A machine learning model based on combined clinical features and serological indicators for predicting persistent organ failure in acute pancreatitis

doi:10.3877/cma.j.issn.1674-3946.2026.06.015

Abstract

Abstract:

Objective

To construct a predictive model for acute pancreatitis (AP) secondary to persistent organ failure (POF) based on clinical characteristics and serological indicators using machine learning.

Methods

A total of 350 AP patients from October 2019 to December 2024 were retrospectively collected and divided into a training set (n=245) and an internal validation set (n=105) in a 7∶3 ratio. Another 82 patients from Xincheng Hospital during the same period were included as the test set. Quantitative data with normal distribution were described as (

x ¯

±s) and analyzed using independent sample t tests; count data were described as [cases (%)] and analyzed using the χ² test. Variables were selected using LASSO regression and models of logistic regression (LR), random forest (RF), and support vector machine (SVM) were constructed to optimize the model using the area under the receiver operating characteristic curve (AUC). The generalization ability of the models was verified using the test set.

Results

The AP secondary POF-related variables selected by LASSO regression included body mass index, Acute Physiology and Chronic Health Evaluation Ⅱ (APACHE Ⅱ) score, random blood glucose, interleukin 6 (IL-6), C-reactive protein (CRP), lactate dehydrogenase (LDH), and neutrophil count/lymphocyte count (NLR). The AUC of the LR, RF, and SVM models based on clinical characteristics were 0.700, 0.862, and 0.707, respectively; the AUC of the LR, RF, and SVM models based on clinical characteristics and serological indicators were 0.805, 0.904, and 0.840, respectively. The AUC of the serological indicators-based models in the three machine learning models were all better than those of the clinical characteristics alone (P < 0.05). The AUC of the RF model based on clinical characteristics and serological indicators in the internal validation set for predicting AP secondary POF was 0.899. The ROC curve was plotted using the test set data, and the results showed that the AUC value of the RF model based on clinical characteristics and serological indicators for predicting AP secondary POF was 0.896, with a sensitivity of 80.7%, specificity of 87.8%, accuracy of 85.7%, precision of 73.5%, and F₁ of 0.831.

Conclusion

The RF model based on the combination of clinical characteristics (body mass index, APACHE II score, random blood glucose) and serological indicators has excellent performance and reliable generalization ability for predicting AP secondary POF, and can be used for early identification of high-risk patients and risk stratification.

Key words: Acute Pancreatitis, Persistent Organ Failure, Clinical Features, Serological Indicators, Machine Learning Model

Dong Li, Liefeng Dong, Wenyue Zhao, Zhenqing Song. A machine learning model based on combined clinical features and serological indicators for predicting persistent organ failure in acute pancreatitis[J]. Chinese Journal of Operative Procedures of General Surgery(Electronic Edition), 2026, 20(06): 576-581.

/ / Recommend

Add to citation manager EndNote|Ris|BibTeX

URL: https://zhpwkssxzz.cma-cmc.com.cn/EN/10.3877/cma.j.issn.1674-3946.2026.06.015

