Predicting age at loss of ambulation in Duchenne muscular dystrophy with deep phenotypic measures

Yinxue Wang, Luca Bello, Yue Wang, Craig M McDonald, Eric P. Hoffman, Guoqiang Yu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Although Duchenne muscular dystrophy (DMD), the most common single-gene lethal disorder, is caused by a homogeneous biochemical defect in all patients, substantial patient-patient variety in disease progression is observed. The loss of ambulation (LoA) is a functional milestone of DMD progression and the age at LoA is often used as an indication of disease severity. But as age at LoA is not always available, such as when patients remain ambulant at study end, its use has been limited. In this paper, we report machine learning approaches to predict age at LoA based on clinical measures of muscular strength and motor function, and validate the algorithms using the CINRG dataset. With extensive experiments and rigorous statistical analysis, we found that (1) the utilization of multiple clinical features yields better prediction than using any of the single measures, and (2) the prediction based on Lasso is more accurate than other multivariate analytical approaches such as ordinary least squares and ridge regression. To our knowledge, we are the first to provide point predictions for age at LoA in DMD using clinical phenotypic measures. Importantly, we find that not all clinical measures contribute to the prediction. Age at the last visit (before LoA), velocity of walking 10 meters, and velocity of climbing 4 steps are selected as important predictors by Lasso. The usefulness of the prediction model is illustrated with evidence that the association between a well-known modifier of DMD severity and age at LoA has better power when the predicted values are utilized.

Original languageEnglish (US)
Title of host publication2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1344-1347
Number of pages4
ISBN (Print)9781479970889
DOIs
StatePublished - Feb 5 2014
Event2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014 - Atlanta, United States
Duration: Dec 3 2014Dec 5 2014

Other

Other2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014
CountryUnited States
CityAtlanta
Period12/3/1412/5/14

Fingerprint

Learning systems
Statistical methods
Genes
Defects
Experiments

ASJC Scopus subject areas

  • Signal Processing
  • Information Systems

Cite this

Wang, Y., Bello, L., Wang, Y., McDonald, C. M., Hoffman, E. P., & Yu, G. (2014). Predicting age at loss of ambulation in Duchenne muscular dystrophy with deep phenotypic measures. In 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014 (pp. 1344-1347). [7032343] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GlobalSIP.2014.7032343

Predicting age at loss of ambulation in Duchenne muscular dystrophy with deep phenotypic measures. / Wang, Yinxue; Bello, Luca; Wang, Yue; McDonald, Craig M; Hoffman, Eric P.; Yu, Guoqiang.

2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 1344-1347 7032343.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wang, Y, Bello, L, Wang, Y, McDonald, CM, Hoffman, EP & Yu, G 2014, Predicting age at loss of ambulation in Duchenne muscular dystrophy with deep phenotypic measures. in 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014., 7032343, Institute of Electrical and Electronics Engineers Inc., pp. 1344-1347, 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014, Atlanta, United States, 12/3/14. https://doi.org/10.1109/GlobalSIP.2014.7032343
Wang Y, Bello L, Wang Y, McDonald CM, Hoffman EP, Yu G. Predicting age at loss of ambulation in Duchenne muscular dystrophy with deep phenotypic measures. In 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014. Institute of Electrical and Electronics Engineers Inc. 2014. p. 1344-1347. 7032343 https://doi.org/10.1109/GlobalSIP.2014.7032343
Wang, Yinxue ; Bello, Luca ; Wang, Yue ; McDonald, Craig M ; Hoffman, Eric P. ; Yu, Guoqiang. / Predicting age at loss of ambulation in Duchenne muscular dystrophy with deep phenotypic measures. 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 1344-1347
@inproceedings{a9639c865a67466792eb3d428039f3e6,
title = "Predicting age at loss of ambulation in Duchenne muscular dystrophy with deep phenotypic measures",
abstract = "Although Duchenne muscular dystrophy (DMD), the most common single-gene lethal disorder, is caused by a homogeneous biochemical defect in all patients, substantial patient-patient variety in disease progression is observed. The loss of ambulation (LoA) is a functional milestone of DMD progression and the age at LoA is often used as an indication of disease severity. But as age at LoA is not always available, such as when patients remain ambulant at study end, its use has been limited. In this paper, we report machine learning approaches to predict age at LoA based on clinical measures of muscular strength and motor function, and validate the algorithms using the CINRG dataset. With extensive experiments and rigorous statistical analysis, we found that (1) the utilization of multiple clinical features yields better prediction than using any of the single measures, and (2) the prediction based on Lasso is more accurate than other multivariate analytical approaches such as ordinary least squares and ridge regression. To our knowledge, we are the first to provide point predictions for age at LoA in DMD using clinical phenotypic measures. Importantly, we find that not all clinical measures contribute to the prediction. Age at the last visit (before LoA), velocity of walking 10 meters, and velocity of climbing 4 steps are selected as important predictors by Lasso. The usefulness of the prediction model is illustrated with evidence that the association between a well-known modifier of DMD severity and age at LoA has better power when the predicted values are utilized.",
author = "Yinxue Wang and Luca Bello and Yue Wang and McDonald, {Craig M} and Hoffman, {Eric P.} and Guoqiang Yu",
year = "2014",
month = "2",
day = "5",
doi = "10.1109/GlobalSIP.2014.7032343",
language = "English (US)",
isbn = "9781479970889",
pages = "1344--1347",
booktitle = "2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Predicting age at loss of ambulation in Duchenne muscular dystrophy with deep phenotypic measures

