The impact of machine learning models on healthcare will depend on the degree of trust that healthcare professionals place in the predictions made by these models. In this paper, we present a method to provide people with clinical expertise with domain-relevant evidence about why a prediction should be trusted. We first design a probabilistic model that relates meaningful latent concepts to prediction targets and observed data. Inference of latent variables in this model corresponds to both making a prediction and providing supporting evidence for that prediction. We present a two-step process to efficiently approximate inference: (i) estimating model parameters using variational learning, and (ii) approximating maximum a posteriori estimation of latent variables in the model using a neural network, trained with an objective derived from the probabilistic model. We demonstrate the method on the task of predicting mortality risk for patients with cardiovascular disease. Specifically, using electrocardiogram and tabular data as input, we show that our approach provides appropriate domain-relevant supporting evidence for accurate predictions.
A novel interpretable machine learning system to generate clinical risk scores: An application for predicting early mortality or unplanned readmission in a retrospective cohort study
Duke-NUS Medical School（杜克-新加坡国立大学医学院）；Singapore Health Services（新加坡保健服务集团）；Singapore General Hospital（新加坡中央医院）；National University of Singapore（新加坡国立大学）；Duke University（杜克大学）；Singapore Eye Research Institute（新加坡眼科研究所）；Singapore National Eye Centre（新加坡国家眼科中心）
Copula Based Fusion of Clinical and Genomic Machine Learning Risk Scores for Breast Cancer Risk Stratification
Department of Mathematics, University of Louisiana at Lafayette（路易斯安那州立大学拉斐特分校数学系）；Department of Physical Sciences & Mathematics, West Liberty University（韦斯特里布利大学物理科学与数学系）；Department of Mathematics and Statistics, Minnesota State University（明尼苏达州立大学数学与统计学系）
Tongji University（同济大学）；Shanghai Innovation Institute（上海创新研究院）；Department of Critical Care Medicine, Shanghai East Hospital, Tongji University School of Medicine（同济大学医学院附属上海东方医院重症医学科）；Harbin Institute of Technology（哈尔滨工业大学）
Improving Hospital Risk Prediction with Knowledge-Augmented Multimodal EHR Modeling
Department of Computer Science, University of Virginia（大学计算机科学系）；University of Virginia School of Medicine（弗吉尼亚大学医学院）；Virginia Polytechnic Institute and State University（弗吉尼亚理工学院和州立大学）；Biocomplexity Institute and Initiative, University of Virginia（大学生物复杂性研究所）；Division of Infectious Diseases & International Health, University of Virginia School of Medicine（大学感染性疾病与国际卫生分会）
Learning Clinical Concepts for Predicting Risk of Progression to Severe COVID-19
Carnegie Mellon University（卡内基梅隆大学）；Highmark Health（高马克健康）；Allegheny Health Network（阿勒格尼健康网络）
随着COVID-19的普遍存在，识别高风险个体至关重要。使用宾夕法尼亚州西南部一家主要医疗保健提供商的数据，我们开发了预测严重COVID-19进展的生存模型。在这项工作中，我们面临着一个权衡更准确的模型依赖于许多功能和不太准确的模型依赖于一些功能与临床医生的直觉。更复杂的是，许多EHR功能往往编码不足，降低了较小模型的准确性。在这项研究中，我们开发了两组高性能风险评分：（i）从所有可用特征构建的无约束模型;以及（ii）在训练风险预测器之前学习一小组临床概念的管道。学习的概念提高了相应特征的性能（C指数0.858 vs. 0.844），并在样本外评估（后续时间段）时证明了（i）的改进。我们的模型优于以前的作品（C-index 0.844-0.872 vs. 0.598-0.810）。
