A two-stage model for predicting postoperative pulmonary infection in esophageal cancer patients_Journal of Biomedical Engineering

Authors：

WANG Yu ¹ , YIN Lixia ² , YAO Bo ¹ ,  WU Jiang ³ ,  PU Jiangbo ¹

1. Institute of Biomedical Engineering, Chinese Academy of Medical Sciences and Peking Union Medical College, Tianjin 300192, P. R. China;
2. Department of Hospital Infection Management, Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, P. R. China;
3. Deep Underground Space Medical Center, West China Hospital, Sichuan University, Chengdu 610041, P. R. China;

Corresponding?author：

WU Jiang, Email: jetterwu@163.com; PU Jiangbo, Email: pujb@bme.pumc.edu.cn

Keywords：

Esophageal cancer; Postoperative pulmonary infection; Machine learning; Risk prediction

DOI：

10.7507/1001-5515.202601010

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

Postoperative pulmonary infection (PPI) after esophageal cancer surgery occurs frequently and severely impairs patients’ prognosis. Most existing prediction models cannot realize staged classification of risk factors, which limits targeted risk identification and intervention. Based on machine learning algorithms, this study integrates preoperative baseline characteristics and perioperative indicators to construct a preoperative-perioperative two-stage risk prediction model for postoperative pulmonary infection. Clinical data of 2 200 patients undergoing esophageal cancer surgery admitted to the Cancer Hospital, Chinese Academy of Medical Sciences between October 2022 and August 2024 were retrospectively enrolled. The least absolute shrinkage and selection operator was combined with multivariate logistic regression to screen independent predictive variables for the two stages, and five machine learning models were established accordingly. Six independent predictive variables were identified. The preoperative predictors included gender, American Society of Anesthesiologists (ASA) physical status classification, and colonization of multidrug-resistant bacteria. All models yielded area under the curve values ranging from 0.71 to 0.72 with a specificity higher than 98%, which can be used for preoperative risk stratification of high-risk individuals. On the basis of preoperative variables, the perioperative stage additionally incorporated operation duration, postoperative intensive care unit (ICU) admission, and peak C-reactive protein level within 0-3 days after surgery, leading to a remarkable improvement in predictive performance with all area under the curve values greater than 0.82. The gradient boosting machine (GBM) model achieved a favorable balance between a sensitivity of 69.07% and a specificity of 82.59%, providing support for risk stratification and clinical management decision-making. Further multicenter studies are required to validate the generalization ability of the model.

Citation： WANG Yu, YIN Lixia, YAO Bo, WU Jiang, PU Jiangbo. A two-stage model for predicting postoperative pulmonary infection in esophageal cancer patients. Journal of Biomedical Engineering, 2026, 43(3): 539-545. doi: 10.7507/1001-5515.202601010 Copy

