Tech News: Fabrics that can think
Share this article:
By: Louis CH Fourie
The COVID-19 pandemic has revealed the dire need to assess the health of large populations in real-time. Current approaches of symptom-driven tests are lagging indicators and is like driving only with a rear-view mirror.
We need real-time systems that are predictive, future-looking and can deliver leading indicators. Systems that can access vital signs continuously and detect subtle, almost indiscernible small changes as early warning signs of health issues of an individual. This data of large populations could be correlated in space and time to accurately detect the spread of the various waves of a pandemic.
No human-made objects are more exposed to masses of vital data than the clothes we wear. If we could teach our fabrics to sense, store, analyse, extract and communicate this potentially useful information, we could make a major difference to health management.
Fortunately, smart clothing and advanced fibres and fabrics is a fast-growing field that will certainly transform many facets of our lives. Whether clothing, upholstery, curtains or bedsheets, we continuously interact with textiles across a variety of personal and professional situations. The ubiquitous presence of textiles close to a person makes them very useful platforms for user interfaces of cloud computing systems or health monitoring sensors.
Fabrics that are in daily contact with humans are exposed to large amounts of data. Important health insights could be revealed if only fabrics could compute or sense, store, analyse, infer, alert and act while simultaneously retaining their traditional textile qualities such as stretching, twisting, knittability and user comfort.
Recent progress in fibre material science has placed some innovative new functional fabrics of the future within reach:
•A new class of fast-evolving fibres that can sense, capture, store, process and communicate information, as well as change colour and even provide power.
•Artificially intelligent fabrics programmed with machine learning algorithms to analyse patterns in human activity and subsequently derive hidden patterns and health insights from the physiological data.
•Active components woven into the fabrics that provide value-added services, eventually changing fabrics from goods into advanced platforms of services.
All the innovations entail a new class of highly technological, rapidly evolving fibre materials that look like normal fibres but behave more like computer chips with power, acoustics, material release, colour change and input-output functions all contained in the strands of the fibre. It is foreseen that electrolyte gels, piezoelectricity (the electric charge that accumulates in certain solid materials in response to applied mechanical stress), carbon nanotubes, and drug compounds, among others, will find their way into fibres.
The Massachusetts Institute of Technology (MIT) have succeeded in producing these smart fibres through thermal drawing. Typical examples are fibres that sense temperature through a thermoresistive effect (the change of electrical resistance corresponding to the variation in temperature), detect chemicals in sweat through chemiluminescence (the emission of light as the result of a chemical reaction), and sense touch and motion through the piezoelectric effect (ability of certain materials to generate an electric charge in response to applied mechanical stress).
More recent work includes fibres that can detect the heart rate through photoplethysmography (an optical technique used to detect volumetric changes in blood in peripheral circulation) and optically communicate this physiological information through visible light or electrical signals to a smart phone. The power to drive these smart fibres will possibly come from harvested low-grade energy from vibration, movement and friction or stored energy in fibre batteries or supercapacitors as was researched in the past number of years.
Until now, most research has revolved around sensing, power and communication. However, there remains a need for fibres that can store and process information and can run software to collect and analyse data.
From intelligent fibres to fabric computers
When individual fibres have the capability to perform sensory, data-storage, power and processing functions, a collection of such fibres (a fabric), can be transformed into a fabric computer that executes programs and software to collect data through fibre sensors and store it in the memory component of the fabric computer.
Researchers at MIT have created the first fibre with digital capabilities, able to sense, store, process, analyse and infer activity after being sewn into a shirt.
Professor Yoel Fink says that digital fibres create new possibilities for fabrics to uncover the context of hidden patterns in the human body that could be used for physical performance monitoring, medical deduction and early disease detection. It could also allow the wearer, for example, to store wedding music in the gown on the wedding day. Digital fibres add a new information content dimension to textiles and allow fabrics to be programmed.