https://zhpwkssxzz.cma-cmc.com.cn/EN/Y2026/V20/I06/576

Figures/Tables 10

因素	训练集（n=245）	内部验证集（n=105）	测试集（n=82）	统计值	P值
性别[例（%）]				χ²=0.528	0.768
男	142(58.0)	60(57.1)	48(58.5)
女	103(42.0)	45(42.9)	34(41.5)
年龄（岁， $x ¯$ ±s）	50.7±9.4	51.3±9.7	50.9±9.5	F=0.125	0.882
身体质量指数（kg/m²， $x ¯$ ±s）	23.4±2.2	23.3±2.2	23.4±2.2	F=0.098	0.907
NLR（ $x ¯$ ±s）	7.3±2.9	7.2±2.9	7.3±2.9	F=0.089	0.915
LDH（U/L, $x ¯$ ±s）	545.0±184.4	552.2±182.7	548.7±183.6	F=0.085	0.918
MMP-9（μg/L， $x ¯$ ±s）	534.5±113.4	550.4±112.3	542.4±112.8	F=0.102	0.903
IL-6（pg/L， $x ¯$ ±s）	15.7±3.6	16.1±3.7	15.9±3.7	F=0.091	0.913
CRP（mg/L， $x ¯$ ±s）	71.9±8.9	70.6±8.5	71.3±8.8	F=0.115	0.892
APACHEⅡ评分（分， $x ¯$ ±s）	10.5±2.9	10.4±2.8	10.5±2.8	F=0.056	0.945
Ranson评分[分，M（Q₁，Q₃）]	1(0,3)	1(0,3)	1(0,3)	H=0.218	0.827
随机血糖（mmol/L， $x ¯$ ±s）	9.9±1.7	9.8±1.7	9.9±1.7	F=0.215	0.807
AST（U/L， $x ¯$ ±s）	26.2±3.4	26.1±3.3	26.2±3.4	F=0.082	0.921
BUN（mmol/L， $x ¯$ ±s）	5.1±1.1	5.1±1.1	5.1±1.1	F=0.058	0.943
Cr（μmol/L， $x ¯$ ±s）	80.8±12.2	78.9±11.5	79.5±12.0	F=1.035	0.356
TG（mmol/L， $x ¯$ ±s）	4.6±1.6	4.5±1.5	4.5±1.5	F=0.074	0.929
LDL-C（mmol/L， $x ¯$ ±s）	3.6±0.6	3.5±0.6	3.6±0.6	F=0.091	0.931
PCT（μg/L， $x ¯$ ±s）	1.9±0.5	1.9±0.5	1.9±0.5	F=0.210	0.811
TNF-α（μg/L， $x ¯$ ±s）	30.9±4.9	31.5±5.0	31.2±4.9	F=0.105	0.901

因素	训练集（n=245）	内部验证集（n=105）	测试集（n=82）	统计值	P值
性别[例（%）]				χ²=0.528	0.768
男	142(58.0)	60(57.1)	48(58.5)
女	103(42.0)	45(42.9)	34(41.5)
年龄（岁， $x ¯$ ±s）	50.7±9.4	51.3±9.7	50.9±9.5	F=0.125	0.882
身体质量指数（kg/m²， $x ¯$ ±s）	23.4±2.2	23.3±2.2	23.4±2.2	F=0.098	0.907
NLR（ $x ¯$ ±s）	7.3±2.9	7.2±2.9	7.3±2.9	F=0.089	0.915
LDH（U/L, $x ¯$ ±s）	545.0±184.4	552.2±182.7	548.7±183.6	F=0.085	0.918
MMP-9（μg/L， $x ¯$ ±s）	534.5±113.4	550.4±112.3	542.4±112.8	F=0.102	0.903
IL-6（pg/L， $x ¯$ ±s）	15.7±3.6	16.1±3.7	15.9±3.7	F=0.091	0.913
CRP（mg/L， $x ¯$ ±s）	71.9±8.9	70.6±8.5	71.3±8.8	F=0.115	0.892
APACHEⅡ评分（分， $x ¯$ ±s）	10.5±2.9	10.4±2.8	10.5±2.8	F=0.056	0.945
Ranson评分[分，M（Q₁，Q₃）]	1(0,3)	1(0,3)	1(0,3)	H=0.218	0.827
随机血糖（mmol/L， $x ¯$ ±s）	9.9±1.7	9.8±1.7	9.9±1.7	F=0.215	0.807
AST（U/L， $x ¯$ ±s）	26.2±3.4	26.1±3.3	26.2±3.4	F=0.082	0.921
BUN（mmol/L， $x ¯$ ±s）	5.1±1.1	5.1±1.1	5.1±1.1	F=0.058	0.943
Cr（μmol/L， $x ¯$ ±s）	80.8±12.2	78.9±11.5	79.5±12.0	F=1.035	0.356
TG（mmol/L， $x ¯$ ±s）	4.6±1.6	4.5±1.5	4.5±1.5	F=0.074	0.929
LDL-C（mmol/L， $x ¯$ ±s）	3.6±0.6	3.5±0.6	3.6±0.6	F=0.091	0.931
PCT（μg/L， $x ¯$ ±s）	1.9±0.5	1.9±0.5	1.9±0.5	F=0.210	0.811
TNF-α（μg/L， $x ¯$ ±s）	30.9±4.9	31.5±5.0	31.2±4.9	F=0.105	0.901

