社区所有版块导航
Python
python开源   Django   Python   DjangoApp   pycharm  
DATA
docker   Elasticsearch  
aigc
aigc   chatgpt  
WEB开发
linux   MongoDB   Redis   DATABASE   NGINX   其他Web框架   web工具   zookeeper   tornado   NoSql   Bootstrap   js   peewee   Git   bottle   IE   MQ   Jquery  
机器学习
机器学习算法  
Python88.com
反馈   公告   社区推广  
产品
短视频  
印度
印度  
Py学习  »  机器学习算法

周一“星”视角|脑死亡VS.循环死亡供体肺移植的短期及长期预后;机器学习预测肺移植后原发性移植肺失功

胸外学术时间 • 1 周前 • 40 次点击  



本期胸小星将为大家带来脑死亡VS.循环死亡供体肺移植的短期及长期预后;机器学习预测肺移植后原发性移植肺失功,一起来看看吧!

2017·EATTS 

01

Short and Long-Term Outcomes of Lung Transplantation from Brain Death vs. Circulatory Death Donors: A Meta-analysis of Comparative Studies 

C. Spadaccio1, A. Salsano2, S. Altarabsheh1, A. Castro-Varela1, C. Gallego Navarro1, F. Juarez Casso1, A Abdelrehim1, K Andi1, RVP Ribeiro3, K Choi1, G Knop1, C. C. Kennedy4, K.M. Pennington4, PJ Spencer1, R Daly1, M Villavicencio1, Marcelo Cypel5, SA Saddoughi1,3 

1 Department of Cardiovascular Surgery, Mayo Clinic, Rochester, MN 

2 DISC Department, University of Genoa, Italy 

3 Division of Thoracic Surgery, Department of Surgery, Mayo Clinic, Rochester, MN 

4 Division of Pulmonary and Critical Care Medicine, Mayo Clinic, Rochester, MN 

5 Department of Thoracic Surgery, Toronto General Hospital, Toronto, ON, Canada 


Objectives: 

To investigate through a meta-analysis of comparative studies the impact of donor type (brain death DBD vs circulatory death DCD) on the short- and long-term outcomes of lung transplantation (LTx). 


Methods: 

Literature search (terms “lung transplantation” AND “donation after circulatory death”) was performed up to July 2022 and studies comparing outcomes of LTx from DCD versus DBD were selected. Primary endpoints were early and long-term mortality. Secondary outcomes included primary graft dysfunction (PGD), acute rejection and postoperative complications. The long-term survival was analyzed by retrieving data from each available Kaplan-Meier and restricted mean survival time difference between DBD and DCD for long-term survival was estimated.


Results: 

21 studies were included comprising 60105 patients (DBD=58548 DCD=1557). Recipient and donor baseline characteristics were similar between the two groups. No significant publication bias was observed. The estimated pooled odds ratio of early mortality favored DBD (OR=0.75, CI=0.56-1.00, I2=0%). No statistically significant difference was observed in the risk of acute rejection (OR=1.33, CI=0.82-2.17), and PGD grade 2-3 (OR=0.88, CI=0.69-1.13). One- and 5 year survival were 82.1% and 51.2%, and 86.2% and 62.7% for DBD and DCD groups, respectively (Log-rank,P<0.0001). Unadjusted hazard ratio was 0.693, with DCD as reference. DCD lungs demonstrated improved survival by 4.82% over 5-years when compared to DBD lungs.


Conclusions: 

This meta-analysis of comparative studies between DCD and DBD demonstrates significant long-term survival advantage of DCD LTx despite an initial small but statistically significant increased mortality risk in the short-term. Data supports the continued implementation of DCD to increase the lung donor pool.


