Advances and current challenges in non-invasive wearable sensors and wearable biosensors—A mini-review

2.1 Epidermal devices

The skin acts as a physical barrier that protects the internal organs from contact with the environment—having several functions in the physiological maintenance of the human body (Chang et al., 2019; Someya & Amagai, 2019). Through contact with the skin, it is possible to monitor different targets and biomarkers.

The development of devices should consider the selection of favourable substrates using functional, biocompatible materials and low-cost fabrication techniques. For epidermal devices, a critical focus is the elasticity of the systems that must be compatible with skin users (Mishra et al., 2018; Zhang, Chen, et al., 2019). Polymeric materials as polyethylene terephthalate (PET), polyimides, among others, are commonly used. Among the most used materials stands out the Ecoflex due to presenting Young's modulus that approximates that of human skin (Kim et al., 2019; Xu et al., 2019; Yang, Kong, & Fang, 2019). The most frequently used active materials for sensor construction are carbon derivatives and silver-based metal nanomaterials (Arenas et al., 2019; Muralee et al., 2020). These can be used to create devices in the form of stickers, tattoos, bandages or even smart clothes for specific monitoring of the sweat, ISF for human motion.

Compared to other biofluids, sweat has advantages, such as the possibility of analytes detections, it is sampling non-invasively, continuously, comfortably, with better correlation with blood levels about diagnoses of diseases, stress or sports monitoring. Usually, its collection does not prevent or restrict the user from performing natural movements.

Sweat is secreted by the sweat glands to regulate body temperature. They are divided into two types, the eccrine and apocrine sweat glands (Legner et al., 2019). The eccrine secretion is odourless and composed of urea, sodium chloride and immunoglobins. While apocrine sweat is composed of steroids, lipids, proteins and pheromones, typically, eccrine sweat is the most relevant for WBS studies (Legner et al., 2019).

Some authors assume that many of the chemical species found in sweat are transport by-products of ISF or blood vessels, which would explain the similarity of the analytes found in them.

The WBS in the monitoring of athletes is one of the leading applications (Seshadri et al., 2019a, 2019b). Since sodium is the primary electrolyte of the human body, being essential for the regulation of osmotic pressure, pH, and water balance, providing critical information about electrolyte imbalance after activities (Qi et al., 2020), long physical exercises or exposed hot and humid environments may also indicate dehydration. Increased lactate in sweat can be associated with changes from aerobic to anaerobic metabolic conditions during activity, while monitoring of ammonium may give evidence of extreme fatigue (Mccaul et al., 2018).

In this context, Zamorra and co-works performed tests on different conductive tissues modified with electrodeposited iridium oxide (IrO₂) to form wearable pH-sensitive sensors. Being these characterized by impedance measurements, the steel mesh, in turn, showed sensitivity to pH changes in addition to biocompatibility regarding the IrO₂ coating. The sensor was simple, inexpensive, and washable, allowing it to be reused (Zamora et al., 2018). Dang and co-works presented a flexible system for sweat pH monitoring. They developed graphite and polyurethane composite pH sensor, sensitive in the range of pH 5 to 9, and the sensor showed no interference from the primary ions present in the sweat. The sensor withstood up to 53% deformation and more than 500 cycles. Besides transmitting power-free data through an RFID antenna to a smartphone, the authors say that this system can be manufactured at less than U$ 3.53 (Dang et al., 2018).

McCaul and co-works developed and tested a fully integrated wristwatch-based 3D printing platform to analyse sodium concentration in real-time. The platform showed a delay of about 5 min between skin surface sweat detection and the detector (Mccaul et al., 2018). It is hoped that in the future, similar devices will be able to have an even faster detection. In amputees, these sensors may be placed at the muscle or skin sites to which nerves may be reinnervated (Willand, 2015), allowing the monitoring of physical signals in the region or even specific biomarkers. Sun and co-works developed a large, subtle human motion sensor based on a polyacrylamide composite hydrogel and oxidized multi-walled carbon nanotubes, with high tensile strength 0.71 Mpa, with elasticity major 700%, response 300 ms fast and capable of over 300 cycles (Sun et al., 2019). Similarly, Zhang and co-works presented a new strain sensor based on an ultra-elastic hydrogel conductive polymer composite, which demonstrated stability and high creep sensitivity for detecting far-reaching human movements. The hydrogels showed a durable adhesive behaviour on different surfaces. They showed no allergy or residue in the human skin, which demonstrated its potential as a detector of movement and torsion in the human skin (Zhang, Liu, et al., 2019). Other relevant studies on epidermal devices were summarized in Table 1 that presents the relevant characteristics of the devices studied in these works, such as position in the body, kind of sensor, sensitivity, tests evaluated, kind of substrate and response time.