因素	POF组（n=75）	Non-POF组（n=170）	t值	P值
身体质量指数（kg/m²）	23.9±2.1	23.1±2.2	2.861	0.005
APACHEⅡ评分（分）	11.6±3.0	10.0±2.7	4.108	<0.001
随机血糖（mmol/L）	10.5±1.8	9.7±1.6	3.432	0.001
IL-6（pg/L）	17.0±3.6	15.1±3.4	4.148	<0.001
LDH（U/L）	602.4±186.4	519.7±178.2	3.302	0.001
NLR	10.3±3.2	5.1±2.1	11.256	<0.001
MMP-9（μg/L）	570.3±118.1	518.7±107.9	3.355	0.001
AST（U/L）	27.0±3.2	25.9±3.5	2.466	0.014
BUN（mmol/L）	5.4±1.0	4.9±1.1	2.547	0.012
Cr（μmol/L）	84.1±11.9	79.3±12.1	2.854	0.005
TG（mmol/L）	5.0±1.4	4.4±1.6	2.843	0.005
LDL-C（mmol/L）	3.4±0.5	3.7±0.6	3.293	0.001
PCT（μg/L）	1.9±0.5	1.8±0.5	2.408	0.017
TNF-α（μg/L）	32.1±4.7	30.4±4.9	2.545	0.012

变量	模型	AUC（95%置信区间）	灵敏度（%）	特异度（%）	准确度（%）	精确率（%）	F₁值
临床特征	LR	0.700（0.638~0.756）	80.0	50.6	59.6	41.7	0.683
	RF	0.862（0.812~0.903）	92.0	65.3	73.5	53.9	0.817
	SVM	0.707（0.646~0.764）	73.3	60.0	64.1	44.7	0.684
临床特征-血清学指标	LR	0.805（0.750~0.853）	72.0	81.8	78.8	63.5	0.752
	RF	0.904（0.881~0.952）	84.0	88.8	87.3	76.8	0.856
	SVM	0.840（0.788~0.883）	74.7	81.8	79.6	64.4	0.771

