Prediction of COVID-19 Patients' Emergency Room Revisit using Multi-Source Transfer Learning
Yuelyu Ji1, Yuhe Gao2, Runxue Bao3
1Department of Information Science, School of Computing and Information, University of Pittsburgh, Pittsburgh,USA.
Insights
Identifying COVID-19 patients likely to revisit the emergency room (ER) is crucial. A new deep transfer learning model, Multi-DANN, effectively predicts 7-day ER revisits using Electronic Health Records (EHRs).
Area of Science:
- Medical Informatics
- Artificial Intelligence in Healthcare
- Public Health
Background:
- Coronavirus disease 2019 (COVID-19) presents a significant global health challenge with varied clinical outcomes.
- A substantial number of COVID-19 patients require emergency room (ER) readmission shortly after discharge, increasing healthcare burdens.
- Early identification of patients at high risk for ER revisits is essential for optimizing medical resource allocation.
Purpose of the Study:
- To develop and evaluate predictive models for identifying COVID-19 patients at high risk of ER revisits within 7 days of discharge.
- To address data heterogeneity across multiple ERs using deep transfer learning techniques.
- To assess the effectiveness of the Domain Adversarial Neural Network (DANN) algorithms in predicting patient revisits.
Main Methods:
- Utilized Electronic Health Records (EHRs) from 3,210 COVID-19 patient encounters across 13 ERs.
- Employed Natural Language Processing (NLP) with ScispaCy to extract key clinical concepts.
- Developed 7-day revisit prediction models using frequent clinical concepts and compared Multi-DANN, Single-DANN, and baseline strategies.
Main Results:
- The Multi-DANN model demonstrated superior performance in predicting 7-day ER revisits for COVID-19 patients, achieving a median AUROC of 0.8.
- Multi-DANN significantly outperformed Single-DANN (median AUROC = 0.5) and baseline models, effectively handling domain differences.
- The study confirmed the utility of EHR data in developing robust predictive models for ER revisit risk.
Conclusions:
- Deep transfer learning, particularly the Multi-DANN approach, is highly effective in predicting COVID-19 patient ER revisits.
- The Multi-DANN strategy successfully mitigates data heterogeneity issues from multiple sources, enhancing model generalizability.
- This predictive capability can aid clinicians in proactive patient management and resource optimization within emergency departments.
Abstract:
The coronavirus disease 2019 (COVID-19) has led to a global pandemic of significant severity. In addition to its high level of contagiousness, COVID-19 can have a heterogeneous clinical course, ranging from asymptomatic carriers to severe and potentially life-threatening health complications. Many patients have to revisit the emergency room (ER) within a short time after discharge, which significantly increases the workload for medical staff. Early identification of such patients is crucial for helping physicians focus on treating life-threatening cases. In this study, we obtained Electronic Health Records (EHRs) of 3,210 encounters from 13 affiliated ERs within the University of Pittsburgh Medical Center between March 2020 and January 2021. We leveraged a Natural Language Processing technique, ScispaCy, to extract clinical concepts and used the 1001 most frequent concepts to develop 7-day revisit models for COVID-19 patients in ERs. The research data we collected were obtained from 13 ERs, which may have distributional differences that could affect the model development. To address this issue, we employed a classic deep transfer learning method called the Domain Adversarial Neural Network (DANN) and evaluated different modeling strategies, including the Multi-DANN algorithm (which considers the source differences), the Single-DANN algorithm (which doesn't consider the source differences), and three baseline methods: using only source data, using only target data, and using a mixture of source and target data. Results showed that the Multi-DANN models outperformed the Single-DANN models and baseline models in predicting revisits of COVID-19 patients to the ER within 7 days after discharge (median AUROC = 0.8 vs. 0.5). Notably, the Multi-DANN strategy effectively addressed the heterogeneity among multiple source domains and improved the adaptation of source data to the target domain. Moreover, the high performance of Multi-DANN models indicates that EHRs are informative for developing a prediction model to identify COVID-19 patients who are very likely to revisit an ER within 7 days after discharge.