Figure 5 shows some more recent and relevant epidermal device examples. Ahyeon Koh and co-works developed and evaluated different properties of a new microfluidic device to monitor sweat. This device allowed the multivariate colorimetric detection of different types of analytes, such as lactate, glucose, water, chloride and pH (Figure 5a–c).

B. Xu and co-works developed multifunctional epidermal devices perfectly integrate sensors for different kinds such as electromyography, temperature and voltage with electrical stimulation of compression electrodes, in a simple architecture that in the future should allow prosthetic control as a sensory response (Figure 5d,e). Yokus and co-works development a new WBS based on multiplexing, hardware, integrated with wireless technology for electrochemical analysis of glucose, lactate, pH and temperature, with continuous detection sensing in simulated and sampled sweat (Figure 5f,g).

Torrente-Rodríguez and co-works evaluated a set of multiplexed adhesive type sensors used for glucose monitoring during exercise, with simultaneous measurement of pH, temperature and humidity to correct the glucose signal (Figure 5h,i). Farooqui and Shamim produced a new bandage used on the forearm, and this sensor informs the presence of blood in the bandage.

2.2 Oral cavity devices

Oral cavity sensor and biosensors, usually based on the electrochemical or optical transducer placed in the oral cavity, has experienced considerable progress and shown advantages faced to other systems, mainly because of the versatility of use saliva as biofluid.

Saliva consists of a complex biofluid constituted by several analytes. They are permeating from the blood in different ways (transcellular or paracellular paths). The analytes (e.g. ions, acids, sugars, enzymes, hormones, peptides, proteins, antibodies, microorganisms, growth factors and so on) have shown high potential for general health surveillance (Bandodkar & Wang, 2014; Lee & Wong, 2010).

To be more readily accessible than the bloody and requiring fewer pretreatments, it has been used to the monitory nutritional, hormonal, emotional and metabolic state of the human body (Aguirre et al., 1993; Ilea et al., 2019; Kim, Campbell, et al., 2019; Malon et al., 2014).

An interesting study was presented by Mannoor and co-works (Mannoor et al., 2012), which developed a dental tattoo based on graphene printed onto water-soluble silk and modified with antimicrobial peptides, for continuous monitoring of pathogenic bacterial on tooth enamel. Despite real time on body applications that have not been described, this oral wearable biosensor showed excellent specificity, response time, with limits detection down to a single bacterium, as well as wirelessly achieving remote powering and readout.

More recently, researches on oral cavity wearables biosensors have resulted in the development of several mouthguard platforms for sensing the biomarkers in saliva. In this context, Ciui et al. (2018) reported an electrochemical oral cavity sensor for glycated end compounds (Nε (carboxymethyl)lysine), a class of biomarkers involved in oxidative stress, which produce long-term damage to proteins in ageing processes, atherosclerosis and diabetes. Briefly, the sensor was fabricated via screen printing method on a standard and cheap, flexible foil, using Ag/AgCl conductive ink for providing reference electrode and carbon ink to give work and counter electrodes, being posteriorly attached to a mouthguard. The sensors exhibited high repeatability, lifetime storage, short timescale measurements, ease of use, and high selectivity and sensitivity with a limit of detection of 166 ng mL⁻¹ in phosphate buffer. Lastly, when tested in raw and untreated human saliva, promising results were observed.

Saliva used as biofluid has not been restricted to the biologic factors monitoring as nutritional, hormonal or metabolic state of the human body but also has been applied to monitoring physical issues as a fracture. In this context, an unusual approach was developed by Hashem and co-workers (Hashem et al., 2020), which demonstrated a wearable bio-tooth sensor to the monitoring of tooth fracture and its metabolites. The idea consists of a multipurpose wireless wearable graphite bio-tooth sensor to monitoring, coaching, drinking, chewing, fracture and infection. Briefly, the biosensor (range of 2 × 3 mm) consists of a human–machine interface responsible by imaging, integrated to biomarkers analysis to diagnose the various motion patterns from the oral cavity. The biosensor showed to be capable of analysing object surface aiming tooth shape reconstruction and, besides, provided a solution to validate the 3D-intraoral images using an image acquisition sensor.