TRACER: Transfer Learning based Real-time Adaptation for Clinical Evolving Risk
Duke University School of Medicine（杜克大学医学学院）；Duke AI Health（杜克AI健康）；Department of Biostatistics and Bioinformatics（生物统计学与生物信息学系）；Duke-NUS Medical School（杜克-新加坡国立大学医学院）；Centre for Quantitative Medicine（定量医学中心）；Duke-NUS AI + Medical Sciences Initiative（杜克-新加坡国立大学AI+医学科学倡议）；Pre-hospital & Emergency Research Centre（急救与应急研究中心）；Health Services and Systems Research（卫生服务与系统研究）；NUS Artificial Intelligence Institute（新加坡国立大学人工智能研究所）
Improving Early Prediction of Type 2 Diabetes Mellitus with ECG-DiaNet: A Multimodal Neural Network Leveraging Electrocardiogram and Clinical Risk Factors
College of Science and Engineering, Hamad Bin Khalifa University（哈马德·本·哈利法大学科学与工程学院）
From Data Lifting to Continuous Risk Estimation: A Process-Aware Pipeline for Predictive Monitoring of Clinical Pathways
Department of Informatics, University of Bari Aldo Moro（巴里阿尔多·莫罗大学信息学院）；Department of Engineering, University of Sannio（萨尼奥大学工程学院）；Department of Economy and Law, UnitelmaSapienza University（单位elma萨皮恩扎大学经济法学院）
Predicting High-Risk Colorectal Polyps in African Americans Using Pre-Colonoscopy Clinical Features: Machine Learning Model Development and Temporal Validation
Howard University College of Medicine（霍华德大学医学院）；University of Oslo（奥斯陆大学）；University College London Hospital（伦敦大学学院医院）
A Latent Risk-Aware Machine Learning Approach for Predicting Operational Success in Clinical Trials based on TrialsBank
Groupe Sorintellis Inc.（索林特利斯集团）；Department of Computer Science and Operations Research, Université de Montréal（蒙特利尔大学计算机科学与运筹学系）；Department of Computer Engineering, École Polytechnique de Montréal（蒙特利尔理工学院计算机工程系）；Faculty of Pharmacy, Université de Montréal（蒙特利尔大学药学院）；Faculty of Pharmacy, Université Laval（拉瓦尔大学药学院）；Vantage BioTrials（Vantage BioTrials公司）；CDS monitoring（CDS监测公司）；Department of Social and Preventive Medicine, School of Public Health, Université de Montréal（蒙特利尔大学公共卫生学院社会与预防医学系）；Department of Epidemiology, Biostatistics and Occupational Health, McGill University（麦吉尔大学流行病学、生物统计学与职业健康系）
Machine Learning for Violence Risk Assessment Using Dutch Clinical Notes
Utrecht University（乌得勒支大学）；Eindhoven University of Technology（埃因霍温理工大学）；University Medical Center Utrecht（乌得勒支大学医学中心）；Leiden University Medical Center（莱顿大学医学中心）；Leiden Institute of Advanced Computer Science（莱顿高级计算机科学研究所）
Machine Learning to Support Triage of Children at Risk for Epileptic Seizures in the Pediatric Intensive Care Unit
Technion - Israel Institute of Technology（以色列理工学院）；The Hospital for Sick Children（病童医院）；University of Toronto（多伦多大学）；KU Leuven（鲁汶大学）；Alberta Children’s Hospital（阿尔伯塔儿童医院）；University of Calgary（卡尔加里大学）；Li Ka Shing Knowledge Institute of St. Michael’s Hospital（圣迈克尔医院李嘉诚知识研究院）
Adaptable Cardiovascular Disease Risk Prediction from Heterogeneous Data using Large Language Models
Eidgenössische Technische Hochschule (ETH), Zürich, Switzerland（瑞士苏黎世联邦理工学院）；Max Planck Institute for Intelligent Systems, Tübingen, Germany（德国图宾根智能系统研究所）；ELLIS Institute, Tübingen, Germany（德国图宾根ELLIS研究所）；Stanford School of Medicine, CA, USA（美国斯坦福医学院）
