Deepr: A Convolutional Net for Medical Records

TL;DR

Deepr is an end-to-end deep learning system that automatically extracts features from irregular medical records, detects clinical motifs, and predicts patient readmission risk with superior accuracy and interpretability.

Contribution

Deepr introduces a convolutional neural network that processes sequential medical records to identify predictive clinical motifs and improve risk prediction.

Findings

Deepr outperforms traditional methods in predicting hospital readmission.

It enables transparent visualization of clinical motifs.

The model uncovers meaningful disease and intervention structures.

Abstract

Feature engineering remains a major bottleneck when creating predictive systems from electronic medical records. At present, an important missing element is detecting predictive regular clinical motifs from irregular episodic records. We present Deepr (short for Deep record), a new end-to-end deep learning system that learns to extract features from medical records and predicts future risk automatically. Deepr transforms a record into a sequence of discrete elements separated by coded time gaps and hospital transfers. On top of the sequence is a convolutional neural net that detects and combines predictive local clinical motifs to stratify the risk. Deepr permits transparent inspection and visualization of its inner working. We validate Deepr on hospital data to predict unplanned readmission after discharge. Deepr achieves superior accuracy compared to traditional techniques, detects meaningful clinical motifs, and uncovers the underlying structure of the disease and intervention space.