey on healthcare in 2030, 55% of doctors and 32% of patients believed that patients would be taking more responsibility for monitoring their health in the future. Electronic devices will assist this development, and we are already seeing the emergence of fitness and health apps that are familiarising patients with the concept of monitoring via a smartphone or other device. For instance, myFitnessPal is already used to record calorie expenditure, as does the Nike FuelBand plus app, helping to encourage a more widespread culture of health consciousness.
Connected medical devices and apps will assist further, especially to measure progress, and to remind patients when and how to use infrequent high value/impact medication. This will allow treatment plan adherence to be recorded and improved. Data fusion technology will also be used to provide patients with suggested advisory actions based on combinations of disparate physiological and other measurements.
– Aggregation of information on patient groups for research
Data gathered automatically on patient management and usage will help assess compliance for clinical studies, allowing better assessment of the quality of the underlying data used. The gathering of sets of data from patients using a device or treatment, perhaps anonymised, will also allow learning from the measured interactions of patients and devices and allow new insights and improvements to be made.
Conclusion
In all these areas one thing is certain — there will be surprises. Some innovative new applications with connectivity over ‘the last inch’ will emerge and unexpectedly succeed whilst others will fail to become established, not least for cost/benefit reasons. New players will also enter the market and come through the learning curve of working within the medical product regulatory frameworks. As a result, it is unlikely that technology will be the constraining factor as we move towards 2030, given its rapid rate of development fuelled by commercial consumer technology investment, more likely it will be regulatory acceptance.
This article was taken from issue 7 of Insight magazine. Get your free copy of the latest issue here.