1.	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. CA Cancer J Clin, 2021, 71(3): 209-249.
2.	Bykowska-Derda A, Darr R, Czlapka-Matyasik M, et al. Short fat-adaptation diet impact on a consecutive day of interval exercise performance. Acta Sci Pol Technol Aliment, 2021, 20(1): 47-54.
3.	Fabian T. Management of postoperative complications after esophageal resection. Surg Clin North Am, 2021, 101(3): 525-539.
4.	Avendano C E, Flume P A, Silvestri G A, et al. Pulmonary complications after esophagectomy. Ann Thorac Surg, 2002, 73(3): 922-926.
5.	Kopp B T, Joseloff E, Goetz D, et al. Urinary metabolomics reveals unique metabolic signatures in infants with cystic fibrosis. J Cyst Fibros, 2019, 18(4): 507-515.
6.	田園, 林志浩, 李瑞, 等. 基于組合優化的機器學習模型預測胃癌術后感染性并發癥的診斷性研究. 中國循證醫學雜志, 2024, 24(9): 993-1003.
7.	Dev D, Wallace E, Sankaran R, et al. Value of C-reactive protein measurements in exacerbations of chronic obstructive pulmonary disease. Respir Med, 1998, 92(4): 664-667.
8.	呂明闖, 曹睿, 丁旭青, 等. 食管癌術后并發肺部感染危險因素和預測模型構建. 中華醫院感染學雜志, 2024, 34(10): 1507-1511.
9.	Li S, Fang C, Tao Z, et al. A nomogram for postoperative pulmonary infections in esophageal cancer patients: a two-center retrospective clinical study. BMC Surg, 2025, 25(1): 70.
10.	Jung J O, Pisula J I, Bozek K, et al. Prediction of postoperative complications after oesophagectomy using machine-learning methods. Br J Surg, 2023, 110(10): 1361-1366.
11.	Goense L, van Dijk W A, Govaert J A, et al. Hospital costs of complications after esophagectomy for cancer. Eur J Surg Oncol, 2017, 43(4): 696-702.
12.	王瑜, 吳江, 伍東升. 卷積神經網絡在職業性塵肺病影像學診斷中的應用研究進展. 生物醫學工程學雜志, 2024, 41(2): 413-420.
13.	Rajkomar A, Dean J, Kohane I. Machine learning in medicine. N Engl J Med, 2019, 380(14): 1347-1358.
14.	詹少強, 曾安, 張逸群, 等. 基于異質屬性融合的危重疾病二階段預測模型. 計算機與現代化, 2024(1): 67-73.
15.	Hassan N, Slight R, Morgan G, et al. Surgical machine learning model for predicting risk of postoperative infections in general elective surgery (SMART): a modelling study. Int J Antimicrob Agents, 2026, 67(6): 107768.
16.	中華人民共和國國家衛生健康委員會醫政醫管局. 食管癌診療指南 (2022年版). 中華消化外科雜志, 2022, 21(12): 1247-1268.
17.	中華人民共和國衛生部. 醫院感染診斷標準 (試行). 北京: 中華醫學雜志, 2001, 81(5): 314-320.
18.	中華人民共和國衛生部. 肺炎診斷: WS 382-2012. 北京: 中華人民共和國衛生部, 2012.
19.	劉欣悅, 姜改英, 王曉娟, 等. 機器學習在腹膜透析并發癥風險預測模型中的研究進展. 中國血液凈化, 2025, 24(10): 834-837, 852.
20.	李旭然, 丁曉紅. 機器學習的五大類別及其主要算法綜述. 軟件導刊, 2019, 18(7): 4-9.
21.	Hu B, Chen X, Chen T, et al. Multimodal machine learning-based marker enables the link between obesity-related indices and future stroke: a prospective cohort study. EClinicalMedicine, 2025, 85: 103331.
22.	Marcy T W, Merrill W W. Cigarette smoking and respiratory tract infection. Clin Chest Med, 1987, 8(3): 381-391.
23.	Li K, Lu S, Li C, et al. Sex-based analysis of clinical outcomes in elderly patients with esophageal squamous cell carcinoma post-esophagectomy: a propensity score matching analysis. Front Oncol, 2025, 15: 1549123.
24.	張樹國, 韓立新. ASA分級和合并癥評估對老年全髖關節置換術后認知功能障礙的預測價值. 中國骨科臨床與基礎研究雜志, 2025, 15(2): 110-116.
25.	Ruzzenente A, Alaimo L, Caputo M, et al. Infectious complications after surgery for perihilar cholangiocarcinoma: a single Western center experience. Surgery, 2022, 172(3): 813-820.
26.	徐英莉, 龐博, 曹姍, 等. 中藥防治耐藥銅綠假單胞菌肺炎感染機制研究進展. 中國藥物警戒, 2023, 20(1): 57-60.
27.	劉力佳, 林倩, 尚威. 肛腸疾病患者術后病原菌分布特點及多重耐藥菌感染預測模型的構建. 鄭州大學學報(醫學版), 2026(2): 151-156.
28.	Alsanad M, Aljanoubi M, Alenezi F K, et al. Preoperative smoking cessation interventions: a systematic review and meta-analysis. Perioper Med, 2025, 14(1): 5.
29.	曹明楠, 王喬宇, 陶驊, 等. 《中國圍手術期感染預防與管理指南》解讀. 臨床藥物治療雜志, 2023, 21(6): 19-25.
30.	Cheng H, Chen B P H, Soleas I M, et al. Prolonged operative duration increases risk of surgical site infections: a systematic review. Surg Infect, 2017, 18(6): 722-735.
31.	Cheng H, Clymer J W, Chen B P H, et al. Prolonged operative duration is associated with complications: a systematic review and meta-analysis. J Surg Res, 2018, 229: 134-144.
32.	Takahashi K, Nishikawa K, Tanishima Y, et al. Risk stratification of postoperative pneumonia in patients undergoing subtotal esophagectomy for esophageal cancer. Anticancer Res, 2022, 42(6): 3023-3028.
33.	Zhang S, Tuerganbayi K, Wang J, et al. Incorporating preoperative and intraoperative data to predict postoperative pneumonia in elderly patients undergoing non-cardiothoracic surgery: the online two-stage prediction tool. Geriatr Nurs, 2025, 62: 244-253.
34.	Papazian L, Klompas M, Luyt C E. Ventilator-associated pneumonia in adults: a narrative review. Intensive Care Med, 2020, 46(5): 888-906.
35.	Cook D J, Walter S D, Cook R J, et al. Incidence of and risk factors for ventilator-associated pneumonia in critically ill patients. Ann Intern Med, 1998, 129(6): 433-440.
36.	Miyamura T, Sakamoto N, Kakugawa T, et al. Postoperative acute exacerbation of interstitial pneumonia in pulmonary and non-pulmonary surgery: a retrospective study. Respir Res, 2019, 20(1): 154.
37.	Kano K, Aoyama T, Nakajima T, et al. Prediction of postoperative inflammatory complications after esophageal cancer surgery based on early changes in the C-reactive protein level. BMC Cancer, 2017, 17(1): 812.
38.	李珊, 李小青, 葉新毅, 等. 食管癌術后肺部感染發生的危險因素分析. 中國胸心血管外科臨床雜志, 2026-04-09.
39.	賈辰, 高巖, 解夕黎, 等. 基于5種機器學習算法構建ICU患者耐碳青霉烯類鮑曼不動桿菌感染風險預測模型. 中華醫院感染學雜志, 2025(17): 2586-2591.
40.	Jin W, Huang K, Ding Z, et al. Global, regional, and national burden of esophageal cancer: a systematic analysis of the Global Burden of Disease Study 2021. Biomark Res, 2025, 13(1): 3.