The new fibre was developed by placing hundreds of square silicon microscale digital chips into a preform that was then used to produce a polymer fibre. By accurately controlling the polymer flow, the researchers were able to create a fibre with continuous electrical connection between the chips over a length of tens of metres. The fibre is thin and flexible, can be passed through a needle and cannot be felt by the wearer.
The research team further devised a digital addressing method that allows them to "switch on" the functionality of only one element within a fibre without turning on all the elements. They were able to write, store and read information on the fibre, including a 767-kilobit full-colour short movie file and a 0.48-megabyte music file. The files can be stored for two months without power.
Artificial Intelligence (AI) in fabrics
The fabrics we wear are exposed to a variety of acoustic, optical, electrical, biochemical and biological signals coming from our body, which provides very useful data. Unfortunately, we delete this data every time we wash the piece of clothing. To be useful the data must be captured and stored in the fabric computer. AI tools and machine-learning algorithms can then be used to identify specific patterns in this dataset and unlock the potential of personalised health care.
The MIT researchers included within the fibre memory, a neural network of 1 650 connections. They use the fibre to collect 270 minutes of surface body temperature data and analyse how the data correspond to different physical activities. Trained on the data, the fibre was able to determine, with 96% accuracy, what activity the person wearing it was engaged in.
With this analytical power, AI enables the fabric to uncover hidden markers for diseases and illnesses in real-time, as well as detect important correlations between human health and physiological parameters.
The machine learning algorithms can also be used to support the human wearer of the fabric, for example, an artificially intelligent fabric with a sensor for heart monitoring can potentially detect minute deviations in heart signals before a heart attack. The fibre could also sense and alert a person to respiratory decline or deliver muscle activation or heart rate data to athletes during training.
Researcher Young and Chen, from the University of Massachusetts, developed a textile-based pulse monitoring device that can identify obstructive sleep apnoea in wearers. They used the intrinsic property of textiles to enable their device function.
Textiles microscopically stretch and twist at specific pulse points, such as the inside of one’s wrist, as the pulse of blood travels through the wearer’s body. This thus allows metal-yarn-based triboelectric devices at pulse points to record the stretches and transducing the pulse as a contact- or tribo-electrification signal.
Fabrics as a platform for services
Scientists predict that eventually smart fabrics will culminate in providing numerous value-added services. The information and deductions that smart fabrics derive about human activity and health open the door to a future where fabrics become a valuable source of relevant and actionable information.
The fusion of software, computing and sensing capabilities in fabrics will enable us to not only extract information but also form a multisensory human interface, thus transforming fabrics into a platform for value-added services and experiences.
Smart fabrics are indeed promising, but some challenges need to be overcome. An important challenge is that fabrics that are worn and enjoyed, must be flexible, machine washable, safe, lightweight, durable, stretchable, breathable, beautiful and comfortable. Textile and garments are demanding technology platforms due to their complex and three-dimensional topologies and the large, diverse mechanical stresses to which they are regularly subjected. These mechanical stress result in frequent and unavoidable delamination, abrasion damage and/or cracking in the microelectronic devices or connections that are embedded into textiles.
No single material answer to all the criteria and researchers will have to look to combinations of traditional fibres and new innovative fibres to create the new computing fabrics that are technologically advanced but also soft to touch and as aesthetically pleasing as traditional fabrics.
Smart fabrics will increasingly become part of our lives. Health monitoring devices in the form of everyday clothing, bedding or upholstery have the promise of being readily adopted and frequently used in a user’s natural environment.
From a knowledge gathering point of view, smart fabrics and garments can provide longitudinal datasets of vital signs from which the intricate interrelationships between lifestyle and everyday habits and selected chronic illnesses can be systematically determined. Perhaps we would even be able to manage the waves of the Covid-19 pandemic better.
Professor Louis Fourie is a technology strategist.
*The views expressed here are not necessarily those of IOL or of title sites.