Abstract: End-tidal carbon dioxide (ETCO₂), as an important physiological parameter reflecting the state of gas exchange in the lungs, has significant application value in medical scenarios such as respiratory disease management. With the increasing demand for respiratory health monitoring and portable medical devices, respiratory monitoring is advancing toward continuous, non-invasive, and portable approaches. ETCO₂ detection technology faces multiple engineering constraints, including dynamic response, high humidity interference, and system miniaturization, which impose stringent requirements on the overall performance of sensing technologies. This paper, starting from the application scenarios of ETCO₂ detection, systematically analyzes its key technical constraints in terms of concentration range, dynamic characteristics, environmental interference, and engineering implementation. Based on this analysis, a comparative evaluation is conducted of CO₂ gas sensing technologies, including non-dispersive infrared (NDIR), photoacoustic spectroscopy (PAS), tunable diode laser absorption spectroscopy (TDLAS), solid-state gas sensing, and semiconductor sensing₂. Through comprehensive multi-parameter analysis, NDIR technology demonstrates optimal overall suitability for ETCO₂ detection requirements in terms of selectivity, measurement range, system complexity, and engineering feasibility. Furthermore, this paper reviews the research progress of NDIR technology in ETCO₂ detection and concludes that future advancements in ETCO₂ detection can achieve continuous improvements in high precision, low power consumption, and multi-parameter fusion within NDIR-based systems through device optimization, system integration, and data-driven approaches, thereby providing critical technical support for continuous monitoring and accurate assessment of respiratory health.