The development of the oral cavity WS and WBS has improved in the last years, and specific issues still have to be carefully approached in the design of an ideal biosensor. Due to complex mouth anatomy, comfort becomes crucial in the choice of adequate substrates. Thus, aiming minimal inconvenience to the user, properties as flexibility, lightness and thickness are encouraged. Also, toxicity assays need to be carried out to investigate the biocompatibility of many electronic components, which shows absent in many reported studies. Lastly, are summarized in Table 2 some critical results about the development and employment of several oral cavity wearables biosensors applied in the analysis of a wide range of analytes.

Some other relevant examples were shown in Figure 6. Y. Lee and co-works developed an intraoral electronic device that provides continuous monitoring of sodium through a hybrid electronic system of microchips in a membrane and evaluated the monitoring of sodium in real time in humans (Figure 6a). Similarly, Xuemeng Li and collaborators developed a mouthguard sensor sensitive to the presence of volatile sulphur to evaluate and diagnose oral diseases. The developed sensor presented a highly sensitive and selective response to the sulphur, in addition to biocompatibility, low toxicity and low cost (Figure 6b). Mohamed Hashem and co-works built and evaluated on a laboratory scale a graphite-based WBS capable of detecting different analytes, including saliva, by analysing the surface of devices (Figure 6c). Similarly, Takahiro Arakawa and co-works developed a mouthguard for non-invasive glucose monitoring by saliva with high sensitivity to artificial glucose (Figure 6d).

Jayoung Kim and co-works developed a WBS for continuous monitoring of salivary uric acid via electrochemical detection and with transmission via Bluetooth 4.0 (Figure 6e). Peter Tseng and co-works developed a device based on permeable membranes for detecting liquid analytes. In vivo evaluation was demonstrated for detecting food during human ingestion (Figure 6f). Lantada and co-works developed a splint-type sensor to assess intraoral tightness to assess bruxism and other pathologies (Figure 6g). Wu and co-works developed a WS to accurately assess head impacts caused during a football match (Figure 6h).

2.3 Ocular WBS

The seek for vision corrections and aesthetic reasons make the ocular research be widely studied (Korde et al., 2019; Lim et al., 2019; Yuan et al., 2019). The most common material in contact with the eyes is the contact lenses, a thin curved lens placed on the film of tears that covers the surface of the eyes (Horne et al., 2020; Mayers et al., 2019). Due to its presence in the ocular environment and being an inexpensive material with uncomplicated production, it can also be a potential WBS for other applications (Kim et al., 2020; Reid & Mahbub, 2020).

The miniaturization of electronic components facilitates the incorporation of sensors into contact lenses (Yu, Yang, & An, 2019). The diagnostic for healthcare through electronic sensors in contact lenses are more straightforward than those obtained by blood analysis (Farandos et al., 2015). Even though there are differences between the blood and tear composition, the barrier between them allows the migration of some metabolites that can be correlated. Therefore, it is possible to analyse the tear fluids and compare to the condition of the blood fluids without needing an invasive procedure (Kim et al., 2017).

The tear fluids present in the eye have a vital function. Lubrication and cleaning (basal tears), protection (reflex tear), emotional (psychic tear) and its composition has proteins, electrolytes and antibodies, providing quantitative information about potential diseases (Aparna & Shanti Iyer, 2020; Salim & Lim, 2019).

In order to ocular WBS to be useful, it is necessary to overcome some tears analysis challenges related to the type of tear due to its difference in composition, the amount of analyte and the evaporation of fluids (approximately 10 µl of tears can be collected; Farandos et al., 2015).

The main application of contact lenses biosensor is the glucose concentration monitoring in tears fluids for diabetes treatment, where it is more advantageous than the blood test, which causes some physical discomfort (Bussan & Robertson, 2019; Forster & Cumba, 2019). Another monitoring made by ocular sensors is the intraocular pressure, directly related to glaucoma issues, which can be detected and treated early (Aitsebaomo et al., 2019; Campigotto et al., 2019).

