Abstract:BACKGROUND:Personalized medicine requires the patient similarity analysis for providing specific treatments tailed for each patient. However, the patient similarity analysis in personalized clinical scenarios encounters challenges, which are twofold. First, heterogeneous and multi-type data are usually recorded to Electronic Health Records (EHRs) during the course of admissions, which makes it difficult to measure the patient similarity. Second, disease progression manifests diverse disease states at different times, which brings sequential complexity to dynamically retrieve similar patients' sequences.MATERIALS AND METHODS:To overcome the above-mentioned challenges, we propose a novel dynamic patient similarity analysis model based on deep learning. Specifically, the proposed model embeds the multi-type and heterogeneous data into hidden representations with a specially designed embedding and attention module. Thereafter, the proposed model retrieves similar patients' sequences based on these hidden representations in a dynamic manner. More importantly, we adopt two clinical tasks, i.e., diagnosis prediction and medication recommendation, to validate the effectiveness of the proposed model. It is worth noticing that the proposed model integrates a drug-drug interaction (DDI) knowledge graph in the medication recommendation task to reduce adverse reactions caused by combinational treatments, such that a more rational strategy can be realized. We evaluate our proposed model using the critical care database MIMIC-III, which includes 5,430 patients covering 14,096 clinical visits.RESULTS:The proposed model outperforms several state-of-the-art methods. For diagnosis prediction, the average PR-AUC score of the proposed model reaches 0.6200, which is significantly higher than that of the baseline models (0.2497∼0.5407). Meanwhile, for medication recommendation, the average PR-AUC of the proposed model is 0.6682 (Jaccard: 0.4070; F1: 0.5672; Recall: 0.7832) whereas the K-nearest model can only reach 0.3805 (Jaccard: 0.3911; F1: 0.5465; Recall: 0.5705). In addition, our proposed model achieves a lower DDI rate.CONCLUSION:We propose a novel dynamic patient similarity analysis model, which can be implemented into a decision support system for clinical tasks including diagnosis prediction, surgical procedure selection, medication recommendation, etc. Also, the proposed model serves as an explainable protocol in clinical practice thanks to its analogy to real clinical reasoning where a doctor diagnoses diseases and prescribes medications according to the previous cured patients empirically.

A Novel Two-stage Deming Regression Framework with Applications to Association Analysis between Clinical Risks

Deep Dynamic Patient Similarity Analysis: Model Development and Validation in ICU.

Risk Prediction Model for In-Hospital Mortality in Women with ST-elevation Myocardial Infarction: A Machine Learning Approach

Distributed Learning from Multi-Site Observational Health Data for Zero-Inflated Count Outcomes

Deep Learning Framework with Uncertainty Quantification for Survey Data: Assessing and Predicting Diabetes Mellitus Risk in the American Population

A novel hyperparameter search approach for accuracy and simplicity in disease prediction risk scoring

A Novel Data-Driven Framework for Risk Characterization and Prediction from Electronic Medical Records: A Case Study of Renal Failure

Directed Disease Networks to Facilitate Multiple-Disease Risk Assessment Modeling

Stabilized Sparse Ordinal Regression for Medical Risk Stratification

A Regularized Deep Learning Approach for Clinical Risk Prediction of Acute Coronary Syndrome Using Electronic Health Records

Dynamic and concordance-assisted learning for risk stratification with application to Alzheimer's disease

A New Approach for Interpretability and Reliability in Clinical Risk Prediction: Acute Coronary Syndrome Scenario

Abstract TP31: Identifying Multivariate Nonlinear Bias in Sammpris Using Graph-Based AI Analysis

Improving dynamic stroke risk prediction in non-anticoagulated patients with and without atrial fibrillation: comparing common clinical risk scores and machine learning algorithms

Abstract 16426: A Risk Model to Predict Atrial Fibrillation in Diabetes Using Machine Learning: The ACCORD Study

Counterfactual Reasoning for Fair Clinical Risk Prediction

A novel deep learning algorithm for real-time prediction of clinical deterioration in the emergency department for a multimodal clinical decision support system

Clinical risk prediction with random forests for survival, longitudinal, and multivariate (RF-SLAM) data analysis

A Novel Explainable AI Method to Assess Associations between Temporal Patterns in Patient Trajectories and Adverse Outcome Risks: Analyzing Fitness as a Risk Factor of ADRD

Relational Regularized Risk Prediction of Acute Coronary Syndrome Using Electronic Health Records