The Journal of Practical Medicine >
An interpretable machine learning model based on bedside lung and diaphragm ultrasound for preoperative prediction of pulmonary dysfunction in gastrointestinal tumor surgery： A clinical study
Received date: 2025-10-16
Online published: 2025-12-25
Objective To develop and validate an interpretable machine learning model based on bedside lung and diaphragm ultrasound for preoperative prediction of pulmonary dysfunction in patients undergoing gastrointestinal tumor surgery. Methods In this prospective study， data from 424 patients （June 2021?December 2023） were used for model development， with external validation conducted on an independent cohort of 101 patients （January 2024-December 2024）. Clinical variables， PFTs results， and ultrasound parameters （LUS score， diaphragmatic excursion， and thickening fraction） were collected. Three feature selection methods-Least Absolute Shrinkage and Selection Operator （LASSO）， Support Vector Machine Recursive Feature Elimination （SVM-RFE）， eXtreme Gradient Boosting Recursive Feature Elimination （XGBoost-RFE）- were employed to identify key predictors. Five machine learning algorithms were trained and evaluated using 5-fold cross-validation. The optimal model was assessed based on the area under the receiver operating characteristic curve （AUC）， accuracy， sensitivity， specificity， calibration， and decision curve analysis. SHapley Additive exPlanations （SHAP） analysis was applied to enhance model interpretability. Results The prevalence of preoperative pulmonary dysfunction was 36.8%. Three key predictors were consistently identified： diaphragmatic excursion during deep breathing ［D-DE （4.26 cm vs. 5.05 cm， P < 0.001）］， LUS score ［LUSs （4 vs. 1， P < 0.001）］， and diaphragmatic thickening fraction during deep breathing ［D-DTF （39.83% vs. 71.80%， P < 0.001）］. The Support Vector Machine （SVM） model demonstrated superior generalizability， achieving an AUC of 0.867 on the internal test set （accuracy of 80.0%， sensitivity of 81.8%， specificity of 79.4%）. External validation confirmed its robust performance， with an AUC of 0.934 （95% CI： 0.881 ~ 0.987） and an accuracy of 88.0%. SHAP analysis revealed D-DTF as the most influential protective factor. Conclusion An interpretable SVM model integrating D-DE， LUSs， and D-DTF accurately predicts preoperative pulmonary dysfunction in patients with gastrointestinal tumors， offering a non-invasive， bedside-compatible new approach for preoperative risk assessment.
Tianyuan LI , Ying TIAN , Dingde LONG , Yang DONG , Huan. FU . An interpretable machine learning model based on bedside lung and diaphragm ultrasound for preoperative prediction of pulmonary dysfunction in gastrointestinal tumor surgery： A clinical study[J]. The Journal of Practical Medicine, 2025 , 41(24) : 3929 -3940 . DOI: 10.3969/j.issn.1006-5725.2025.24.018
| [1] | MAO X， ZHANG W， NI Y Q， et al. A Prediction Model for Postoperative Pulmonary Complication in Pulmonary Function-Impaired Patients Following Lung Resection［J］. J Multidiscip Healthc， 2021， 14： 3187-3194. doi:10.2147/jmdh.s327285 |
| [2] | 郭高锋， 阮孝国， 王洋洋， 等. 床旁膈肌超声在预测胸腔镜肺叶切除术后肺部并发症中的应用价值［J］. 实用医学杂志， 2024， 40（2）： 207-212. |
| [3] | FERNANDEZ-BUSTAMANTE A， FRENDL G， SPRUNG J， et al. Postoperative Pulmonary Complications， Early Mortality， and Hospital Stay Following Noncardiothoracic Surgery： A Multicenter Study by the Perioperative Research Network Investigators［J］. JAMA Surg， 2017， 152（2）： 157-166. doi:10.1001/jamasurg.2016.4065 |
| [4] | 胡荣贵， 傅剑华， 罗孔嘉， 等. 术前放化疗对食管癌患者肺功能和术后肺部并发症的影响［J］. 中华胃肠外科杂志， 2013， 16（9）： 827-830. |
| [5] | SUNG H， FERLAY J， SIEGEL R L， et al. Global Cancer Statistics 2020： GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries［J］. CA Cancer J Clin， 2021， 71（3）： 209-249. doi:10.3322/caac.21660 |
| [6] | DOHERTY C， KUBASKI F， TOMATSU S， et al. Non-invasive pulmonary function test on Morquio patients［J］. J Rare Dis Res Treat， 2017， 2（2）： 55-62. doi:10.29245/2572-9411/2017/2.1097 |
| [7] | KAWAMOTO H， KAMBE M. ［Pulmonary function tests］［J］. Rinsho Byori， 2003， 51（6）： 568-573. |
| [8] | BRUCK O， NAOFAL A， SENUSSI M H. Lung， Pleura， and Diaphragm Point-of-Care Ultrasound［J］. Semin Ultrasound CT MR， 2024， 45（1）： 120-131. doi:10.1053/j.sult.2024.01.001 |
| [9] | TENZA-LOZANO E， LLAMAS-ALVAREZ A， JAIMEZ-NAVARRO E， et al. Lung and diaphragm ultrasound as predictors of success in weaning from mechanical ventilation［J］. Crit Ultrasound J， 2018， 10（1）： 12. doi:10.1186/s13089-018-0094-3 |