AU - Wang, Yinxue

AU - Bello, Luca

AU - Wang, Yue

AU - McDonald, Craig M

AU - Hoffman, Eric P.

AU - Yu, Guoqiang

PY - 2014/2/5

Y1 - 2014/2/5

N2 - Although Duchenne muscular dystrophy (DMD), the most common single-gene lethal disorder, is caused by a homogeneous biochemical defect in all patients, substantial patient-patient variety in disease progression is observed. The loss of ambulation (LoA) is a functional milestone of DMD progression and the age at LoA is often used as an indication of disease severity. But as age at LoA is not always available, such as when patients remain ambulant at study end, its use has been limited. In this paper, we report machine learning approaches to predict age at LoA based on clinical measures of muscular strength and motor function, and validate the algorithms using the CINRG dataset. With extensive experiments and rigorous statistical analysis, we found that (1) the utilization of multiple clinical features yields better prediction than using any of the single measures, and (2) the prediction based on Lasso is more accurate than other multivariate analytical approaches such as ordinary least squares and ridge regression. To our knowledge, we are the first to provide point predictions for age at LoA in DMD using clinical phenotypic measures. Importantly, we find that not all clinical measures contribute to the prediction. Age at the last visit (before LoA), velocity of walking 10 meters, and velocity of climbing 4 steps are selected as important predictors by Lasso. The usefulness of the prediction model is illustrated with evidence that the association between a well-known modifier of DMD severity and age at LoA has better power when the predicted values are utilized.

AB - Although Duchenne muscular dystrophy (DMD), the most common single-gene lethal disorder, is caused by a homogeneous biochemical defect in all patients, substantial patient-patient variety in disease progression is observed. The loss of ambulation (LoA) is a functional milestone of DMD progression and the age at LoA is often used as an indication of disease severity. But as age at LoA is not always available, such as when patients remain ambulant at study end, its use has been limited. In this paper, we report machine learning approaches to predict age at LoA based on clinical measures of muscular strength and motor function, and validate the algorithms using the CINRG dataset. With extensive experiments and rigorous statistical analysis, we found that (1) the utilization of multiple clinical features yields better prediction than using any of the single measures, and (2) the prediction based on Lasso is more accurate than other multivariate analytical approaches such as ordinary least squares and ridge regression. To our knowledge, we are the first to provide point predictions for age at LoA in DMD using clinical phenotypic measures. Importantly, we find that not all clinical measures contribute to the prediction. Age at the last visit (before LoA), velocity of walking 10 meters, and velocity of climbing 4 steps are selected as important predictors by Lasso. The usefulness of the prediction model is illustrated with evidence that the association between a well-known modifier of DMD severity and age at LoA has better power when the predicted values are utilized.

UR - http://www.scopus.com/inward/record.url?scp=84949927284&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949927284&partnerID=8YFLogxK

U2 - 10.1109/GlobalSIP.2014.7032343

DO - 10.1109/GlobalSIP.2014.7032343

M3 - Conference contribution

AN - SCOPUS:84949927284

SN - 9781479970889

SP - 1344

EP - 1347

BT - 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -