ETT-LDx: transformer-based landmark detection system for endotracheal tube placement verification in chest radiographs

This study examines the use of transformer-based networks with declarative knowledge in assessing endotracheal tube (ETT) placement in chest radiographs.∗ The optimal placement of the ETT is crucial for patient safety, typically residing 5±2 cm above the Carina. Misplacement of an ETT can result in serious complications, highlighting the necessity for precise and prompt verification techniques in intensive care environments. In this work, we propose the ETT-LDx system (Endotracheal Tube Landmark Detection and Assessment System), which utilizes a Transformer-based UNet with declarative knowledge. The system was trained and evaluated on a dataset of 200 anonymized chest X-ray images to determine ETT position effectively. It localizes the tip of the ETT and Carina, ensuring the tube’s placement within a predefined safe zone. Addressing limitations of previous works in the literature, the proposed system is more robust and less prone to error. The initial findings are encouraging and motivate further integration of the ETT-LDx system into clinical practice for real-time use.