Designing Wearable Medical Electronics for Maximal Impact: Breaking the Physical Sensing Barrier

Joshua R.Windmiller, PhD, MSc, Founder and CTO, Biolinq Inc.

Joshua R.Windmiller, PhD, MSc, Founder and CTO, Biolinq Inc.

Whereas infectious disease plagued society as the leading healthcare epidemic in the 20th century, chronic disease will come to define healthcare in the 21st century. Indeed, seven of the ten leading causes of death in the past few years were attributed to chronic disease, according to the US Centers for Disease Control and Prevention. In fact, 86 percent of the nation’s $2.7 trillion annual healthcare expenditure is utilized to care for the sequelae of chronic disease. These sobering statistics underscore the magnitude of the challenge confronting healthcare systems of developed and many developing nations.

"A new breed of biosenor has begun to emerge, one that is positioned to quantify circulating levels of various metabolites and electrolytes in a very minimally-invasive fashion"

Despite these alarming trends, a very overt disconnect has materialized in healthcare today. On the one hand, whereas the current cadre of wearable devices, including fitness trackers and smartwatches, are quite accessible to the average individual, the information content that can be gleaned from such devices is extremely limited and very seldom canbe leveraged to attain clinical endpoints. On the other hand, conventional centralized, laboratory-based diagnostic tests often ordered by a physician and performed by a trained operator provide a very clinically-relevant assessment of an individual’s health status but remain far from accessible to the individual at their own volition. The truth of the matter is that the sensors buried in conventional wearable devices are limited in scope and capability to measuring a few basic vital signs (via electrophysiological measurements) or the wearer’s motion (kinesthetics). Not surprisingly, vital signs constitute a small, though not inconsequential, perspective of one’s metabolic health. In fact, over 70 percent of the data required for accurate diagnoses or monitoring can only be found in one’s biochemistry.

Recent aims are directed at augmenting conventional electrophysiological and kinesthetic sensors with biosensors capable of providing the user with added dimensions of rich chemical information that can be exploited to understand the causality linking one’s daily lifestyle choices–diet, exercise, sleep patterns, stress management–and their metabolic health. Indeed, it is insufficient to provide a mere snapshot such as a single point-in-time blood panel–one must tender this information continuously, in real-time, with high contextual relevance so that corrective action can be taken prior to embarking on unhealthy behaviors that might be an underlying contributor to chronic disease.

In order to address these aims, a new breed of biosenor has begun to emerge, one that is positioned to quantify circulating levels of various metabolites and electrolytes in a very minimally-invasive fashion. Fundamentally, the key difference between biosensors and the physical sensors in which most are familiar reside in the incorporation of a biorecognition element within the biosensor that selectively undergoes a physical, electrical, optical, or chemical alteration upon exposure to a biomarker of interest most often circulating in a physiological fluid.

Such biosensors embody substantial potential for integration into conventional wearable form-factors as well as skin-adhesive patches. Despite this potential, it is imperative that the supporting electronic backbone configured for the control and interpretation of these sensors be architected with some unique constraints in mind:

(1) Sensitivity: Signal levels originating from biochemical reactions are often exceedingly weak. It is not out of the ordinary for electronics incorporated at the sensing front end to be capable of transducing current levels at the picoampere level and, in some cases, with femtoampere-level resolution.

(2) Signal-to-noise: Sources of noise may be endogenous to the sensing medium (co-circulating biochemicals occupying the physiological fluid in which the sensor exhibits residual selectivity) or exogenous, arising from the wearer’s motion, autonomic response, accumulation of static charge, or even ingress from AC mains power in the wearer’s vicinity. Effective PCB design for EMI rejection, including the implementation of guard rings and isolated grounding planes separating the analog and digital components, is mandatory, especially as current levels descend into the nanoampere range.

(3) Stability: Owing to the limited lifetime of the biorecognition element, any design must incorporate a means of compensating for a decaying or drifting signal. This is often achieved in embedded firmware by means of convolution of time-series data with the biosensor’s steady-state temporal response.

(4) Power: Current user-centric requirements for low-profile wearables impose great demands on battery capacity; batteries often require the majority of the volume in a wearable device today. Sleep or hibernate modes must be invoked for the majority of an MCU’s and/or wireless transceiver’s duty cycle, taking great care to limit the frequency of RF transmission and analog-to-digital conversion, which often take the most significant toll on battery life.

(5) Miniaturization: As mentioned above, low-profile wearables are all the rage today; this trend will likely continue into the foreseeable future. Minimizing the number of discrete components and integrating discrete units of silicon, either by means of an ASIC or chip stacking packaging methods, can markedly condense a device’s footprint. Here, a minimalist design philosophy is crucial.

(6) Compliance with Established Regulatory Standards: Although this one might seem obvious, it is surprising how often medical electronic designs are pursued with little or no regard to compliance with IEC 60601 or other standards imposed by regulatory bodies in various jurisdictions. A design philosophy that begins with comprehensive knowledge of the appropriate regulatory standards is essential to safe and reliable device operation and, ultimately, timely regulatory approval.

In the coming years we will begin to observe the emergence of this new archetype of wearable sensor, one which mandates a very unique electronics design paradigm to address the performance requirements of these novel devices. With some ingenuity and creativity, this new generation of biosensors will begin to proliferate, increasingly finding home in our favorite wearable platforms, and will thus empower the end-user to take the appropriate steps to maintain a healthy lifestyle and manage chronic disease or potentially avert it entirely.