Probabilistic Modeling of Imaging, Genetics and Diagnosis

Alzheimer's Disease Neuroimaging Initiative

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

We propose a unified Bayesian framework for detecting genetic variants associated with disease by exploiting image-based features as an intermediate phenotype. The use of imaging data for examining genetic associations promises new directions of analysis, but currently the most widely used methods make sub-optimal use of the richness that these data types can offer. Currently, image features are most commonly selected based on their relevance to the disease phenotype. Then, in a separate step, a set of genetic variants is identified to explain the selected features. In contrast, our method performs these tasks simultaneously in order to jointly exploit information in both data types. The analysis yields probabilistic measures of clinical relevance for both imaging and genetic markers. We derive an efficient approximate inference algorithm that handles the high dimensionality of image and genetic data. We evaluate the algorithm on synthetic data and demonstrate that it outperforms traditional models. We also illustrate our method on Alzheimer's Disease Neuroimaging Initiative data.

Original languageEnglish (US)
Article number7404010
Pages (from-to)1765-1779
Number of pages15
JournalIEEE Transactions on Medical Imaging
Volume35
Issue number7
DOIs
StatePublished - Jul 1 2016

Fingerprint

Imaging techniques
Neuroimaging
Phenotype
Genetic Markers
Alzheimer Disease
Genetics
Direction compound

Keywords

  • Bayesian models
  • imaging genetics
  • probabilistic graphical model
  • variational inference

ASJC Scopus subject areas

  • Software
  • Radiological and Ultrasound Technology
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Probabilistic Modeling of Imaging, Genetics and Diagnosis. / Alzheimer's Disease Neuroimaging Initiative.

In: IEEE Transactions on Medical Imaging, Vol. 35, No. 7, 7404010, 01.07.2016, p. 1765-1779.

Research output: Contribution to journalArticle

Alzheimer's Disease Neuroimaging Initiative 2016, 'Probabilistic Modeling of Imaging, Genetics and Diagnosis', IEEE Transactions on Medical Imaging, vol. 35, no. 7, 7404010, pp. 1765-1779. https://doi.org/10.1109/TMI.2016.2527784
Alzheimer's Disease Neuroimaging Initiative. / Probabilistic Modeling of Imaging, Genetics and Diagnosis. In: IEEE Transactions on Medical Imaging. 2016 ; Vol. 35, No. 7. pp. 1765-1779.
@article{253f76851ad042379ba909422598fda4,
title = "Probabilistic Modeling of Imaging, Genetics and Diagnosis",
abstract = "We propose a unified Bayesian framework for detecting genetic variants associated with disease by exploiting image-based features as an intermediate phenotype. The use of imaging data for examining genetic associations promises new directions of analysis, but currently the most widely used methods make sub-optimal use of the richness that these data types can offer. Currently, image features are most commonly selected based on their relevance to the disease phenotype. Then, in a separate step, a set of genetic variants is identified to explain the selected features. In contrast, our method performs these tasks simultaneously in order to jointly exploit information in both data types. The analysis yields probabilistic measures of clinical relevance for both imaging and genetic markers. We derive an efficient approximate inference algorithm that handles the high dimensionality of image and genetic data. We evaluate the algorithm on synthetic data and demonstrate that it outperforms traditional models. We also illustrate our method on Alzheimer's Disease Neuroimaging Initiative data.",
keywords = "Bayesian models, imaging genetics, probabilistic graphical model, variational inference",
author = "{Alzheimer's Disease Neuroimaging Initiative} and Batmanghelich, {Nematollah K.} and Adrian Dalca and Gerald Quon and Mert Sabuncu and Polina Golland",
year = "2016",
month = "7",
day = "1",
doi = "10.1109/TMI.2016.2527784",
language = "English (US)",
volume = "35",
pages = "1765--1779",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "7",

}

TY - JOUR

T1 - Probabilistic Modeling of Imaging, Genetics and Diagnosis

AU - Alzheimer's Disease Neuroimaging Initiative

AU - Batmanghelich, Nematollah K.

AU - Dalca, Adrian

AU - Quon, Gerald

AU - Sabuncu, Mert

AU - Golland, Polina

PY - 2016/7/1

Y1 - 2016/7/1

N2 - We propose a unified Bayesian framework for detecting genetic variants associated with disease by exploiting image-based features as an intermediate phenotype. The use of imaging data for examining genetic associations promises new directions of analysis, but currently the most widely used methods make sub-optimal use of the richness that these data types can offer. Currently, image features are most commonly selected based on their relevance to the disease phenotype. Then, in a separate step, a set of genetic variants is identified to explain the selected features. In contrast, our method performs these tasks simultaneously in order to jointly exploit information in both data types. The analysis yields probabilistic measures of clinical relevance for both imaging and genetic markers. We derive an efficient approximate inference algorithm that handles the high dimensionality of image and genetic data. We evaluate the algorithm on synthetic data and demonstrate that it outperforms traditional models. We also illustrate our method on Alzheimer's Disease Neuroimaging Initiative data.

AB - We propose a unified Bayesian framework for detecting genetic variants associated with disease by exploiting image-based features as an intermediate phenotype. The use of imaging data for examining genetic associations promises new directions of analysis, but currently the most widely used methods make sub-optimal use of the richness that these data types can offer. Currently, image features are most commonly selected based on their relevance to the disease phenotype. Then, in a separate step, a set of genetic variants is identified to explain the selected features. In contrast, our method performs these tasks simultaneously in order to jointly exploit information in both data types. The analysis yields probabilistic measures of clinical relevance for both imaging and genetic markers. We derive an efficient approximate inference algorithm that handles the high dimensionality of image and genetic data. We evaluate the algorithm on synthetic data and demonstrate that it outperforms traditional models. We also illustrate our method on Alzheimer's Disease Neuroimaging Initiative data.

KW - Bayesian models

KW - imaging genetics

KW - probabilistic graphical model

KW - variational inference

UR - http://www.scopus.com/inward/record.url?scp=84978204513&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84978204513&partnerID=8YFLogxK

U2 - 10.1109/TMI.2016.2527784

DO - 10.1109/TMI.2016.2527784

M3 - Article

VL - 35

SP - 1765

EP - 1779

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 7

M1 - 7404010

ER -