| [10] | LLAMAS-áLVAREZ A M， TENZA-LOZANO E M， LATOUR-PéREZ J. Diaphragm and Lung Ultrasound to Predict Weaning Outcome： Systematic Review and Meta-Analysis［J］. Chest， 2017， 152（6）： 1140-1150. doi:10.1016/j.chest.2017.08.028 |
| [11] | LI T， WU X Z， LONG D， et al. Ultrasonographic modeling of lung and diaphragm mechanics： Clinical trial of a novel non-invasive method to evaluate pre-operative pulmonary function［J］. PeerJ， 2024， 12： e18677. doi:10.7717/peerj.18677 |
| [12] | 中华医学会超声医学分会. 肺部超声临床应用专家共识［J］. 中华超声影像学杂志， 2022， 31（9）： 757-768. |
| [13] | DANISH M， AGARWAL A， GOYAL P， et al. Diagnostic Performance of 6-Point Lung Ultrasound in ICU Patients： A Comparison with Chest X-Ray and CT Thorax［J］. Turk J Anaesthesiol Reanim， 2019， 47（4）： 307-319. doi:10.5152/tjar.2019.73603 |
| [14] | KONG S， WANG J， LI Y， et al. Value of Bedside Lung Ultrasound in Severe and Critical COVID-19 Pneumonia［J］. Respir Care， 2021， 66（6）： 920-927. doi:10.4187/respcare.08382 |
| [15] | ZAMBON M， GRECO M， BOCCHINO S， et al. Assessment of diaphragmatic dysfunction in the critically ill patient with ultrasound： A systematic review［J］. Intensive Care Med， 2017， 43（1）： 29-38. doi:10.1007/s00134-016-4524-z |
| [16] | MOINGEON P， KUENEMANN M， GUEDJ M. Artificial intelligence-enhanced drug design and development： Toward a computational precision medicine［J］. Drug Discov Today， 2022， 27（1）： 215-222. doi:10.1016/j.drudis.2021.09.006 |
| [17] | BIFARIN O O. Interpretable machine learning with tree-based shapley additive explanations： Application to metabolomics datasets for binary classification［J］. PLoS One， 2023， 18（5）： e0284315. doi:10.1371/journal.pone.0284315 |
| [18] | GUO Z， WANG P， YE S， et al. Interpretable Machine Learning Models Based on Shapley Additive Explanations for Predicting the Risk of Cerebrospinal Fluid Leakage in Lumbar Fusion Surgery［J］. Spine （Phila Pa 1976）， 2024， 49（18）： 1281-1293. doi:10.1097/brs.0000000000005087 |
| [19] | ZARGARI MARANDI R. ExplaineR： An R package to explain machine learning models［J］. Bioinform Adv， 2024， 4（1）： vbae049. doi:10.1093/bioadv/vbae049 |
| [20] | BRUSASCO V， CRAPO R， VIEGI G， et al. Coming together： The ATS/ERS consensus on clinical pulmonary function testing［J］. Eur Respir J， 2005， 26（1）： 1-2. doi:10.1183/09031936.05.00034205 |
| [21] | BALL L， ROCCO P R M， PELOSI P. Editorial： Lung Imaging in Respiratory Failure［J］. Front Physiol， 2022， 13： 862647. doi:10.3389/fphys.2022.862647 |
| [22] | DESANTI R L， GILL K G， SWANSON J O， et al. Comparison of chest radiograph and lung ultrasound in children with acute respiratory failure［J］. J Ultrasound， 2023， 26（4）： 861-870. doi:10.1007/s40477-023-00827-y |
| [23] | 路子蕴， 孙行， 徐璐， 等. 日间胸腔镜手术患者术后肺部并发症危险因素分析［J］. 实用医学杂志， 2023， 39（24）： 3205-3209. |
| [24] | QIAN Z， YANG M， LI L， et al. Ultrasound assessment of diaphragmatic dysfunction as a predictor of weaning outcome from mechanical ventilation： A systematic review and meta-analysis［J］. BMJ Open， 2018， 8（9）： e021189. doi:10.1136/bmjopen-2017-021189 |
| [25] | XU Q， YANG X， QIAN Y， et al. Speckle tracking quantification parasternal intercostal muscle longitudinal strain to predict weaning outcomes： A multicentric observational study［J］. Shock， 2023， 59（1）： 66-73. doi:10.1097/shk.0000000000002044 |
| [26] | PODDIGHE D， VAN HOLLEBEKE M， CHOUDHARY Y Q， et al. Accuracy of respiratory muscle assessments to predict weaning outcomes： A systematic review and comparative meta-analysis［J］. Crit Care， 2024， 28（1）： 70. doi:10.1186/s13054-024-04823-4 |
| [27] | STAHR C S， SAMARAGE C R， DONNELLEY M， et al. Quantification of heterogeneity in lung disease with image-based pulmonary function testing［J］. Sci Rep， 2016， 6： 29438. doi:10.1038/srep29438 |
| [28] | STENWIG E， SALVI G， ROSSI P S， et al. Comparative analysis of explainable machine learning prediction models for hospital mortality［J］. BMC Med Res Methodol， 2022， 22（1）： 53. doi:10.1186/s12874-022-01540-w |
| [29] | ABBAS S， QAISAR A， FAROOQ M S， et al. Smart Vision Transparency： Efficient Ocular Disease Prediction Model Using Explainable Artificial Intelligence［J］. Sensors （Basel）， 2024， 24（20）： 6618. doi:10.3390/s24206618 |
| [30] | BRANKOVIC A， HUANG W， COOK D， et al. Elucidating Discrepancy in Explanations of Predictive Models Developed Using EMR［J］. Stud Health Technol Inform， 2023， 310： 865-869. doi:10.3233/shti231088 |
/
| 〈 |
|
〉 |