Prediction of Postoperative Hospital Stay with Deep Learning Based on 101 654 Operative Reports in Neurosurgery.

This study was aimed at testing the informativeness of neurosurgical operative reports for predicting the duration of postoperative stay in a hospital using deep learning techniques. The recurrent neuronal networks (GRU) were applied to the word-embedded texts in our experiments. The mean absolute error of prediction in 90% of cases was 2.8 days. These results demonstrate the potential utility of narrative medical texts as a substrate for decision support technologies in neurosurgery. PMID: 30942728 [PubMed - in process]
Source: Studies in Health Technology and Informatics - Category: Information Technology Tags: Stud Health Technol Inform Source Type: research