Page 109 - Read Online

Page 16 of 19 Huang et al. Soft Sci. 2025, 5, 24 https://dx.doi.org/10.20517/ss.2025.07

recognition system [Figure 7C]. Furthermore, the preprocessed data were input into t-distributed stochastic
neighborhood embedding (t-SNE) for dimensionality reduction and clustering. As shown in Figure 7D, the
captured data for different ball actions are clearly distinguished and the data points corresponding to the
same ball action colored identically, categorizing the data into five distinct groups. Figure 7E presents the
feature importance analysis, quantifying the contribution of each input signal to the model’s predictions.
This highlights the significant role of signal 3 in the classification process, providing valuable insights into
the underlying principles of the model’s decision-making. Finally, the learning curve in Figure 7F illustrates
the model’s performance across both the training and validation datasets as the number of training samples
increases. The progressive convergence of the curve emphasizes the model’s outstanding generalization
ability, effectively avoiding overfitting or underfitting.

CONCLUSIONS
In summary, MXene-nanocomposited organohydrogels with ultrastretchability, robust adhesiveness, self-
healing ability, and environmental stability were successfully fabricated to realize dual sensory
functionalities. By incorporating CS-encapsulated MXene nanosheets into a PAM-crosslinked network, the
system’s conductivity was significantly enhanced. The CS molecular forms a protective shell around the
MXene nanosheets via hydrogen bonds and electrostatic interactions, improving their stability and
dispersibility within the organohydrogel. Thanks to abundance of reversible noncovalent bonds, the PCM
organohydrogel exhibits ultrastretchability (2,800%), reversible adhesiveness, and self-healing ability,
making it ideal for constructing conformal electrode interfaces for high-efficiency electrophysiological
signal detection. The synergistic hydrogen bonds between GL, PA, and water molecules endow the PCM
organohydrogel with environmental adaptability (-30~60 °C), ensuring stable performance across a wide
range of operating conditions and long-term applicability. Notably, the multimodal organohydrogel
-1
demonstrates reliable temperature response ability with a satisfactory thermosensation (-43% °C ) and
excellent strain sensing performances with high sensitivity of 3.24 at strains of 2,300%-2,800% and high
stability. Impressively, the PCM sensor can detect and differentiate a broad range of human activities and
subtle electrophysiological signals. Moreover, the organohydrogel sensing system assisted by DLAs achieves
100% accuracy in ball sports recognition, highlighting its promising potential in intelligent wearable
electronics.

DECLARATIONS
Acknowledgements
The authors would like to thank the National Supercomputing Centre in Zhengzhou and the funding of
Zhengzhou University for the DFT calculations.

Authors’ contributions
Formal analysis, investigation, data curation, writing - original draft, and visualization: Huang, M.
Formal analysis, software: Liu, S.
Investigation, formal analysis, and validation: Chi, Y.
Visualization, validation, formal analysis: Li, J.
Conceptualization, methodology, resources, writing - review and editing, and supervision: Sun, H.
Methodology, and writing - review and editing: Dong, L.
Conceptualization, methodology, resources, writing - review and editing, supervision, and funding
acquisition: Liu, H.
Methodology, writing - review and editing, and funding acquisition: Liu, C.
Funding acquisition: Shen, C.

Availability of data and materials
All data are available in the main text or the Supplementary Materials. Information requests should be
directed to the corresponding authors upon reasonable request.