Interpretable Patient Mortality Prediction with Multi-value Rule Sets
Tippie College of Business \ of Iowa；Carver College of Medicine\ of Iowa；School of Business\ Island College；College of Dentistry University of Illinois at Chicago
We propose a Multi-vAlue Rule Set (MRS) model for in-hospital predicting patient mortality. Compared to rule sets built from single-valued rules, MRS adopts a more generalized form of association rules that allows multiple values in a condition. Rules of this form are more concise than classical single-valued rules in capturing and describing patterns in data. Our formulation also pursues a higher efficiency of feature utilization, which reduces possible cost in data collection and storage. We propose a Bayesian framework for formulating a MRS model and propose an efficient inference method for learning a maximum \emph{a posteriori}, incorporating theoretically grounded bounds to iteratively reduce the search space and improve the search efficiency. Experiments show that our model was able to achieve better performance than baseline method including the current system used by the hospital.
A Transformer-based survival model for prediction of all-cause mortality in heart failure patients: a multi-cohort study
University of Oxford（牛津大学）；Health Data Research UK（英国健康数据研究中心）；Katholieke Universiteit Leuven（鲁汶大学）；University of Amsterdam（阿姆斯特丹大学）；University College London（伦敦大学学院）；The George Institute for Global Health（乔治全球健康研究所）；University of New South Wales（新南威尔士大学）；Imperial College London（伦敦帝国学院）；University of Auckland（奥克兰大学）；University of Glasgow（格拉斯哥大学）
本文提出基于 Transformer 的生存模型 TRisk，利用 EHR 纵向病程预测心衰患者36个月全因死亡，并在英、美多队列中验证其优于 MAGGIC-EHR 的准确性、可迁移性与风险因素发现能力。
Predicting Intensive Care Unit Length of Stay and Mortality Using Patient Vital Signs: Machine Learning Model Development and Validation
New Mexico Institute of Mining and Technology（新墨西哥矿业技术学院）；Oak Ridge National Laboratory（橡树岭国家实验室）；The University of Tennessee Health Science Center（田纳西大学健康科学中心）；Illinois State University（伊利诺伊州立大学）
Mortality Prediction with Adaptive Feature Importance Recalibration for Peritoneal Dialysis Patients: a deep-learning-based study on a real-world longitudinal follow-up dataset
Peking University（北京大学）；Health Data Research（健康数据研究所）；University of Edinburgh（爱丁堡大学）；Peking University Third Hospital（北京大学第三医院）；Peking University People’s Hospital（北京大学人民医院）；University of Exeter（埃克塞特大学）
Development and external validation of a multimodal artificial intelligence mortality prediction model of critically ill patients using multicenter data
University of California, San Diego Health（加州大学圣地亚哥分校健康中心）；University of California, San Diego（加州大学圣地亚哥分校）
Imaging-Based Mortality Prediction in Patients with Systemic Sclerosis
Department of Computer Science, Northwestern University, Evanston, IL 60208, USA Division of Pulmonary；Critical Care Medicine, Northwestern University, Chicago, IL 60611, USA Machine \& Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, IL 60611, USA Division of Rheumatology, Northwestern University, Chicago, IL 60611, USA Division of Rheumatology, University of Michigan, Ann Arbor, MI 48109, USA Section of Rheumatology, Allergy \& Immunology, Yale University, New Haven, CT 06520, USA Department of Electrical；Computer Engineering, Northwestern University, Evanston, IL 60208, USA
Rolling Day-Wise Mortality Prediction in Critically Ill Patients With AKI on CRRT Utilizing Machine Pressure Waveforms
University of Alabama at Birmingham（阿拉巴马大学伯明翰分校）；University of Cincinnati（辛辛那提大学）；Icahn School of Medicine at Mount Sinai（西奈山伊坎医学院）；UT Southwestern Medical Center（德克萨斯大学西南医学中心）
A Novel Multi-Task Teacher-Student Architecture with Self-Supervised Pretraining for 48-Hour Vasoactive-Inotropic Trend Analysis in Sepsis Mortality Prediction
University of Southern California（南加州大学）；California State University, Long Beach（加州州立大学长滩分校）；Iran University of Medical Sciences（伊朗医科大学）；California State Polytechnic University, Pomona（加州州立理工大学波莫纳分校）
TA-RNN-Medical-Hybrid: A Time-Aware and Interpretable Framework for Mortality Risk Prediction
Department of Computer Engineering, SRC.,Islamic Azad university（计算机工程系，伊斯兰教Azad大学）；School of Computer Engineering, Iran university of Science and technology（计算机工程学院，伊朗科学技术大学）
AI-powered skin spectral imaging enables instant sepsis diagnosis and outcome prediction in critically ill patients
German Cancer Research Center (DKFZ)（德国癌症研究中心）；Helmholtz Information and Data Science School for Health（亥姆霍兹健康信息与数据科学学院）；Heidelberg University（海德堡大学）；National Center for Tumor Diseases (NCT)（国家肿瘤疾病中心）；Heidelberg University Hospital（海德堡大学医院）；University Medical Center Mannheim（曼海姆大学医学中心）；DKFZ Hector Cancer Institute（DKFZ赫克托癌症研究所）
Machine learning predicts long-term mortality after acute myocardial infarction using systolic time intervals and routinely collected clinical data
École de technologie supérieure, Université du Québec（魁北克大学高等技术学院）；K. N. Toosi University of Technology（K. N. 图什理工大学）；Centre de recherche de l’Institut universitaire de gériatrie de Montréal（蒙特利尔大学老年学研究所）
DeepSOFA: A Continuous Acuity Score for Critically Ill Patients using Clinically Interpretable Deep Learning
University of Florida（佛罗里达大学）
传统的危重病患者病情严重程度评估和院内死亡率预测方法需要使用静态变量阈值进行耗时、易出错的计算。这些方法没有利用流电子健康记录数据的新兴可用性或捕获时间敏感的个体生理模式，这是重症监护室中的关键任务。我们提出了一种新的敏锐度评分框架（DeepSOFA），该框架利用时间测量和可解释的深度学习模型来评估ICU住院期间任何时间点的疾病严重程度。我们使用相同的模型输入将DeepSOFA与SOFA（序贯器官衰竭评估）基线模型进行比较，发现在ICU入院期间的任何时间点，DeepSOFA对院内死亡率的预测都更加准确。在公共数据库中开发并在单个机构队列中验证的DeepSOFA模型在整个ICU住院期间的平均AUC为0.90（95% CI 0.90-0.91），而基线SOFA模型的平均AUC为0.79（95% CI 0.79-0.80）和0.85（95% CI 0.85-0.86）。深度模型非常适合在发生意外不良事件之前识别需要救生干预的ICU患者，并告知患者，提供者和家庭之间关于护理目标和最佳资源利用的共享决策过程。
Designing a Deep Learning-Driven Resource-Efficient Diagnostic System for Metastatic Breast Cancer: Reducing Long Delays of Clinical Diagnosis and Improving Patient Survival in Developing Countries
University of Maryland, Baltimore County（马里兰大学巴尔的摩县分校）；Meta Platforms Inc.（元平台公司）
