MedML: Fusing medical knowledge and machine learning models for early pediatric COVID-19 hospitalization and severity prediction.
MedML: Fusing medical knowledge and machine learning models for early pediatric COVID-19 hospitalization and severity prediction.
复制标题
DOI：
10.1016/j.isci.2022.104970
复制
发表时间：
2022-09-16
期刊：
影响因子：
5.8
通讯作者：
Sun, Jimeng
中科院分区：
文献类型：
--
作者：
Gao, Junyi;Yang, Chaoqi;Heintz, Joerg;Barrows, Scott;Albers, Elise;Stapel, Mary;Warfield, Sara;Cross, Adam;Sun, Jimeng
COVID-19疫情对经济及社会造成破坏性影响。这导致全国范围内呼吁建立模型来预测COVID-19患者的住院和严重疾病，以告知有限医疗资源的分配。为了应对这一挑战，我们提出了一个机器学习模型MedML，使用电子健康记录对儿科人群进行住院和严重程度预测。MedML基于医学知识和倾向分数从超过600万个医学概念中提取最具预测性的特征，并通过图形神经网络将特征间关系纳入医学知识图中。我们在国家队列协作（N3 C）数据集上评估MedML。与最佳基线机器学习模型相比，MedML的AUROC和AUPRC分别高出7%和14%。MedML是一种新的机器学习框架，可以整合临床领域知识，比当前的数据驱动方法更具预测性和可解释性。MedML从超过600万个医学概念中提取最具预测性的特征MedML通过图神经网络将医学知识融合到机器学习模型中MedML使用N3 C数据集在儿科COVID-19预测方面优于其他方法呼吸医学;儿科;人工智能;人工智能应用
The COVID-19 pandemic has caused devastating economic and social disruption. This has led to a nationwide call for models to predict hospitalization and severe illness in patients with COVID-19 to inform the distribution of limited healthcare resources. To address this challenge, we propose a machine learning model, MedML, to conduct the hospitalization and severity prediction for the pediatric population using electronic health records. MedML extracts the most predictive features based on medical knowledge and propensity scores from over 6 million medical concepts and incorporates the inter-feature relationships in medical knowledge graphs via graph neural networks. We evaluate MedML on the National Cohort Collaborative (N3C) dataset. MedML achieves up to a 7% higher AUROC and 14% higher AUPRC compared to the best baseline machine learning models. MedML is a new machine learnig framework to incorporate clinical domain knowledge and is more predictive and explainable than current data-driven methods. MedML extracts the most predictive features from over 6 million medical concepts MedML fuses medical knowledge into machine learning models via graph neural networks MedML outperforms other methods on pediatric COVID-19 predictions using N3C dataset Respiratory medicine; Pediatrics; Artificial intelligence; Artificial intelligence applications
登录
查看更多内容
影响因子：
--
作者：
Jamshidi E;Asgary A;Tavakoli N;Zali A;Setareh S;Esmaily H;Jamaldini SH;Daaee A;Babajani A;Sendani Kashi MA;Jamshidi M;Jamal Rahi S;Mansouri N
通讯作者：
Mansouri N
影响因子：
16.6
作者：
Feng Z;Yu Q;Yao S;Luo L;Zhou W;Mao X;Li J;Duan J;Yan Z;Yang M;Tan H;Ma M;Li T;Yi D;Mi Z;Zhao H;Jiang Y;He Z;Li H;Nie W;Liu Y;Zhao J;Luo M;Liu X;Rong P;Wang W
通讯作者：
Wang W
DOI：
10.1016/j.jpeds.2022.01.001
发表时间：
2022-05
期刊：
The Journal of pediatrics
影响因子：
--
作者：
Oliveira EA;Simões E Silva AC;Oliveira MCL;Colosimo EA;Mak RH;Vasconcelos MA;Miranda DM;Martelli DB;Silva LR;Pinhati CC;Martelli-Júnior H
通讯作者：
Martelli-Júnior H
影响因子：
13.8
作者：
Martin B;DeWitt PE;Russell S;Anand A;Bradwell KR;Bremer C;Gabriel D;Girvin AT;Hajagos JG;McMurry JA;Neumann AJ;Pfaff ER;Walden A;Wooldridge JT;Yoo YJ;Saltz J;Gersing KR;Chute CG;Haendel MA;Moffitt R;Bennett TD
通讯作者：
Bennett TD
DOI：
10.1093/jamia/ocaa196
发表时间：
2021-03-01
影响因子：
6.4
作者：
Haendel, Melissa A.;Chute, Christopher G.;Gersing, Ken R.
通讯作者：
Gersing, Ken R.