A Framework for Combining a Motion Atlas with Non-Motion Information to Learn Clinically Useful Biomarkers: Application to Cardiac Resynchronisation Therapy Response Prediction

Devis Peressutti; Matthew David Maurice Sinclair; Wenjia Bai; Thomas Andrew Jackson; Jacobus Ruijsink; David Nordsletten; Liia Asner; Myrianthi Hadjicharalambous; Christopher Aldo Rinaldi; Daniel Rueckert; Andrew Peter King

doi:10.1016/j.media.2016.10.002

A Framework for Combining a Motion Atlas with Non-Motion Information to Learn Clinically Useful Biomarkers: Application to Cardiac Resynchronisation Therapy Response Prediction

Devis Peressutti, Matthew David Maurice Sinclair, Wenjia Bai, Thomas Andrew Jackson, Jacobus Ruijsink, David Nordsletten, Liia Asner, Myrianthi Hadjicharalambous, Christopher Aldo Rinaldi, Daniel Rueckert, Andrew Peter King

Imperial College London

Research output: Contribution to journal › Article › peer-review

37 Citations (Scopus)

245 Downloads (Pure)

Abstract

We present a framework for combining a cardiac motion atlas with non-motion data. The atlas represents cardiac cycle motion across a number of subjects in a common space based on rich motion descriptors capturing 3D displacement, velocity, strain and strain rate. The non-motion data are derived from a variety of sources such as imaging, electrocardiogram (ECG) and clinical reports. Once in the atlas space, we apply a novel supervised learning approach based on random projections and ensemble learning to learn the relationship between the atlas data and some desired clinical output. We apply our framework to the problem of predicting response to Cardiac Resynchronisation Therapy (CRT). Using a cohort of 34 patients selected for CRT using conventional criteria, results show that the combination of motion and non-motion data enables CRT response to be predicted with 91.2% accuracy (100% sensitivity and 62.5% specificity), which compares favourably with the current state-of-the-art in CRT response prediction.

Original language	English
Pages (from-to)	669–684
Journal	Medical Image Analysis
Volume	35
Early online date	11 Oct 2016
DOIs	https://doi.org/10.1016/j.media.2016.10.002
Publication status	Published - Jan 2017

Access to Document

10.1016/j.media.2016.10.002Licence: CC BY

A Framework for Combining_SINCLAIR_Accepted 6Oct2016_GOLD VoRFinal published version, 3.88 MBLicence: CC BY

Cite this

Peressutti, D., Sinclair, M. D. M., Bai, W., Jackson, T. A., Ruijsink, J., Nordsletten, D., Asner, L., Hadjicharalambous, M., Rinaldi, C. A., Rueckert, D., & King, A. P. (2017). A Framework for Combining a Motion Atlas with Non-Motion Information to Learn Clinically Useful Biomarkers: Application to Cardiac Resynchronisation Therapy Response Prediction. Medical Image Analysis, 35, 669–684. https://doi.org/10.1016/j.media.2016.10.002

@article{f369f517bc5e4e3e8207ccd1afa7e9b4,

title = "A Framework for Combining a Motion Atlas with Non-Motion Information to Learn Clinically Useful Biomarkers: Application to Cardiac Resynchronisation Therapy Response Prediction",

abstract = "We present a framework for combining a cardiac motion atlas with non-motion data. The atlas represents cardiac cycle motion across a number of subjects in a common space based on rich motion descriptors capturing 3D displacement, velocity, strain and strain rate. The non-motion data are derived from a variety of sources such as imaging, electrocardiogram (ECG) and clinical reports. Once in the atlas space, we apply a novel supervised learning approach based on random projections and ensemble learning to learn the relationship between the atlas data and some desired clinical output. We apply our framework to the problem of predicting response to Cardiac Resynchronisation Therapy (CRT). Using a cohort of 34 patients selected for CRT using conventional criteria, results show that the combination of motion and non-motion data enables CRT response to be predicted with 91.2% accuracy (100% sensitivity and 62.5% specificity), which compares favourably with the current state-of-the-art in CRT response prediction.",