In order to succeed the ocular WBS uses, it will be necessary not to interfere with the transparency and flexibility conditions of the contact lenses and nor to affect the optical necessities (Kim, Kukjoo, et al., 2019; Wei et al., 2019). The electronics compounds incorporated must be capable of having their mechanical, chemical and electrical properties for a highly sensitive, selective and wireless response, which needs to be connected in real time with the user by a computer or a smartphone (Takamatsu et al., 2019).

Many materials have been proposed in the ocular devices. Due to the specification of the required condition, the most well succeeded are polymers and its hybrids (Li et al., 2019), graphene-based structures (Lee & Yook, 2019), metal nano and mesh structures (Kodama & Yamada, 2020) and carbon nanotubes (Kahng, 2019). All these materials can provide a cheap contact lens with the biosensor incorporated and allow a non-invasive diagnostic through the human eye.

In this context, Zou and co-works (Zou et al., 2019) developed an intraocular biosensor made by Chitosan functionalized nitrogen/graphene for glucose detection. Due to the sp₂-hybridization from graphene structures, it allows a reactive site for electrochemical reactions, and this reactivity is intensified when associated with nitrogen. The well-synthetized ocular WBS was tested with the presence of other reactive materials such as uric acid, ascorbic acid and dopamine, commonly present in tears. No change in the detection of glucose was observed, confirming then the selectivity of the biosensor.

Xiao et al. (2018) have demonstrated a gold-based for a potential power source in contact lenses where the Au-Ag alloys can support enzymes (lactate oxidase and bilirubin oxidase) for an electrocatalytic response. This biosensor worked for 5.5 hours at 150 mV (artificial tear solution) but lost his density drastically due to the presence of ascorbate present in tears composition, which quickly oxidizes the nanostructured gold electrode. The stability found in this biosensor can be used for one 1-day disposable lenses. In Table 3 are summarized the other critical results about the ocular WBS. More examples are shown in Figure 7. Agaoglu and co-works developed a microfluidic sensor with a transduction mechanism detectable by a smartphone camera, with a low detection limit <0.06% for uniaxial and <0.004% for biaxial strain. This sensor was evaluated measured the intraocular pressure-induced strain in porcine eyes in the physiological range, with continuous monitoring for >19 hr and a lifetime reaching >7 months (Figure 7a,b).

Chiou and co-works developed a biocompatible contact lens sensor wireless with a capacitive circuit interface and a wirelessly powered radio-frequency identification (RFID) addressable system for sensor control and data communication with a low sensibility and low energy consumption, 120pF and 110 μW, respectively (Figure 7c). Joohee Kim and co-works reported the manufacture of sensors in soft and transparent lenses for the detection and monitoring of glucose with wireless energy transfer. Live tests showed that the sensors are suitable for determining the fasting glucose level (Figure 7d,h)).

Hongbin An and co-works developed and evaluated a sensor system to monitor intraocular pressure based on a microfluidic lens. The tests on enucleated porcine eyes show a linear response, and these lenses can track the intraocular pressure changes (Figure 7e). Similarly, Yin and co-works developed lenses with graphene electrodes to register full-cornea recording of electroretinography (ERG) in monkeys. These sensors show a higher signal amplitude than conventional ERG electrodes for in vivo visual ERG (Figure 7f). Chu and co-works developed a biocompatible WBS based on contact lenses to glucose monitoring. This WBS shows a good relationship between output current in different glucose concentrations 0.03–5.0 mM that is commons in normal and diabetics patients (Figure 7g).

Leonardi and co-works reported a wireless sensor with a good for the continuous monitoring of intraocular pressure based on WS lens. This system allows minimally invasive glaucoma monitoring (Figure 7i). More recently, Takamatsu and co-works developed a WS through electrochemically printing a wireless-powered circuit onto a moist, soft contact lens. This new sensor allows an energy transfer between the glasses transmitter and the receiver lens and will allow the integration of more complex circuits in the future (Figure 7j).

Kouhani and co-works reported another excellent example of wireless WS to real-time monitoring of intraocular pressure. In this work, the authors report a WS with design simplicity, high responsivity and low cost (Figure 7k). Finally, Lee and co-works reported a new glass WS not only to monitor various biological phenomena but to coordinate the recorded data for active commands and game operations for human–machine interaction applications and to monitor electroencephalogram and electrooculogram (Figure 7i).