[CITATION]: C. Spadaccio, A. Salsano, S. Altarabsheh, et al.Short and Long-Term Outcomes of Lung Transplantation from Brain Death vs. Circulatory Death Donors: A Meta-analysis of Comparative StudiesRunning title: DCD vs DBD lung transplantation meta-analysis, Journal of Heart and Lung Transplantation, (2024)

[DOI]: 10.1016/j.healun.2024.12.010                

[IF]: 6.4

向下滑动查看所有内容 


脑死亡VS.循环死亡供体肺移植的短期及长期预后:基于比较研究的荟萃分析

胸“星”外科学术团队成员 林恒涛 译


目的



通过对比较研究的荟萃分析,探讨供体类型(脑死亡供肺(brain death, DBD) vs. 循环死亡供肺(circulatory death, DCD))对肺移植(lung transplantation, LTx)短期和长期预后的影响。

方法




检索截至2022年7月的文献,检索词为“lung transplantation” AND “donation after circulatory death”,并筛选出比较DCD与DBD供肺移植结局的研究。主要终点为早期和长期死亡率,次要结局指标包括原发性移植肺失功(primary graft dysfunction, PGD)、急性排斥反应及术后并发症。通过检索每项可用的Kaplan-Meier生存曲线数据,分析长期生存情况,并估算DBD与DCD之间长期生存的限制平均生存时间差异。

结果


共纳入21项研究,涵盖60105例患者(DBD组:58548例,DCD组:1557例)。两组受者及供者的基线特征相似,未发现显著的发表偏倚。早期死亡率的汇总比值比(odds ratio, OR)为0.75(95%CI:0.56-1.00,I²=0%),提示DBD在短期内具有一定的生存优势。急性排斥反应(OR=1.33,95%CI:0.82-2.17)和PGD 2-3级发生率(OR=0.88,95%CI:0.69-1.13)在两组间无显著差异。DCD与DBD供肺移植的一年生存率为86.2% vs. 82.1%,五年生存率为62.7% vs. 51.2%(Log-rank检验,P<0.0001)。未校正的风险比(HR)为0.693(以DCD为参考)。与DBD供体肺相比,DCD供体肺在5年内的生存率提高了4.82%。

结论

本项荟萃分析比较了DCD与DBD供肺移植的预后,结果显示尽管DCD供肺移植在短期内存在轻微但具有统计学意义的死亡风险增加,但其长期生存优势显著。本研究结果支持继续推广DCD供肺移植,以扩大肺移植供体库。

Figure 3. Pooled survival curves derived from reconstructed individual patient data. 

2017·EATTS 

02

Machine Learning for Predicting Primary Graft Dysfunction After Lung Transplantation: An Interpretable Model Study

Wei Xia1, Weici Liu2, Zhao He2, Chenghu Song2, Jiwei Liu2, Ruo Chen2, Jingyu Chen3, Xiaokun Wang2, Hongyang Xu1, Wenjun Mao2

1 Department of Intensive Care Unit, The Affiliated Wuxi People’s Hospital of Nanjing Medical University, Wuxi People’s Hospital, Wuxi Medical Center, Nanjing Medical University, Wuxi, Jiangsu, China. 

2 Department of Thoracic Surgery, The Affiliated Wuxi People’s Hospital of Nanjing Medical University, Wuxi People’s Hospital, Wuxi Medical Center, Nanjing Medical University, Wuxi, Jiangsu, China. 

3 Department of Lung Transplantation, The Affiliated Wuxi People’s Hospital of Nanjing Medical University, Wuxi People’s Hospital, Wuxi Medical Center, Nanjing Medical University, Wuxi, Jiangsu, China.


Background: 

Primary graft dysfunction (PGD) develops within 72h after lung transplantation (Lung Tx) and greatly influences patients’ prognosis. This study aimed to establish an accurate machine learning (ML) model for predicting grade 3 PGD (PGD3) after Lung Tx. 


Methods: 

This retrospective study incorporated 802 patients receiving Lung Tx between July 2018 and October 2023 (640 in the derivation cohort and 162 in the external validation cohort), and 640 patients were randomly assigned to training and internal validation cohorts in a 7:3 ratio. Independent risk factors for PGD3 were determined by integrating the univariate logistic regression and least absolute shrinkage and selection operator regression analyses. Subsequently, 9 ML models were used to construct prediction models for PGD3 based on selected variables. Their prediction performances were further evaluated. Besides, model stratification performance was assessed with 3 posttransplant metrics. Finally, the SHapley Additive exPlanations algorithm was used to understand the predictive importance of selected variables. 


