Page 68 - Read Online

Page 2 of 25 Nagwade et al. Soft Sci 2023;3:24 https://dx.doi.org/10.20517/ss.2023.12

[2]
technologies, it was evident that HMI technology would have to adapt and evolve to improve its usability .
Eventually, wearable devices with HCI capabilities were introduced to the world, and considering human
aspects such as physical, cognitive, and emotional characteristics of these devices is essential during their
[3]
development . Refining designs, configurations, and ergonomics of rigid wearable devices for HCI
technologies would not be enough to enhance the user experience. Therefore, using soft and advanced
engineering materials is crucial for making HMIs more competent and up to par to pass as a mode of
communication in wearable HCI devices .
[4]

With the constant progress in health and lifestyle technologies, there has been an uprising in wearable
devices and the features they can offer . Wearable devices are essential tools for HCI technologies that can
[5-9]
[10]
record individual biological and neural activities through close contact with human skin . All kinds of
physical and mental activities in the human body generate electrical potentials, called biopotentials, that
provide colossal amounts of information, which can be used for various applications, such as health
monitoring and controlling electronic devices [11-16] . For instance, neural activities in the brain can be
recorded, analyzed, and used for various purposes. Motor actions that are intention-based, such as walking
or moving arms, can be monitored in the form of neural activities in the brain. Similarly, when sensory
inputs are perceived in the brain, such as touch or taste, the associated neural activities can be observed in
the brain. The most common technique for non-invasive brain recording is electroencephalography
(EEG) . Although EEG allows the recording of neuronal activity non-invasively, it is difficult to isolate and
[17]
record specific areas of the brain . Neural signals from the brain travel in the human body and branch out
[18]
to different areas through the nervous system. Similarly, the sensory inputs from the body are sent to the
brain via these nervous systems. Apart from EEG, electromyogram (EMG), electrocardiograph (ECG), and
electrooculogram (EOG) signals are the major biopotential signals that can be acquired using non-invasive
interfaces . Other than acquiring data from these signals, non-invasive interfaces can also perform
[19]
stimulation-based applications for specific rehabilitation purposes [20-22] .

Wearable devices can utilize biopotential signals, such as EMG (from muscular activity) and EOG (from eye
movement), for delivering intention-based output signals to allow applications such as gesture control, end-
effector manipulation, and prosthesis. Unintentional or autonomous biopotential signals, such as ECG
(from heartbeats), can allow for health monitoring applications. In HCI, such information is valuable in
various industries, including medical healthcare, assistive robotics, lifestyle, and more. Figure 1 illustrates
the possible HCI applications that can be controlled via wearable soft biopotential interfaces. Figure 2
illustrates the advantages of using biopotential signals for wearable devices.

The most widely accepted and used interfaces for recording biopotentials are silver/silver-chloride (Ag/
AgCl) electrodes . These electrodes are the gold standard for conducting tests and research in the medical
[23]
and bioinstrumentation industry . Their performance has been constantly under study, and their
[24]
[25]
parameters are well-defined, displaying remarkable consistency . However, Ag/AgCl electrodes, due to
their use of electrolyte gel, cannot be used for the long term, which is an important consideration for
producing convenient wearable devices. Moreover, the materials used have a short life cycle. Other issues,
such as inelasticity, obtrusiveness, and non-reusability, make them unfit for integration with wearable HCI
devices [26-28] . Figure 3 shows the basic Ag/AgCl biopotential electrode and its electrical circuit model while
interfacing with the skin. A usable biopotential signal is generally acquired by amplifying the surface
recording by an electrode interface. Figure 4 shows a schematic for surface EMG (sEMG) biopotential signal
acquisition.