MXene/perovskite-based bionic human odor sensor array with machine learning

Human identification is crucial in many fields. However, it is challenging to achieve instant and accurate human identification through body odors. In this work, an in-situ growth strategy was utilized to establish a gas-sensitive nanocomposite material library with MXene and perovskite. A plug-and-play bionic sensor array module based on the material library was manufactured and assembled into an instant detection platform (IDP). Machine learning (ML) algorithms were introduced to assist IDP in identifying human odors. Due to its unique Schottky Barrier structure, the material library exhibited higher performance and responded 37 % similar to 70 % higher than initial MXene. IDP was applied to detect the odors of volunteers' breaths and clothes with an accuracy rate of 69.2 % and 51.1 %, respectively, with the assistance of ML. In all, an instant, convenient, and accurate human identification prototype machine was fabricated, providing a general solution for more complex application scenarios.

Automated summary

Achieving instant and accurate human identification through body odors is challenging. This study utilized an in-situ growth strategy to establish a gas-sensitive nanocomposite material library with MXene and perovskite. A plug-and-play bionic sensor array module based on the material library was manufactured and assembled into an instant detection platform (IDP). Machine learning algorithms were introduced to assist IDP in identifying human odors. The material library exhibited higher performance and responded 37% to 70% higher than initial MXene due to its unique Schottky Barrier structure. IDP was successfully applied to detect the odors of volunteers' breaths and clothes with an accuracy rate of 69.2% and 51.1%, respectively. Overall, an instant, convenient, and accurate human identification prototype machine was fabricated, providing a general solution for more complex application scenarios.