Interpretable prediction of necrotizing enterocolitis from machine learning analysis of premature infant stool microbiota.
Interpretable prediction of necrotizing enterocolitis from machine learning analysis of premature infant stool microbiota.
复制标题
早产儿粪便微生物群的机器学习分析对坏死性小肠结肠炎的可解释性预测。
DOI：
10.1186/s12859-022-04618-w
复制
发表时间：
2022-03-25
影响因子：
3
通讯作者：
Hooven TA
中科院分区：
文献类型：
--
作者：
Lin YC;Salleb-Aouissi A;Hooven TA
坏死性小肠结肠炎(NEC)是一种在极低出生体重早产儿中常见的、可能是灾难性的肠道疾病。体重低于1500微克的新生儿中，高达15%会受到影响，NEC会导致突然发作的进行性肠道炎症和坏死，可能导致严重的肠道失血、多器官损伤或死亡。目前还没有确定NEC的统一原因，也没有任何可靠的生物标记物表明单个患者患这种疾病的风险。由于无法提前预测NEC，目前的医疗策略涉及密切的临床监测，以努力在无法恢复的肠道损伤发生之前尽快治疗患有NEC的婴儿。在这份报告中，我们描述了一种新的机器学习应用程序，用于基于肠道微生物区系数据生成动态的、个性化的NEC风险评分，这些数据可以通过对其他被丢弃的婴儿粪便中的细菌DNA进行测序来确定。我们与过去工作的一个核心见解是认识到，从粪便微生物群预测疾病代表了机器学习问题的一个特定子类型，称为多实例学习(MIL)。我们使用了基于神经网络的MIL架构，我们在来自两个队列的独立数据集上进行了测试，这些队列包括来自261名高危婴儿的3595份粪便样本。我们的报告还引入了一个名为“成长袋”分析的新概念，它随着时间的推移应用MIL，允许将过去的数据纳入每个新的风险计算。这种方法可以早期、准确地预测NEC，平均灵敏度为86%，特异度为90%。真阳性的NEC预测平均发生在疾病发作前8天。我们还证明，我们的MIL算法中包含的注意力门控机制可以解释NEC的风险，识别过去工作与NEC相关的几个细菌分类群，并可能为NEC发病机制的新假说指明方向。我们的系统是灵活的，可以接受从靶向16S或“猎枪”全基因组DNA测序产生的微生物区系数据。它在常见的、可能令人混淆的早产儿临床事件中表现良好，例如围产期心肺抑制、抗生素给药、喂养中断或母乳喂养和配方奶之间的过渡。我们已经开发并验证了一种基于MIL的可靠系统，用于从无害的早产儿粪便中预测NEC。虽然该系统是为预测NEC而开发的，但我们的MIL方法也可能适用于以人类微生物区系变化为特征的其他疾病。网上版载有补充材料，可在10.1186/s12859-022-04618-w查阅。
Necrotizing enterocolitis (NEC) is a common, potentially catastrophic intestinal disease among very low birthweight premature infants. Affecting up to 15% of neonates born weighing less than 1500 g, NEC causes sudden-onset, progressive intestinal inflammation and necrosis, which can lead to significant bowel loss, multi-organ injury, or death. No unifying cause of NEC has been identified, nor is there any reliable biomarker that indicates an individual patient’s risk of the disease. Without a way to predict NEC in advance, the current medical strategy involves close clinical monitoring in an effort to treat babies with NEC as quickly as possible before irrecoverable intestinal damage occurs. In this report, we describe a novel machine learning application for generating dynamic, individualized NEC risk scores based on intestinal microbiota data, which can be determined from sequencing bacterial DNA from otherwise discarded infant stool. A central insight that differentiates our work from past efforts was the recognition that disease prediction from stool microbiota represents a specific subtype of machine learning problem known as multiple instance learning (MIL). We used a neural network-based MIL architecture, which we tested on independent datasets from two cohorts encompassing 3595 stool samples from 261 at-risk infants. Our report also introduces a new concept called the “growing bag” analysis, which applies MIL over time, allowing incorporation of past data into each new risk calculation. This approach allowed early, accurate NEC prediction, with a mean sensitivity of 86% and specificity of 90%. True-positive NEC predictions occurred an average of 8 days before disease onset. We also demonstrate that an attention-gated mechanism incorporated into our MIL algorithm permits interpretation of NEC risk, identifying several bacterial taxa that past work has associated with NEC, and potentially pointing the way toward new hypotheses about NEC pathogenesis. Our system is flexible, accepting microbiota data generated from targeted 16S or “shotgun” whole-genome DNA sequencing. It performs well in the setting of common, potentially confounding preterm neonatal clinical events such as perinatal cardiopulmonary depression, antibiotic administration, feeding disruptions, or transitions between breast feeding and formula. We have developed and validated a robust MIL-based system for NEC prediction from harmlessly collected premature infant stool. While this system was developed for NEC prediction, our MIL approach may also be applicable to other diseases characterized by changes in the human microbiota. The online version contains supplementary material available at 10.1186/s12859-022-04618-w.
登录
查看更多内容
影响因子：
5.2
作者：
Dobbler PT;Procianoy RS;Mai V;Silveira RC;Corso AL;Rojas BS;Roesch LFW
通讯作者：
Roesch LFW
影响因子：
6.4
作者：
Brown CT;Xiong W;Olm MR;Thomas BC;Baker R;Firek B;Morowitz MJ;Hettich RL;Banfield JF
通讯作者：
Banfield JF
影响因子：
6.5
作者：
Capone, Kimberly A.;Dowd, Scot E.;Stamatas, Georgios N.;Nikolovski, Janeta
通讯作者：
Nikolovski, Janeta
DOI：
10.1007/bf01029321
发表时间：
1984-01-01
期刊：
JOURNAL OF THE INTERNATIONAL ASSOCIATION FOR MATHEMATICAL GEOLOGY
影响因子：
--
作者：
AITCHISON, J
通讯作者：
AITCHISON, J
DOI：
10.1073/pnas.1002601107
发表时间：
2010-06-29
影响因子：
11.1
作者：
Dominguez-Bello, Maria G.;Costello, Elizabeth K.;Knight, Rob
通讯作者：
Knight, Rob