The fusion of wearable technology and advanced physiochemical sensing holds immense potential for real-time health monitoring. However, traditional manufacturing struggles to meet the demands of personalized at-home applications. In a recent paper published in the journal Science Advances, researchers proposed a semisolid extrusion 3D printing method to create versatile epifluidic elastic electronic skin (e3-skin) with top-tier sensing abilities.
Study: Revolutionizing Health Monitoring with 3D-Printed e3-Skin and AI. Image credit: Kitreel/Shutterstock
Background
Maintaining a balanced lifestyle and early symptom recognition are pivotal for overall well-being and longevity. Wearable technology is revolutionizing healthcare, offering personalized medicine and digital health solutions. Skin-interfaced wearables provide real-time insights into vital signs such as body temperature and heart rate. Microfluidic sweat analysis offers rich biomolecular data linked to health. Developing multimodal wearables that combine molecular sensing and vital sign tracking is crucial for comprehensive health monitoring and can unlock various health surveillance and clinical applications.
However, creating these devices is complex, requiring customizable materials and designs. Nanomaterials and composites enhance sweat sensor efficiency, and biomolecule incorporation is needed for selective biomarker detection. Polymer hydrogels enable transdermal drug delivery, while microfluidic channels regulate sweat flow. Three-dimensional structures are essential for pressure and strain sensing. Traditional fabrication involves multiple cleanroom facilities and manual assembly, limiting scalability.
To address these challenges, the e3-skin is introduced. It utilizes semisolid extrusion (SSE)-based 3D printing, employing direct ink writing and selective phase elimination. Functional inks are customized for precise patterning of multidimensional components. Rheological properties matching SSE requirements are achieved. Phase elimination enhances performance, enabling low-cost prototyping of sustainable, multifunctional physiochemical sensing systems ideal for remote healthcare.
Integrating machine learning and e3-Skin
The biophysical sensors of e3-skin and interconnects were crafted with precision using semisolid extrusion (SSE) technology employing an aqueous MXene ink. MXene nanosheets, with their unique properties, allowed for stable and high-resolution printing on flexible substrates. MXene was also utilized to create a temperature sensor in the e3-skin, displaying sensitivity to temperature changes within a physiological range. This sensor maintained stable performance even during mechanical bending and skin contact.
For pulse monitoring, a pressure sensor was designed with an interdigital MXene electrode and a porous carbon nanotube (CNT)-polydimethylsiloxane (PDMS) foam. The porosity of foam contributed to its high-pressure sensitivity and could be optimized for robust mechanical stability. It effectively monitored radial pulses in human subjects and demonstrated resilience through extensive cycles of compression and release.
The e3-skin incorporated various electrochemical biosensors via 3D printing. Enzymatic biosensors were created by printing porous CNT-styrene-butadiene-styrene (CNT-SBS) as the working electrode, MXene-Prussian blue (MX-PB) as the redox mediator, and bioactive polymers with enzymes for target recognition. The optimized 3D-printed porous CNT-SBS working electrode showed superior electrochemical performance compared to commercial electrodes. Enzymatic sensors for glucose and alcohol were printed, exhibiting sensitivity and stability across physiological concentrations. A pH sensor based on a CNT-SBS-polyaniline (PANI) electrode displayed near-Nernstian sensitivity and stability within a physiologically relevant pH range.
Microfluidic components, including an iontophoretic sweat induction module, were 3D-printed to enable on-demand and continuous molecular monitoring. The e3-skin adhered well to the skin and maintained stable sensor performance during mechanical deformations. It exhibited high biocompatibility and low cytotoxicity in cell culture studies.
To power the e3-skin, 3D-printed MXene micro-supercapacitors (MSCs) were designed to interface with a solar cell for energy harvesting. These MSCs demonstrated high capacitance and energy density, even with mechanical bending. Multiple MSCs could be connected in series to provide sufficient voltage for powering the wearable sensors. The fully integrated e3-skin and wearable system, equipped with energy harvesting and wireless communication capabilities, demonstrated its potential for continuous health monitoring and data transmission.
The ability of e3-skin to perform continuous multimodal sensing was evaluated on human subjects. In glucose tolerance tests, it simultaneously monitored vital signs and glucose levels, revealing rapid glucose increases after meals. Over 12 hours, the e3-skin provided stable sensor responses during various daily activities, including exercise. This comprehensive wearable system holds significant potential for real-time health assessment and early prediction of health states.
Artificial intelligence-driven 3D-printed e3-skin assesses behavioral responses. The impact of alcohol consumption on cardiac, metabolic, cognitive, and behavioral functions is well documented. Even moderate alcohol intake can impair reaction time, coordination, judgment, and self-control. Traditional methods struggle to assess behavioral responses accurately due to alcohol's varying effects among individuals. In a pilot study, the e3-skin collected real-time data, including alcohol levels and vital signs, and, coupled with machine learning, accurately predicted behavioral responses, such as reaction time and inhibitory control impairment. The study involved five subjects, and physiological responses were correlated with alcohol intake. Machine learning achieved high prediction accuracy for these responses, with sweat and alcohol levels playing a crucial role.
Conclusion
In summary, SSE-based 3D printing offers an elegant solution for crafting sustainable wearable health monitors. The all-3D-printed e3-skin detects various biomarkers, including glucose, alcohol, pH, heart rate, and temperature. It features an iontophoretic module for sweat induction, microfluidic sampling, and an MSC for energy storage.
Machine learning enhances health assessment, predicting behavioral impairments with over 90 percent accuracy after alcohol consumption. This low-cost 3D printing technology holds potential for remote health monitoring and clinical applications.
Machine Learning Optimizes Polymer Analysis