Year: 2019
Type: article
Abstract: This paper presents a novel method for predicting suicidal ideation from electronic health records (EHR) and ecological momentary assessment (EMA) data using deep sequential models. Both EHR longitudinal data and EMA question forms are defined by asynchronous, variable length, randomly sampled data sequences. In our method, we model each of them with a recurrent neural network, and both sequences are aligned by concatenating the hidden state of each of them using temporal marks. Furthermore, we incorporate attention schemes to improve performance in long sequences and time-independent pre-trained schemes to cope with very short sequences. Using a database of 1023 patients, our experimental results show that the addition of EMA records boosts the system recall to predict the suicidal ideation diagnosis from 48.13% obtained exclusively from EHR-based state-of-the-art methods to 67.78%. Additionally, our method provides interpretability through the t-distributed stochastic neighbor embedding (t-SNE) representation of the latent space. Furthermore, the most relevant input features are identified and interpreted medically.
Authors: Ignacio Peis, Pablo M. Olmos, Constanza Vera-Varela, Maria Luisa Barrigon, Philippe Courtet, Enrique Baca-Garcia, Antonio Artes-Rodriguez
Institutions: Instituto de Investigación Sanitaria Gregorio Marañón, Universidad Carlos III de Madrid, Hospital Universitario Fundación Jiménez Díaz, Universidad Autónoma de Madrid, Université de Montpellier, Hôpital Lapeyronie, Catholic University of the Maule, Instituto de Salud Carlos III, Centro de Investigación Biomédica en Red de Salud Mental
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: G1075404386, G1520406698, G1650125035, G278303889, G3210393749, G4124952967, G4207158616, G4494357670, G4500004730, G4652866256, G47552203, G4771197469, G5040582290, G5279340442, G6742695155, G6858479917, G696673246, Monitorización basada en Ecological Momentary Assessment y marcadores geneticos para la personalización del tratamiento en pacientes depresivos con riesgo suicida, G7571065899, G894356020
OpenAlex ID: W2946874644