Wearables: Communications, Miniaturization Challenges Abound

For wearable medical diagnostic devices to transition from the lab to the patient, engineers must overcome several distinct technological hurdles. First and foremost, they must integrate wireless communications and create miniaturized and flexible substrates.

At the MEDevice San Diego conference on Wednesday, September 10 from 4:30-5:00 p.m., Fred Beyette, professor of electrical and computer engineering at the University of Cincinnati, will take up these challenges in a presentation titled "Disappearing Noninvasive Medical Electronics--The Challenges/Benefits of Wireless Diagnostics No Larger than a Band-Aid." In the following Q&A Beyette provides a preview of the themes he will take up in his talk.

MPMN: Please describe your work with noninvasive electronic technologies for wireless medical diagnostic devices.

Beyette: The work that we have been focusing on is how to make medical devices that can be worn like a Band-Aid and can noninvasively extract information from the subject about their health status. We have looked at a range of obvious technologies, such as temperature sensors and skin conductance devices. We have also investigated technologies that have been around for quite a while, such as devices for performing pulse oximetry.

However, our objective is to extend beyond what has already been investigated to look at the kinds of devices that can be developed for detecting biomarkers from such body fluids as sweat. It turns out that many of the biomarkers that clinicians gain access to through a blood draw can actually be quantified in sweat itself. Thus, we are looking at ways of miniaturizing sensors to detect specific biomarkers in sweat that can provide information about the health status of the patient.

The devices we are trying to develop will 'disappear.' In other words, we seek to shift from bulky objects such as handheld devices that are pressed against the skin to devices that are miniaturized to the point that the form factor, weight, and energy requirements become so minimal that they are like wearing a bandage. Thus, although such devices won't disappear physically, they will disappear from the patient's consciousness.

MPMN: What about the electronic components contained in such devices--how are they being integrated into the device or the body?

Beyette: There are three groups of electronic functionality. The first is the sensor. Sensors are specially designed to be reactive to the specific biomarkers of interest, essentially producing an analog signal response based on the concentration of the given biomarker. The sensor output then passes into onboard signal processing data-acquisition circuitry. Second, we need a microcontroller, that captures the sensor data and provides filtering and other low-level processing of the data. Third, we have to be able to transmit the initially processed data wirelessly to some type of electronic device, such as a smartphone or laptop. All of these electronic components are being miniaturized and integrated into flexible substrates that allow them to conform to the body. And they look very much like Band-Aids.

MPMN: Could you go into the wireless aspect of such diagnostic devices?

Beyette: From the wireless standpoint, we've been exploring two different pathways. The first and perhaps the simplest approach is to use a very simple communication system, such as RFID. Such systems consist of a simple reader that interrogates the patch electronics and takes a very small payload for communicating information back. In a more sophisticated way, we have been turning to Bluetooth communication, the benefit of which is that it provides a much larger information packet to transmit information. It can also perform such tasks as encryption so that we can deal with such regulatory issues as protecting patient information.

MPMN: What avenues are you exploring to achieve medical device miniaturization and flexibility?

Beyette: We are investigating two different paths for achieving device miniaturization. The first involves the use of commercial, off-the-shelf chips. Either we will customize them at the level of the flexible circuit board or we will use bare dies and flip-mount them directly onto the substrate. And while it is also possible to design custom chips, we have not found it necessary at this point to go down that path. By using commercially available microcontrollers and analog components, we're still able to get all the functionality we need while maintaining a small form factor and the appropriate level of flexibility.

MPMN: How advanced are the technologies that your laboratory is working to design, and what are the challenges facing the development and integration of miniaturized wireless wearables?

Beyette: We're at a stage of development at which we're able to make some very simple sensors that can detect sweat and analyze such constituent parts as potassium levels, which can then be used as an indicator of hydration. We have developed prototype patches that can detect a single biomarker and perform RIFD-type communication.

From those simple prototypes, we're in the process of going in two directions. The first is to put multiple sensors onto a single patch in order to look at multiple biomarkers or to put different kinds of sensors on the patch so that we can analyze skin conductance, a biomarker concentration such as potassium, and perhaps heart rate simultaneously. The main objective of this technology is to perform physiological monitoring for physically active people such as firefighters or construction workers or to monitor athletes that need to be conscious of electrolyte replacement. The other area into which we are expanding our efforts is transitioning from simple RFID-type devices to more of a Bluetooth communications platform so that we can ensure patient information security.

As we move into more-complex patches, the types of challenges we face are those that one would typically expect in any type of complex electronic system design. For example, we are dealing with real estate and how much area is required. In our case, this issue is compounded by that fact that the larger the area, the larger the patch must be and the more difficult it is to make the patch conformal to the shape of the body part to which it will be applied. That's the real-estate problem that all electronic devices have--becoming more complex while maintaining a small footprint.

We also face a problem that other electronic systems face: When shifting over to Bluetooth communications, we start to have larger power requirements. With larger power requirements, it becomes much more of a challenge to ensure that we will maintain sufficient battery power and can communicate data wirelessly through a more-complex communication protocol. And with shrinking real estate, miniaturized or flexible batteries become essential. We are looking at the use of polymer batteries that are essentially that same size as our Band-Aid. However, if we use such a large battery, it must be flexible enough to conform to the human body shape.

Another challenge we face is how to charge the battery. If you want to develop a technology that's going to disappear from patients' consciousness, they must be able to wear it for days at a time without having to think about plugging it in and recharging it. Therefore, we are starting to look at energy harvesting--how to collect energy from the world around is to charge the battery. Technologies that solve this problem include solar cells and radio-frequency collectors that can hack into Wi-Fi energy and convert it into electronic energy. For example, we have been looking at flexible solar cells and top layers for our patch in order to charge the flexible battery.

The bottom line is that all such technologies must be conformal to the human body. But creating diagnostic devices that will continue to operate over days while being so lightweight, flexible, and wearable that patients will forget about them raises a whole bunch of interesting engineering problems.

Bob Michaels is senior technical editor at UBM Canon.

[email protected]