This study aims to develop an explainable machine learning model that takes into account the characteristics of cardiopulmonary interactions. This model will enable early prediction of acute lung injury (ALI) in patients undergoing major liver surgery. The research will create a digital early-warning system for ALI, thereby supporting clinical diagnosis and treatment decisions. This, in turn, should help reduce the incidence and mortality rates associated with ALI.
Age ≥ 18 years
Undergoing major liver surgery (including two-segment or more hepatectomy, liver transplantation, etc.)
Voluntary participation with signed informed consent
None
* Patient refusal * Presence of comorbidities affecting outcome assessment * Incomplete outcome records * Postoperative follow-up failure * Participation in other studies that may interfere with outcome assessment
Other
Other
Not listed
Occurrence of ALI within 7 Days after Surgery
Berlin Definition: 1. Onset: Acute exacerbation of known injury or new/worsening respiratory symptoms within 1 week. 2. Chest Imaging (X-ray or CT): Bilateral pulmonary shadows not fully explained by exudation, atelectasis, or nodules. 3. Pulmonary Edema Etiology: Respiratory failure not fully attributed to heart failure or fluid overload; if no related risk factors, objective tests (e.g., Doppler echocardiography) are needed to exclude hydrostatic pulmonary edema. 4. Oxygenation Levels: Mild - With CPAP/PEEP \>5 cmH2O, 200 mmHg \< PaO2/FiO2 \< 300 mmHg; Moderate - With CPAP/PEEP \>5 cmH2O, 100 mmHg \< PaO2/FiO2 \< 200 mmHg; Severe - With CPAP/PEEP \>5 cmH2O, PaO2/FiO2 \< 100 mmHg.
Click on the lead sponsor to view all of their trials.
Beijing Tsinghua Chang Gung Hospital
Lead sponsor
Huangdao District People's Hospital of Qingdao
Collaborator
Peking University International Hospital
Collaborator
The First Affiliated Hospital of Army Medical University (Southwest Hospital)
Collaborator
