Predicting Length of Stay in the Intensive Care Unit with Temporal Pointwise Convolutional Networks
This addresses bed allocation challenges for clinical staff in ICUs, though it is incremental as it builds on existing deep learning methods for EHR data.
The paper tackled predicting ICU length of stay to aid hospital bed management by proposing a Temporal Pointwise Convolutional network, achieving performance improvements of 18-51% over LSTM and Transformer models on the eICU dataset.
The pressure of ever-increasing patient demand and budget restrictions make hospital bed management a daily challenge for clinical staff. Most critical is the efficient allocation of resource-heavy Intensive Care Unit (ICU) beds to the patients who need life support. Central to solving this problem is knowing for how long the current set of ICU patients are likely to stay in the unit. In this work, we propose a new deep learning model based on the combination of temporal convolution and pointwise (1x1) convolution, to solve the length of stay prediction task on the eICU critical care dataset. The model - which we refer to as Temporal Pointwise Convolution (TPC) - is specifically designed to mitigate for common challenges with Electronic Health Records, such as skewness, irregular sampling and missing data. In doing so, we have achieved significant performance benefits of 18-51% (metric dependent) over the commonly used Long-Short Term Memory (LSTM) network, and the multi-head self-attention network known as the Transformer.