Abstract: Sensitive reconstruction of peripheral airway branches is critical for bronchoscopic navigation, yet existing deep learning (DL) methods often struggle with incomplete annotations and ...