Predicting patients hospital length of stay (LOS) is essential for improving
resource allocation and supporting decision-making in healthcare organizations.
This paper proposes a novel approach for predicting LOS by modeling patient
information as sequences of events. Specifically, we present a
transformer-based model, termed Medic-BERT (M-BERT), for LOS prediction using
the unique features describing patients medical event sequences. We performed
empirical experiments on a cohort of more than 45k emergency care patients from
a large Danish hospital. Experimental results show that M-BERT can achieve high
accuracy on a variety of LOS problems and outperforms traditional
nonsequence-based machine learning approaches.

此研究提出了一种基于 Transformer 模型的 Medic-BERT（M-BERT），通过将病人信息建模为事件序列来预测其住院天数，并在丹麦大型医院的 45000 多名急诊病人上进行了实验，结果表明 M-BERT 在各种住院时间问题上表现出色，优于传统非序列机器学习方法。

利用患者事件序列预测住院天数

Hospitalization Length of Stay Prediction using Patient Event Sequences

Hospital length of stay (LOS) is one of the most essential healthcare metrics
that reflects the hospital quality of service and helps improve hospital
scheduling and management. LOS prediction helps in cost management because
patients who remain in hospitals usually do so in hospital units where
resources are severely limited. In this study, we reviewed papers on LOS
prediction using machine learning and statistical approaches. Our literature
review considers research studies that focus on LOS prediction for stroke
patients. Some of the surveyed studies revealed that authors reached
contradicting conclusions. For example, the age of the patient was considered
an important predictor of LOS for stroke patients in some studies, while other
studies concluded that age was not a significant factor. Therefore, additional
research is required in this domain to further understand the predictors of LOS
for stroke patients.

该研究回顾了机器学习和统计方法在预测中风病人住院时间方面的研究，分析了目前研究领域中存在的一些矛盾之处，需要进一步研究了解中风患者住院时间的预测因素。