Graph-guided deep hashing networks for similar patient retrieval

Yifan Gu, Xuebing Yang, Mengxuan Sun, Chutong Wang, Hongyu Yang, Chao Yang, Jinwei Wang, Guilan Kong, Jicheng Lv, Wensheng Zhang

Abstract: Highlights•A novel framework named guided Deep Hashing Networks (GDHN) is proposed to derive more informative hash codes for efficient and accurate similar patient retrieval on electronic health records (EHRs).•GDHN organizes multi-labels in EHRs as a graph and employs a graph convolutional network to extract the correlation dependencies and generate multi-label embeddings for guiding the patient hashing procedure.•The experiments conducted on two datasets demonstrate that GDHN outperforms the competitors at different hash code lengths, allowing patients with higher similarity to the query to be ranked higher in the retrieved results.