Page 14 - Read Online

Page 12 of 35 Huang et al. Soft Sci 2024;4:40 https://dx.doi.org/10.20517/ss.2024.37

Sensing module
The sensing module is designed to monitor various parameters related to physical health. Sensing fibers can
be integrated into textiles to establish direct contact with the skin or other body parts. For textile-based
pressure sensors, the working principle relies on the increased contact area between sensing fibers and
[177]
electrode fibers, as well as the interactions between the sensor fibers themselves . The contact and
compression within and between the fibers lead to an increase in the conductive pathways, thereby
enhancing the overall conductivity of the sensor. Additionally, the sensitivity of acoustic sensors to sound
pressures and frequencies at various distances suggests the potential for harnessing energy from ambient
noise and vibrations . In addition, ultraviolet (UV)-sensitive textile electronics are mainly realized
[178]
through materials that respond to UV exposure . Temperature sensors can be created using temperature-
[179]
sensitive materials as textile electrodes . If the body’s electrolyte balance, kidney function, or breath odor
[180]
needs monitoring, sensors can be designed to detect biochemical signals related to liquids or gases [181,182] .

Interconnections
Textile-based sensors require connection to other circuit elements, additional sensors, and/or data
acquisition circuits to function effectively. Therefore, flexible or rigid interconnects and connectors are
critical to ensuring the functionality and durability of the final monitoring system. Interconnects are
conductive pathways that link various textile electronics within or between textiles. These pathways can be
made from conductive fibers, tracks printed with conductive inks, or other forms of conductive
materials . Typical interconnects found in textile electronics that are suitable for textile-based sensors are
[70]
conductive yarns . Connectors are physical interfaces used to connect conductive pathways in textiles to
[40]
rigid electronic components, such as integrated circuits (ICs) and sensors. These connectors provide a
reliable electrical connection between the softness of the textile and the rigidity of the electronic
components. Connectors can be categorized into three types: physical connections (such as soldering and
[31]
welding) , mechanical connections (such as gripping, crimping, and sewing) , or connections using
[183]
conductive adhesives (such as conductive epoxy) [184,185] . Moreover, the electrodes in textile electronics, often
integrated as conductive fibers or yarns, play a crucial role in the detection, collection, and transmission of
signals . The synergistic integration of the various textile electronics, electrodes, interconnects and
[186]
connectors holds great promise for realizing a comprehensive textile-based health monitoring system.

Data processing module
The data processing module serves as the central component of the health monitoring system. After
capturing the initial sensing signals, a signal conditioning circuit that includes amplifiers, filters, and analog-
to-digital converters (ADC) is utilized to process these signals. Subsequently, the processed data is
transmitted to the microcontroller for further analysis and processing.

This module is underpinned by a suite of algorithms that leverage statistical analysis, machine learning, and
other sophisticated data analysis techniques to enhance the accuracy and efficiency of health monitoring
protocols. In textile-based health monitoring systems, machine learning provides a range of powerful data
analysis techniques. Compared to traditional data processing methods, machine learning can handle
datasets with high complexity and multiple variables, which is critical for processing extensive health
monitoring data. Algorithms such as support vector machines (SVM), convolutional neural networks
(CNN), K-nearest neighbors (KNN), and decision trees can make predictions about new, unseen data after
undergoing training. Although the data processing module is crucial, it often receives less attention in the
literature compared to the more prominent sensing modules. Table 1 summarizes several machine learning
algorithms utilized in textile-based health monitoring systems, providing a reference for the future
development and optimization of new algorithms.