Abstract: Early and accurate detection of pathogens in pulmonary infections is crucial for the clinical treatment of severe cases, the prevention of mild infections progressing to severe ones, and for control of widespread transmission of infectious pulmonary diseases. In the past, specimens such as sputum and gastric fluid were collected for etiological testing. When traditional specimens are difficult to obtain, bronchoscopy has gained increasing clinical attention as a method for acquiring more accurate specimens to diagnose lower respiratory tract infections. Following bronchoscopic intervention at the lesion site in the lower respiratory tract or the opening of the lesion's drainage bronchus, diagnostic techniques such as biopsy, needle aspiration, brushing, washing, and lavage can be precisely performed under direct visualization and guidance. The specimens obtained through these sampling methods can then be accurately analyzed for etiology, pathology and molecular biology. Conventional bronchoscopy allows for easy observation and specimen collection from central airway lesions, however, sampling peripheral lesions remains challenging. Accurate localization and sampling of such lesions are currently a hot topic and a significant challenge in clinical research. Advancement of specialized bronchoscopy techniques and navigational bronchoscopy,particularly the exploration and application of AI technology such as robot-assisted bronchoscopy, the era of precise sampling and detection--identifying true infectious pathogens through bronchoscopy—is becoming a reality.

Key words: Lung infection, Bronchoscopy, Artificial Intelligence technology, Etiology, Pathology, Molecular biology