1. 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. CA Cancer J Clin, 2021, 71(3): 209-249.
2. Bykowska-Derda A, Darr R, Czlapka-Matyasik M, et al. Short fat-adaptation diet impact on a consecutive day of interval exercise performance. Acta Sci Pol Technol Aliment, 2021, 20(1): 47-54.
3. Fabian T. Management of postoperative complications after esophageal resection. Surg Clin North Am, 2021, 101(3): 525-539.
4. Avendano C E, Flume P A, Silvestri G A, et al. Pulmonary complications after esophagectomy. Ann Thorac Surg, 2002, 73(3): 922-926.
5. Kopp B T, Joseloff E, Goetz D, et al. Urinary metabolomics reveals unique metabolic signatures in infants with cystic fibrosis. J Cyst Fibros, 2019, 18(4): 507-515.
6. 田園, 林志浩, 李瑞, 等. 基于組合優化的機器學習模型預測胃癌術后感染性并發癥的診斷性研究. 中國循證醫學雜志, 2024, 24(9): 993-1003.
7. Dev D, Wallace E, Sankaran R, et al. Value of C-reactive protein measurements in exacerbations of chronic obstructive pulmonary disease. Respir Med, 1998, 92(4): 664-667.
8. 呂明闖, 曹睿, 丁旭青, 等. 食管癌術后并發肺部感染危險因素和預測模型構建. 中華醫院感染學雜志, 2024, 34(10): 1507-1511.
9. Li S, Fang C, Tao Z, et al. A nomogram for postoperative pulmonary infections in esophageal cancer patients: a two-center retrospective clinical study. BMC Surg, 2025, 25(1): 70.
10. Jung J O, Pisula J I, Bozek K, et al. Prediction of postoperative complications after oesophagectomy using machine-learning methods. Br J Surg, 2023, 110(10): 1361-1366.
11. Goense L, van Dijk W A, Govaert J A, et al. Hospital costs of complications after esophagectomy for cancer. Eur J Surg Oncol, 2017, 43(4): 696-702.
12. 王瑜, 吳江, 伍東升. 卷積神經網絡在職業性塵肺病影像學診斷中的應用研究進展. 生物醫學工程學雜志, 2024, 41(2): 413-420.
13. Rajkomar A, Dean J, Kohane I. Machine learning in medicine. N Engl J Med, 2019, 380(14): 1347-1358.
14. 詹少強, 曾安, 張逸群, 等. 基于異質屬性融合的危重疾病二階段預測模型. 計算機與現代化, 2024(1): 67-73.
15. Hassan N, Slight R, Morgan G, et al. Surgical machine learning model for predicting risk of postoperative infections in general elective surgery (SMART): a modelling study. Int J Antimicrob Agents, 2026, 67(6): 107768.
16. 中華人民共和國國家衛生健康委員會醫政醫管局. 食管癌診療指南 (2022年版). 中華消化外科雜志, 2022, 21(12): 1247-1268.
17. 中華人民共和國衛生部. 醫院感染診斷標準 (試行). 北京: 中華醫學雜志, 2001, 81(5): 314-320.
18. 中華人民共和國衛生部. 肺炎診斷: WS 382-2012. 北京: 中華人民共和國衛生部, 2012.
19. 劉欣悅, 姜改英, 王曉娟, 等. 機器學習在腹膜透析并發癥風險預測模型中的研究進展. 中國血液凈化, 2025, 24(10): 834-837, 852.
20. 李旭然, 丁曉紅. 機器學習的五大類別及其主要算法綜述. 軟件導刊, 2019, 18(7): 4-9.
21. Hu B, Chen X, Chen T, et al. Multimodal machine learning-based marker enables the link between obesity-related indices and future stroke: a prospective cohort study. EClinicalMedicine, 2025, 85: 103331.
22. Marcy T W, Merrill W W. Cigarette smoking and respiratory tract infection. Clin Chest Med, 1987, 8(3): 381-391.
23. Li K, Lu S, Li C, et al. Sex-based analysis of clinical outcomes in elderly patients with esophageal squamous cell carcinoma post-esophagectomy: a propensity score matching analysis. Front Oncol, 2025, 15: 1549123.
