A probabilistic framework for predicting disease dynamics: A case study of psychotic depression

Marcos L.P. Bueno*, Arjen Hommersom, Peter J.F. Lucas, Joost Janzing

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Unsupervised learning is often used to obtain insight into the underlying structure of medical data, but it is not always clear how to use such structure in an effective way. In this paper, we propose a probabilistic framework for predicting disease dynamics guided by latent states. The framework is based on hidden Markov models and aims to facilitate the selection of hypotheses that might yield insight into the dynamics. We demonstrate this by using clinical trial data for psychotic depression treatment as a case study. The discovered latent structure and proposed outcome are then validated using standard depression criteria, and are shown to provide new insight into the heterogeneity of psychotic depression in terms of predictive symptoms for different interventions.
Original languageEnglish
Article number103232
Number of pages12
JournalJournal of Biomedical Informatics
Volume95
DOIs
Publication statusPublished - Jul 2019

Fingerprint

Depression
Unsupervised learning
Hidden Markov models
Clinical Trials
Learning

Keywords

  • Machine learning
  • Psychiatry
  • Depression
  • Temporal data
  • Latent variables
  • Hidden Markov model

Cite this

@article{55f135379aa342bebfbaf3cb5cae0181,
title = "A probabilistic framework for predicting disease dynamics: A case study of psychotic depression",
abstract = "Unsupervised learning is often used to obtain insight into the underlying structure of medical data, but it is not always clear how to use such structure in an effective way. In this paper, we propose a probabilistic framework for predicting disease dynamics guided by latent states. The framework is based on hidden Markov models and aims to facilitate the selection of hypotheses that might yield insight into the dynamics. We demonstrate this by using clinical trial data for psychotic depression treatment as a case study. The discovered latent structure and proposed outcome are then validated using standard depression criteria, and are shown to provide new insight into the heterogeneity of psychotic depression in terms of predictive symptoms for different interventions.",
keywords = "Machine learning, Psychiatry, Depression, Temporal data, Latent variables, Hidden Markov model",
author = "Bueno, {Marcos L.P.} and Arjen Hommersom and Lucas, {Peter J.F.} and Joost Janzing",
year = "2019",
month = "7",
doi = "10.1016/j.jbi.2019.103232",
language = "English",
volume = "95",
journal = "Journal of Biomedical Informatics",
issn = "1532-0464",
publisher = "ACADEMIC PRESS INC ELSEVIER SCIENCE",

}

A probabilistic framework for predicting disease dynamics : A case study of psychotic depression. / Bueno, Marcos L.P.; Hommersom, Arjen; Lucas, Peter J.F.; Janzing, Joost.

In: Journal of Biomedical Informatics, Vol. 95, 103232, 07.2019.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - A probabilistic framework for predicting disease dynamics

T2 - A case study of psychotic depression

AU - Bueno, Marcos L.P.

AU - Hommersom, Arjen

AU - Lucas, Peter J.F.

AU - Janzing, Joost

PY - 2019/7

Y1 - 2019/7

N2 - Unsupervised learning is often used to obtain insight into the underlying structure of medical data, but it is not always clear how to use such structure in an effective way. In this paper, we propose a probabilistic framework for predicting disease dynamics guided by latent states. The framework is based on hidden Markov models and aims to facilitate the selection of hypotheses that might yield insight into the dynamics. We demonstrate this by using clinical trial data for psychotic depression treatment as a case study. The discovered latent structure and proposed outcome are then validated using standard depression criteria, and are shown to provide new insight into the heterogeneity of psychotic depression in terms of predictive symptoms for different interventions.

AB - Unsupervised learning is often used to obtain insight into the underlying structure of medical data, but it is not always clear how to use such structure in an effective way. In this paper, we propose a probabilistic framework for predicting disease dynamics guided by latent states. The framework is based on hidden Markov models and aims to facilitate the selection of hypotheses that might yield insight into the dynamics. We demonstrate this by using clinical trial data for psychotic depression treatment as a case study. The discovered latent structure and proposed outcome are then validated using standard depression criteria, and are shown to provide new insight into the heterogeneity of psychotic depression in terms of predictive symptoms for different interventions.

KW - Machine learning

KW - Psychiatry

KW - Depression

KW - Temporal data

KW - Latent variables

KW - Hidden Markov model

U2 - 10.1016/j.jbi.2019.103232

DO - 10.1016/j.jbi.2019.103232

M3 - Article

VL - 95

JO - Journal of Biomedical Informatics

JF - Journal of Biomedical Informatics

SN - 1532-0464

M1 - 103232

ER -