Autonomous system supporting patient-specific transfer and discharge decisions

支持患者特定转移和出院决策的自主系统

• 批准号: 9256278
• 负责人: Ritankar Das
• 金额: $ 34.78万
• 依托单位: DASCENA, INC.
• 依托单位国家: 美国
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-01-01 至 2018-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9256278
• 关键词: Address; Antibiotics; Area; Cessation of life; Chronic; Clinical; Clinical Decision Support Systems; Computer software; Cues; Data; Data Quality; Data Set; Databases; Deterioration; Event; Future; Geography; Goals; Gold; Health; Healthcare Systems; Homeostasis; Hospital Mortality; Hour; Inpatients; Intensive Care Units; Intervention; Knowledge; Life; Liquid substance; Machine Learning; Measurement; Medical; Methods; Patient Care; Patient Discharge; Patient Transfer; Patients; Performance; Phase; Physiology; Process; ROC Curve; Receiver Operating Characteristics; Recommendation; Research; Risk; SCAP2 gene; Sensitivity and Specificity; Series; Site; Small Business Innovation Research Grant; Support System; System; Techniques; Testing; Time; Training; Triage; Update; Validation; Work; base; cost; experimental study; high standard; improved; mortality; novel; patient population; success; support tools; tool; trend

Significance: In this SBIR project, we propose to improve the utility of AutoTriage, a machine-learning based clinical decision support (CDS) system, by integrating clinician intervention medical information into its predictions. Despite identified needs for CDS systems in patient transfer and discharge decisions, existing tools do not meet high standards for sensitivity and specificity. This is because current CDS methods are unable to distinguish changes in patient health due to clinician intervention from those arising due to an internal homeostatic mechanism. Thus, for example, existing tools may erroneously suggest discharge for a patient currently undergoing a life-sustaining treatment. Research Question: Can machine learning principles be used to create a classifier which incorporates signs of clinical intervention to inform transfer and discharge decision support, ultimately leading to higher quality predictions? In addition, will such a tool be able to maintain its performance when tested on a different patient population or one for which the data quality is much poorer? Prior Work: We have developed AutoTriage, a machine learning-based CDSS for 12-hour mortality prediction. On the publicly available MIMIC-III retrospective data set, this system attains an area under the receiver operating characteristic curve (AUROC) of 0.88, which is superior to commonly used triage scores MEWS (AUROC = 0.75), SOFA (0.71), and SAPS-II (0.72) on the same data set. Specific Aims: To integrate clinician intervention information into existing AutoTriage software (Aim 1), and to test the robustness of this modified tool to changes in patient population and data quality (Aim 2). Methods: We will create gold standards for periods of clinician intervention, using chart events and keywords from clinician notes. Then, we will train a binary classifier for identifying these periods and, ultimately, use the classifier to modify AutoTriage scores. Robustness studies will be performed on the retrospective UC ReX and sparse MIMIC III databases. Successful completion of Aim 1will be demonstrated if 75% of all hours of clinician intervention are correctly classified, if the test-set area under the ROC curve improves by 5% over its current value, and if 30-day readmission predictions are 10% more accurate for patients treated within the last hour. Aim 2 will be completed if AutoTriage ROC area performance is within ± 0.10 of its original value for both UC ReX and sparse MIMIC III sets. Future Directions: Following the proposed work, the AutoTriage system will be deployed at the sites of our ongoing clinical implementations. During this study, we project that AutoTriage will assess mortality risk for 25,000 ICU patients per year, helping clinicians more effectively allocate interventions totaling $15 million.

意义：在这个SBIR项目中，我们建议提高基于机器学习的自动分类的效用 临床决策支持(CDS)系统，通过将临床医生干预的医疗信息集成到ITS中 预测。尽管在患者转移和出院决策中确定了CDS系统的需求，但现有 工具不符合敏感度和特异度的高标准。这是因为当前的CDS方法 无法区分由于临床医生干预引起的患者健康变化和由于内部 动态平衡机制。因此，例如，现有工具可能错误地建议患者出院 目前正在接受维持生命的治疗。研究问题：机器学习原理能否 用于创建包含临床干预迹象的分类器，以通知转院和出院 决策支持，最终带来更高质量的预测？此外，这样的工具是否能够 在不同的患者群体或数据质量较高的人群上进行测试时，保持其性能 更穷？前期工作：我们已经开发了AutoTriage，这是一个基于机器学习的CDSS，用于预测12小时死亡率 预测。在公开可用的MIMIC-III追溯数据集上，该系统在 接受者操作特征曲线(AUROC)为0.88，优于常用的分诊评分 MEWS(AUROC=0.75)、SOFA(0.71)和SAPS-II(0.72)。具体目标：整合 将临床医生干预信息输入现有的AutoTriage软件(Aim 1)，并测试其稳健性 根据患者群体和数据质量的变化修改工具(目标2)。方法：我们将创造黄金 临床医生干预周期的标准，使用临床医生笔记中的图表事件和关键字。然后，我们 我将训练一个二进制分类器来识别这些周期，并最终使用该分类器来修改自动分类 得分。将在追溯的UC Rex和Sparse Mimic III数据库上进行稳健性研究。 如果临床医生干预的所有小时中有75%是正确的，则成功完成目标1将被证明 分类，如果ROC曲线下的测试集区域比其当前值改善5%，并且如果30天 对于在过去一小时内接受治疗的患者，重新入院预测的准确性要高出10%。目标2将是 如果UC Rex和的自动分类ROC区域性能均在其原始值的±0.10以内，则完成 稀疏模拟III集。未来方向：根据拟议的工作，自动分类系统将 部署在我们正在进行的临床实施的地点。在这项研究中，我们预计AutoTriage将 每年评估25,000名ICU患者的死亡风险，帮助临床医生更有效地分配干预措施 总计1500万美元。

项目成果

Effect of a machine learning-based severe sepsis prediction algorithm on patient survival and hospital length of stay: a randomised clinical trial.

• DOI: 10.1136/bmjresp-2017-000234
• 发表时间: 2017
• 期刊: BMJ open respiratory research
• 影响因子: 4.1
• 作者: Shimabukuro DW;Barton CW;Feldman MD;Mataraso SJ;Das R
• 通讯作者: Das R

Machine learning landscapes and predictions for patient outcomes.

• DOI: 10.1098/rsos.170175
• 发表时间: 2017-07
• 期刊: Royal Society open science
• 影响因子: 3.5
• 作者: Das R;Wales DJ
• 通讯作者: Wales DJ

Using Transfer Learning for Improved Mortality Prediction in a Data-Scarce Hospital Setting.

• DOI: 10.1177/1178222617712994
• 发表时间: 2017
• 期刊: Biomedical informatics insights
• 作者: Desautels T;Calvert J;Hoffman J;Mao Q;Jay M;Fletcher G;Barton C;Chettipally U;Kerem Y;Das R
• 通讯作者: Das R

Prediction of early unplanned intensive care unit readmission in a UK tertiary care hospital: a cross-sectional machine learning approach.

• DOI: 10.1136/bmjopen-2017-017199
• 发表时间: 2017-09-15
• 期刊: BMJ open
• 影响因子: 2.9
• 作者: Desautels T;Das R;Calvert J;Trivedi M;Summers C;Wales DJ;Ercole A
• 通讯作者: Ercole A