Results: 

We identified 9 independent clinical risk factors as selected variables. Among 9 ML models, the random forest (RF) model displayed optimal performance (area under the curve [AUC] = 0.9415, sensitivity [Se] = 0.8972, specificity [Sp] = 0.8795 in the training cohort; AUC = 0.7975, Se = 0.7520, Sp = 0.7313 in the internal validation cohort; and AUC = 0.8214, Se = 0.8235, Sp = 0.6667 in the external validation cohort). Further assessments on calibration and clinical usefulness indicated the promising applicability of the RF model in PGD3 prediction. Meanwhile, the RF model also performed best in terms of risk stratification for postoperative support (extracorporeal membrane oxygenation time: P < 0.001, mechanical ventilation time: P = 0.006, intensive care unit time: P < 0.001). 


Conclusions: 

The RF model had the optimal performance in PGD3 prediction and postoperative risk stratification for patients after Lung Tx.


[CITATION]: Xia W, Liu W, He Z, et al. Machine Learning for Predicting Primary Graft Dysfunction After Lung Transplantation: An Interpretable Model Study. Transplantation. 2025 Jan 10.   

[DOI]: 10.1097/TP.0000000000005326

[IF]: 5.3                                                       

向下滑动查看所有内容


机器学习预测肺移植后原发性移植肺失功:一项可解释性模型研究

胸“星”外科学术团队成员 古明宇 

背景



原发性移植肺失功(primary graft dysfunction, PGD)在肺移植(lung transplantation, Lung Tx)后72小时内发生,对患者预后有较大影响。本研究旨在建立一个准确的机器学习(machine learning, ML)模型,用于预测Lung Tx后3级PGD(grade 3 PGD, PGD3)的发生。

方法


本研究为一项回顾性研究,纳入了2018年7月至2023年10月期间接受肺移植的802例患者(其中640例患者被纳入模型构建队列,162例患者被纳入外部验证队列)。在模型构建队列中,将640例患者按照7:3的比例随机分配至训练队列和内部验证队列。通过整合单因素逻辑回归分析和最小绝对收缩与选择算子回归分析,确定了PGD3的独立风险因素。并基于选定的因素,采用9种ML模型构建了PGD3的预测模型,并进一步评估了其预测性能。此外,通过3种移植后指标对模型的分层预测性能进行了评估。最后,采用SHapley Additive exPlanations算法对所选因素的预测重要性进行了分析。

结果




本研究识别出9个独立的临床风险因素作为预测模型的选定因素。在9种ML模型中,随机森林(random forest, RF)模型展现出最佳的预测性能:训练队列中,曲线下面积(area under the curve, AUC)为0.9415,敏感性(sensitivity, Se)为0.8972,特异性(specificity, Sp)为0.8795;在内部验证队列中,AUC为0.7975,Se为0.7520,Sp为0.7313;在外部验证队列中,AUC为0.8214,Se为0.8235,Sp为0.6667。对模型的校准能力和临床实用性进一步评估表明,RF模型在预测PGD3方面具有较好的适用性。同时,RF模型在术后支持的风险分层方面也表现最佳(体外膜氧合时间:P < 0.001;机械通气时间:P = 0.006;重症监护病房住院时间:P < 0.001)。

结论


RF模型在Lung Tx患者预测PGD3和术后风险分层方面表现最佳。

Figure 1. The flowchart of Lung Tx patient enrollment.

Figure 3. Performance comparison of 9 machine learning models on training, internal validation, and external validation cohorts. 

Figure 4. Construction and evaluation of the RF model for PGD3. 

2017·EATTS




Python社区是高质量的Python/Django开发社区
本文地址:http://www.python88.com/topic/178683
 
40 次点击