How to quickly develop medical-grade wearable devices for vital signs monitoring

If you are creating custom designs for health and fitness wearables that provide accurate measurement results, you may be very familiar with this phenomenon and Murphy’s law.

This article comes from Digikey, author Stephen Evanczuk

If you are creating custom designs for health and fitness wearables that provide accurate measurement results, you may be very familiar with this phenomenon and Murphy’s law.

The concept behind using photoplethysmography (PPG) or electrocardiogram (ECG) to measure vital signs is certainly well understood. The heart rate of ECG can be determined by measuring changes in peripheral blood volume or by monitoring the bioelectric activity produced by the myocardium, so as to find the heart rate of PPG. They used the difference in the absorption spectra of oxyhemoglobin and deoxyhemoglobin to understand the simple theory of estimating blood oxygen saturation (SpO2). Engineers are also becoming familiar with more complex measurement functions, such as using pulse propagation time (PTT) or pulse arrival time (PAT) to build cuffless blood pressure monitors.

These different measurements rely on signal chains such as amplifiers and filters to be adjusted and sent to the analog-to-digital converter (ADC). Using the converted data, the microcontroller (MCU) executes algorithms to generate heart rate, SpO2, and blood pressure. Equivalent.

Get clean biosignals

Developers can use a large number of low-power, high-precision devices to build customized signal chains and processing subsystems to differentiate health and fitness products. However, in most cases, it is difficult for specialized biosensors to meet the needs of biosensor signal chains.

Devices such as ADI’s MAX86140 and MAX86141 are specifically designed for the optical PPG method. For biopotential ECG measurement, ADI’s MAX30003, AD8232A and AD8233A can meet the required signal chain. ADPD4100 and ADPD4101 of ADI Company can support these two types of measurements. These multi-mode analog front ends (AFE) integrate a pair of multi-channel signal conditioning chains, including transimpedance amplifiers (TIA), band pass filters (BPF), integrators, and ADCs.

Developers can use this AFE as the basis for bioelectric-based single-lead ECG measurement (Figure 1, left) and optical-based PPG measurement (Figure 1, right), because it is very suitable for consumer wearable devices.

How to quickly develop medical-grade wearable devices for vital signs monitoring
Figure 1: ADI’s ADPD4100 and ADPD4101 AFE support PPG (left) and ECG (right) measurements. (Image source: Analog Devices)

These specialized biosensors help speed up development, but they do not save you from all the problems that may arise when dealing with biological systems. Unpredictable (but not accidental) artifacts such as transient environmental sources and skin unevenness can affect PPG, while a series of physiological electrical signal sources such as electromagnetic interference (EMI) and skeletal muscle contraction complicate ECG. (As I discovered during my thesis research, the impact of these conditions on the signal-to-noise ratio (SNR) can sometimes be very bad. I had to postpone my main development goal to build a machine learning (ML)-based Subsystem to obtain clean biological signals.)

Given the nature of biological systems, even if they fully understand PPG, ECG, PAT/PTT and other biophysics, developers will find that designing health or fitness wearable devices is not so easy. If they only focus on their signal chain and algorithms, developers can easily find that their own work will mainly revolve around how to acquire and process clean biological signals.

Using the biosensor development kit, developers can quickly build prototypes and start exploring different light wavelengths, electrode placement, or many other development efforts to optimize biosignal acquisition (or just to make it work in the first place).

Dedicated kit for prototype health wearable devices

ADI’s EVAL-ADPD4100Z-PPG evaluation kit and MAXREFDES103# health sensor are designed to accelerate the development of health wearable devices. During the development process, the EVAL-ADPD4100Z-PPG uses the company’s EVAL-ADPDUCZ Cortex-M3 microcontroller-based motherboard for programming, which is connected via a USB port.

How to quickly develop medical-grade wearable devices for vital signs monitoring
Figure 2. Analog Devices’ EVAL-ADPD4100Z-PPG evaluation board

The MAXREFDES103# kit combines the sensor subsystem based on the MAX86141 biosensor and the integrated host subsystem based on the MAX32630 MCU in a pre-built kit. In addition to buttons and LEDs for displaying the status of the device, the kit also provides a USB Type-C connector for connecting to the provided adapter board for firmware updates (Figure 3).

How to quickly develop medical-grade wearable devices for vital signs monitoring
Figure 3: The MAXREFDES103# health reference design includes wearable devices for researching on-site biosensor applications. (Image source: Analog Devices)

More importantly, each kit comes with a software package for analyzing measurement data, allowing developers to study the waveforms generated during continuous measurements under different sensing configurations and check the effects of artifacts. ADI’s Wavetool evaluation software allows developers to run EVAL-ADPD4100Z-PPG in different application modes, including SpO2 and ECG modes.

ADI’s MAXREFDES103# reference design package includes its DeviceStudio application, which allows developers to configure biosensors and embedded algorithms for heart rate and SpO2. The company’s health sensor platform is also available, which is an Android application that provides additional algorithms for sleep quality, respiration rate, and heart rate variability (HRV). The latter indicator, as a non-invasive method to monitor changes in the individual’s autonomic nervous system, has attracted special attention from the medical community.

The Links:   PM100RSD120 LM190E08-TLG6 IGBTS

Author: Yoyokuo