A GA-BP compensated PEDOT:PSS/MXene wearable NH3 detection device for food safety, environmental monitoring, and disease diagnosis

The real-time and accurate detection of ammonia is critical for food freshness testing, agricultural monitoring, and respiratory health diagnostics. Recently, flexible gas sensors have garnered attention owing to their potential for integration into smart wearable electronics and display devices. However, the development of high-performance, low-power, and stable devices remains challenging. In this study, a chemoresistive sensor based on poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS)/MXene composites is designed, wherein the application of a genetic algorithm-optimized back propagation neural network to compensate for humidity leads to a significantly enhanced detection accuracy. Notably, the use of a genetic algorithm-optimized back propagation neural network reduces the detection error from 8.06 % to 1.3 % in the presence of 50 ppm NH 3 . A wireless Bluetooth monitoring system is integrated to enable real-time data transmission and remote monitoring. This sensor detects ammonia within the range of 0.6–1000 ppm, demonstrating an excellent stability, selectivity, and flexibility, thereby confirming the synergistic effects between PEDOT:PSS and MXene. Importantly, in application tests, the developed system reliably assesses meat spoilage and farm ventilation levels, indicating its high applicability and reliability. Successful human breath tests also open new possibilities for the application of this sensor in respiratory health monitoring. Ultimately, this study offers a convenient and efficient approach for food quality assurance, disease diagnosis, and environmental monitoring.