author = "Devis Peressutti and Sinclair, {Matthew David Maurice} and Wenjia Bai and Jackson, {Thomas Andrew} and Jacobus Ruijsink and David Nordsletten and Liia Asner and Myrianthi Hadjicharalambous and Rinaldi, {Christopher Aldo} and Daniel Rueckert and King, {Andrew Peter}",

year = "2017",

month = jan,

doi = "10.1016/j.media.2016.10.002",

language = "English",

volume = "35",

pages = "669–684",

journal = "Medical Image Analysis",

issn = "1361-8415",

publisher = "Elsevier",

}

Peressutti, D , Sinclair, MDM, Bai, W, Jackson, TA , Ruijsink, J , Nordsletten, D , Asner, L , Hadjicharalambous, M , Rinaldi, CA, Rueckert, D & King, AP 2017, 'A Framework for Combining a Motion Atlas with Non-Motion Information to Learn Clinically Useful Biomarkers: Application to Cardiac Resynchronisation Therapy Response Prediction', Medical Image Analysis, vol. 35, pp. 669–684. https://doi.org/10.1016/j.media.2016.10.002

A Framework for Combining a Motion Atlas with Non-Motion Information to Learn Clinically Useful Biomarkers: Application to Cardiac Resynchronisation Therapy Response Prediction. / Peressutti, Devis ; Sinclair, Matthew David Maurice; Bai, Wenjia et al.
In: Medical Image Analysis, Vol. 35, 01.2017, p. 669–684.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - A Framework for Combining a Motion Atlas with Non-Motion Information to Learn Clinically Useful Biomarkers

T2 - Application to Cardiac Resynchronisation Therapy Response Prediction

AU - Peressutti, Devis

AU - Sinclair, Matthew David Maurice

AU - Bai, Wenjia

AU - Jackson, Thomas Andrew

AU - Ruijsink, Jacobus

AU - Nordsletten, David

AU - Asner, Liia

AU - Hadjicharalambous, Myrianthi

AU - Rinaldi, Christopher Aldo

AU - Rueckert, Daniel

AU - King, Andrew Peter

PY - 2017/1

Y1 - 2017/1

N2 - We present a framework for combining a cardiac motion atlas with non-motion data. The atlas represents cardiac cycle motion across a number of subjects in a common space based on rich motion descriptors capturing 3D displacement, velocity, strain and strain rate. The non-motion data are derived from a variety of sources such as imaging, electrocardiogram (ECG) and clinical reports. Once in the atlas space, we apply a novel supervised learning approach based on random projections and ensemble learning to learn the relationship between the atlas data and some desired clinical output. We apply our framework to the problem of predicting response to Cardiac Resynchronisation Therapy (CRT). Using a cohort of 34 patients selected for CRT using conventional criteria, results show that the combination of motion and non-motion data enables CRT response to be predicted with 91.2% accuracy (100% sensitivity and 62.5% specificity), which compares favourably with the current state-of-the-art in CRT response prediction.

AB - We present a framework for combining a cardiac motion atlas with non-motion data. The atlas represents cardiac cycle motion across a number of subjects in a common space based on rich motion descriptors capturing 3D displacement, velocity, strain and strain rate. The non-motion data are derived from a variety of sources such as imaging, electrocardiogram (ECG) and clinical reports. Once in the atlas space, we apply a novel supervised learning approach based on random projections and ensemble learning to learn the relationship between the atlas data and some desired clinical output. We apply our framework to the problem of predicting response to Cardiac Resynchronisation Therapy (CRT). Using a cohort of 34 patients selected for CRT using conventional criteria, results show that the combination of motion and non-motion data enables CRT response to be predicted with 91.2% accuracy (100% sensitivity and 62.5% specificity), which compares favourably with the current state-of-the-art in CRT response prediction.

U2 - 10.1016/j.media.2016.10.002

DO - 10.1016/j.media.2016.10.002

M3 - Article

SN - 1361-8415

VL - 35

SP - 669

EP - 684

JO - Medical Image Analysis

JF - Medical Image Analysis

ER -

A Framework for Combining a Motion Atlas with Non-Motion Information to Learn Clinically Useful Biomarkers: Application to Cardiac Resynchronisation Therapy Response Prediction

Abstract

Access to Document

Fingerprint

Cite this