24. 張樹國, 韓立新. ASA分級和合并癥評估對老年全髖關節置換術后認知功能障礙的預測價值. 中國骨科臨床與基礎研究雜志, 2025, 15(2): 110-116.
25. Ruzzenente A, Alaimo L, Caputo M, et al. Infectious complications after surgery for perihilar cholangiocarcinoma: a single Western center experience. Surgery, 2022, 172(3): 813-820.
26. 徐英莉, 龐博, 曹姍, 等. 中藥防治耐藥銅綠假單胞菌肺炎感染機制研究進展. 中國藥物警戒, 2023, 20(1): 57-60.
27. 劉力佳, 林倩, 尚威. 肛腸疾病患者術后病原菌分布特點及多重耐藥菌感染預測模型的構建. 鄭州大學學報(醫學版), 2026(2): 151-156.
28. Alsanad M, Aljanoubi M, Alenezi F K, et al. Preoperative smoking cessation interventions: a systematic review and meta-analysis. Perioper Med, 2025, 14(1): 5.
29. 曹明楠, 王喬宇, 陶驊, 等. 《中國圍手術期感染預防與管理指南》解讀. 臨床藥物治療雜志, 2023, 21(6): 19-25.
30. Cheng H, Chen B P H, Soleas I M, et al. Prolonged operative duration increases risk of surgical site infections: a systematic review. Surg Infect, 2017, 18(6): 722-735.
31. Cheng H, Clymer J W, Chen B P H, et al. Prolonged operative duration is associated with complications: a systematic review and meta-analysis. J Surg Res, 2018, 229: 134-144.
32. Takahashi K, Nishikawa K, Tanishima Y, et al. Risk stratification of postoperative pneumonia in patients undergoing subtotal esophagectomy for esophageal cancer. Anticancer Res, 2022, 42(6): 3023-3028.
33. Zhang S, Tuerganbayi K, Wang J, et al. Incorporating preoperative and intraoperative data to predict postoperative pneumonia in elderly patients undergoing non-cardiothoracic surgery: the online two-stage prediction tool. Geriatr Nurs, 2025, 62: 244-253.
34. Papazian L, Klompas M, Luyt C E. Ventilator-associated pneumonia in adults: a narrative review. Intensive Care Med, 2020, 46(5): 888-906.
35. Cook D J, Walter S D, Cook R J, et al. Incidence of and risk factors for ventilator-associated pneumonia in critically ill patients. Ann Intern Med, 1998, 129(6): 433-440.
36. Miyamura T, Sakamoto N, Kakugawa T, et al. Postoperative acute exacerbation of interstitial pneumonia in pulmonary and non-pulmonary surgery: a retrospective study. Respir Res, 2019, 20(1): 154.
37. Kano K, Aoyama T, Nakajima T, et al. Prediction of postoperative inflammatory complications after esophageal cancer surgery based on early changes in the C-reactive protein level. BMC Cancer, 2017, 17(1): 812.
38. 李珊, 李小青, 葉新毅, 等. 食管癌術后肺部感染發生的危險因素分析. 中國胸心血管外科臨床雜志, 2026-04-09.
39. 賈辰, 高巖, 解夕黎, 等. 基于5種機器學習算法構建ICU患者耐碳青霉烯類鮑曼不動桿菌感染風險預測模型. 中華醫院感染學雜志, 2025(17): 2586-2591.
40. Jin W, Huang K, Ding Z, et al. Global, regional, and national burden of esophageal cancer: a systematic analysis of the Global Burden of Disease Study 2021. Biomark Res, 2025, 13(1): 3.

Journal of Biomedical Engineering

A two-stage model for predicting postoperative pulmonary infection in esophageal cancer patients

Abstract Full text Figures/Tables Video References Cited by

Previous Article

Next Article

Format

Content