Assessing The Impact Of CO₂ Exposure On Pneumonia Detection Using A Spatially-Aware Hybrid CNN-Transformer Model

The increasing level of atmospheric CO 2 is not only a concern as a driver of climate change, but is also one of the major factors influencing air quality and human respiratory health. Because carbon dioxide content in the atmosphere is rising, we might expect respiratory infections, including pneumonia, to become even more prevalent and more severe due to compromised pulmonary defenses and increased environmental challenges. In this paper, we develop a spatially-aware hybrid CNN-Transformer for pneumonia detection from chest X-ray (CXR) images,as a context of the larger environmental effect of exposure to CO₂. The model combines localized spatial feature with Convolutional Neural Networks (CNN).Performance on an experimental evaluation shows that the proposed model clearly beats state of the art conventional CNN based methods both in terms of accuracy and recall and in terms of the localization precision. Aside from the algorithmic enhancement, the work highlights the importance of integration between environmental science and medical diagnosis. The authors further point to how the escalation of CO₂ burden and resultant degradation in air-quality have potential for exacerbating risks of pneumonia, especially within high-risk groups, and confirms the demand for smart diagnostics. This work offers a path toward incorporating environmental sensing into deep learning-enabled health applications for sustainable respiratory care.