References 18

［1］	Tang J，Chen T，Ni W，et al. Dynamic nomogram for persistent organ failure in acute biliary pancreatitis: Development and validation in a retrospective study［J］. Dig Liver Dis，2022，54（6）: 805-811.
［2］	Mahapatra SJ，Garg PK. Organ Failure and Prediction of Severity in Acute Pancreatitis［J］. Gastroenterol Clin North Am，2025，54（1）: 1-19.
［3］	中华医学会急诊医学分会，上海市医学会急诊专科分会. 急性胰腺炎急诊诊治专家共识［J］. 中华急诊医学杂志，2024，33（4）: 470-479.
［4］	Pandey R，Pathak R，Jha A，et al. Comparison of Acute Physiology and Chronic Health Evaluation Ⅱ，Bedside Index for Severity in Acute Pancreatitis and Modified Computed Tomography Severity Index Scores in Predicting the Outcome in Acute Pancreatitis in a Tertiary Care Centre in Nepal［J］. J Nepal Health Res Counc，2023，21（2）: 203-206.
［5］	Capurso G，Ponz de Leon Pisani R，Lauri G，et al. Clinical usefulness of scoring systems to predict severe acute pancreatitis: A systematic review and meta-analysis with pre and post-test probability assessment［J］. United European Gastroenterol J，2023，11（9）: 825-836.
［6］	中华医学会急诊分会，京津冀急诊急救联盟，北京医学会急诊分会，等. 急性胰腺炎急诊诊断及治疗专家共识［J］. 临床肝胆病杂志，2021，37（5）: 1034-1041.
［7］	Leppäniemi A，Tolonen M，Tarasconi A，et al. 2023 WSES guidelines for the management of severe acute pancreatitis［J］. World J Emerg Surg，2023，18（1）: 38.
［8］	Machicado JD，Gougol A，Tan X，et al. Mortality in acute pancreatitis with persistent organ failure is determined by the number，type，and sequence of organ systems affected［J］. United European Gastroenterol J，2021，9（2）: 139-149.
［9］	刘德仁，刘加宁，苏峰. 入院血清肝素结合蛋白水平与急性胰腺炎持续器官衰竭的相关研究［J］. 现代消化及介入诊疗，2021，26（10）: 1225-1229.
［10］	Zhang Y，Peng W，Zheng X. The prognostic value of the combined neutrophil-to-lymphocyte ratio（NLR）and neutrophil-to-platelet ratio（NPR）in sepsis［J］. Sci Rep，2024，14（1）: 15075-15084.
［11］	陈系东，王绍闯，赵何伟，等. 高龄BCLC B期肝癌患者常规TACE术后急性肝功能恶化的危险因素研究［J/OL］. 中华普外科手术学杂志（电子版），2025，19（6）: 689-692.
［12］	Zhou J，Yang R，Zhu W，et al. NLR inflammasome pathways: key targets for pathogenesis and therapy of metabolic diseases［J］. Front Immunol，2026，17（1）: 1823724-1823749.
［13］	王迎芝，张先进，杨士彦. 基于Logistic回归分析急性胰腺炎并发持续性器官功能衰竭的因素［J］. 肝胆外科杂志，2022，30（5）: 365-369.
［14］	中华医学会外科学分会胰腺外科学组. 中国急性胰腺炎诊治指南（2021）［J］. 中华消化外科杂志，2021，20（7）: 730-739.
［15］	张洁，罗小霞，余鸿. 系统性免疫炎症指数对急性胰腺炎患者并发器官功能损伤的预测价值［J/OL］. 中华普外科手术学杂志（电子版），2025，19（1）: 68-71.
［16］	Gu K，Wang Q. Establishment and Validation of a dynamic nomogram for persistent organ failure in acute biliary pancreatitis: a retrospective study［J］. J Inflamm Res，2024，17（1）: 8513-8530.
［17］	Guo P，Liu L，Yang X，et al. Irisin improves BBB dysfunction in SAP rats by inhibiting MMP-9 via the ERK/NF-κB signaling pathway［J］. Cell Signal，2022，93（1）: 110300.
［18］	罗竹. 基于机器学习的随机森林模型预测重症急性胰腺炎的价值［D］. 温州医科大学，2022.

[1]	Wenxia Bian, Dongya Huang. Association between early serum anion gap and delirium risk in ICU patients with acute pancreatitis [J]. Chinese Journal of Critical Care Medicine(Electronic Edition), 2026, 19(04): 315-321.
[2]	Chuanyu Liu, Jianguo Hu, Li Wang, Xinyue Xu, Shuai Diao, Qian Yang, Lina Hu. Multicenter retrospective clinical analysis of special types of cervical adenocarcinoma [J]. Chinese Journal of Obstetrics & Gynecology and Pediatrics(Electronic Edition), 2026, 22(02): 128-139.
[3]	Mei Liu, Yonghui Yang, Guiyun Zhu, Weili Ma, Yunxiao Han. Correlation between the expression levels of chemokine CXC chemokine receptor 3, the ligands and clinical characteristics and lesion severity of patients with pulmonary tuberculosis [J]. Chinese Journal of Experimental and Clinical Infectious Diseases(Electronic Edition), 2026, 20(03): 155-166.
[4]	Weifeng Wang, Guifang Zha, Maoyue Yuan, Yixin Mou, Qing Yang. Clinical characteristics of 567 children with pertussis [J]. Chinese Journal of Experimental and Clinical Infectious Diseases(Electronic Edition), 2026, 20(03): 182-187.
[5]	Xin Zhao, Wenqiang Li, Ying Yang. Diagnostic value and clinical characteristics of metagenomic nextgeneration sequencing in nontuberculous Mycobacterial pulmonary disease [J]. Chinese Journal of Lung Diseases(Electronic Edition), 2026, 19(04): 632-637.
[6]	Jiayao Wang, Kaiting Ye, Qingqing Zhu, Liming Fei, Jiahui Dong. Retrospective analysis of clinical characteristics and prognosis of 54 cases with severe Chlamydia psittaci pneumonia [J]. Chinese Journal of Lung Diseases(Electronic Edition), 2026, 19(03): 444-450.
[7]	Yanan Tang, Xiaoliang Liu, Xiuying Ma, Lushan Zhao, Liying Wang, Pengxiang Zhou, Xiaoyan Gai, Yan Xing. Retrospective analysis of clinical features and current status of diagnosis and treatment in 577 children with community-acquired pneumonia [J]. Chinese Journal of Lung Diseases(Electronic Edition), 2026, 19(02): 303-308.
[8]	Donglian Du, Zhiwei Shi, Jie Yao, Shumin Zhang, Weibin Dai. Clinical significance of predicting the growth of solitary pulmonary nodules based on CT radiomics and clinical features [J]. Chinese Journal of Lung Diseases(Electronic Edition), 2026, 19(01): 56-61.
[9]	Hao Wang, Weijie Zhang, Baihui Yuan, Zhiyuan Song, Yanlong Yu. Meta-analysis of triglyceride glucose index in predicting the severity of acute pancreatitis [J]. Chinese Journal of Hepatic Surgery(Electronic Edition), 2026, 15(03): 405-412.
[10]	Ning Yang, Li’ao Zhang, Renjun Li, Shui Jin. Clinical characteristics and analysis of factors influencing polyp recurrence in patients with non-hereditary colorectal polyposis [J]. Chinese Journal of Colorectal Diseases(Electronic Edition), 2026, 15(02): 168-174.
[11]	Dongya Huang, Chaoqun Hou, Qiang Li. Impact of the red blood cell distribution width-to-albumin ratio on in-hospital mortality risk in patients with acute pancreatitis [J]. Chinese Journal of Critical Care & Intensive Care Medicine(Electronic Edition), 2026, 12(02): 147-153.
[12]	Yi Lu, Guoxing Wang, Miaorong Xie. Role of the "First-aid Philosophy" in the management of severe acute pancreatitis [J]. Chinese Journal of Critical Care & Intensive Care Medicine(Electronic Edition), 2026, 12(02): 134-140.
[13]	Zhendong Fang, Weipeng Huang, Yifan Gao, Hong Yu, Bo Shen, Feng Guo. Prognosis of patients with infected pancreatic necrosis undergoing laparoscopic retroperitoneal pancreatic necrosectomy [J]. Chinese Journal of Critical Care & Intensive Care Medicine(Electronic Edition), 2026, 12(02): 141-146.
[14]	Weijie Ren, Ni Mao, Ningdan Hu, Feng Shen. KLF4 activates NLRP3 inflammasomes in AR42J cells by modulating TXNIP-mediated mitochondrial oxidative stress [J]. Chinese Journal of Digestion and Medical Imageology(Electronic Edition), 2026, 16(04): 294-301.
[15]	Yimei Han, Shichuan Feng, Zhijuan Chen. Risk factors for recurrence of hyperlipidemic acute pancreatitis and construction of a nomogram prediction model [J]. Chinese Journal of Digestion and Medical Imageology(Electronic Edition), 2026, 16(03): 222-228.

Please choose a citation manager

Content to export