Prediction of longterm outcome of neuropsychological tests of mtbi patients using imaging features

Shervin Minaee, Yao Wang, Yvonne W. Lui

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

Abstract

Mild Traumatic Brain Injury (MTBI) is a growing public health problem with an underestimated incidence of over one million people annually in the U.S. Neuropsychological tests are used to both assess the patient condition and to monitor patient progress. This work aims to use features extracted from MR images taken shortly after injury to predict the performance of MTBI patients on neuropsychological tests one year after injury. Successful prediction can enable early patient stratification and proper treatment planning. The goal of this project is to determine the most effective features and corresponding prediction method for such prediction. The main challenge is that we have only limited training data, from which we need to develop the prediction method that can be expected to provide accurate prediction results for unseen data. In the machine learning literature, feature selection is usually done to minimize the cross validation error, which still has a chance for overfitting if the available data set is small and noisy. We propose a novel criterion for feature selection, which considers both the cross validation error and the prediction model variance, to further reduce the chance for overfitting. The algorithm is applied to a data set of 15 MTBI patients. The proposed method was able to determine a subset of MR image features for predicting each neuropsychological test to yield both small prediction error and prediction variance.

Original languageEnglish (US)
Title of host publication2013 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2013
PublisherIEEE Computer Society
ISBN (Print)9781479930074
DOIs
StatePublished - 2013
Event2013 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2013 - Brooklyn, NY, United States
Duration: Dec 7 2013Dec 7 2013

Other

Other2013 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2013
CountryUnited States
CityBrooklyn, NY
Period12/7/1312/7/13

Fingerprint

Imaging techniques
Brain
Feature extraction
Public health
Medical problems
Learning systems
Planning

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Minaee, S., Wang, Y., & Lui, Y. W. (2013). Prediction of longterm outcome of neuropsychological tests of mtbi patients using imaging features. In 2013 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2013 [6736783] IEEE Computer Society. https://doi.org/10.1109/SPMB.2013.6736783

Prediction of longterm outcome of neuropsychological tests of mtbi patients using imaging features. / Minaee, Shervin; Wang, Yao; Lui, Yvonne W.

2013 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2013. IEEE Computer Society, 2013. 6736783.

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

Minaee, S, Wang, Y & Lui, YW 2013, Prediction of longterm outcome of neuropsychological tests of mtbi patients using imaging features. in 2013 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2013., 6736783, IEEE Computer Society, 2013 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2013, Brooklyn, NY, United States, 12/7/13. https://doi.org/10.1109/SPMB.2013.6736783
Minaee S, Wang Y, Lui YW. Prediction of longterm outcome of neuropsychological tests of mtbi patients using imaging features. In 2013 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2013. IEEE Computer Society. 2013. 6736783 https://doi.org/10.1109/SPMB.2013.6736783
Minaee, Shervin ; Wang, Yao ; Lui, Yvonne W. / Prediction of longterm outcome of neuropsychological tests of mtbi patients using imaging features. 2013 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2013. IEEE Computer Society, 2013.
@inproceedings{bb8c14471e1242f0be2fb3e757ece133,
title = "Prediction of longterm outcome of neuropsychological tests of mtbi patients using imaging features",
abstract = "Mild Traumatic Brain Injury (MTBI) is a growing public health problem with an underestimated incidence of over one million people annually in the U.S. Neuropsychological tests are used to both assess the patient condition and to monitor patient progress. This work aims to use features extracted from MR images taken shortly after injury to predict the performance of MTBI patients on neuropsychological tests one year after injury. Successful prediction can enable early patient stratification and proper treatment planning. The goal of this project is to determine the most effective features and corresponding prediction method for such prediction. The main challenge is that we have only limited training data, from which we need to develop the prediction method that can be expected to provide accurate prediction results for unseen data. In the machine learning literature, feature selection is usually done to minimize the cross validation error, which still has a chance for overfitting if the available data set is small and noisy. We propose a novel criterion for feature selection, which considers both the cross validation error and the prediction model variance, to further reduce the chance for overfitting. The algorithm is applied to a data set of 15 MTBI patients. The proposed method was able to determine a subset of MR image features for predicting each neuropsychological test to yield both small prediction error and prediction variance.",
author = "Shervin Minaee and Yao Wang and Lui, {Yvonne W.}",
year = "2013",
doi = "10.1109/SPMB.2013.6736783",
language = "English (US)",
isbn = "9781479930074",
booktitle = "2013 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2013",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Prediction of longterm outcome of neuropsychological tests of mtbi patients using imaging features

AU - Minaee, Shervin

AU - Wang, Yao

AU - Lui, Yvonne W.

PY - 2013

Y1 - 2013

N2 - Mild Traumatic Brain Injury (MTBI) is a growing public health problem with an underestimated incidence of over one million people annually in the U.S. Neuropsychological tests are used to both assess the patient condition and to monitor patient progress. This work aims to use features extracted from MR images taken shortly after injury to predict the performance of MTBI patients on neuropsychological tests one year after injury. Successful prediction can enable early patient stratification and proper treatment planning. The goal of this project is to determine the most effective features and corresponding prediction method for such prediction. The main challenge is that we have only limited training data, from which we need to develop the prediction method that can be expected to provide accurate prediction results for unseen data. In the machine learning literature, feature selection is usually done to minimize the cross validation error, which still has a chance for overfitting if the available data set is small and noisy. We propose a novel criterion for feature selection, which considers both the cross validation error and the prediction model variance, to further reduce the chance for overfitting. The algorithm is applied to a data set of 15 MTBI patients. The proposed method was able to determine a subset of MR image features for predicting each neuropsychological test to yield both small prediction error and prediction variance.

AB - Mild Traumatic Brain Injury (MTBI) is a growing public health problem with an underestimated incidence of over one million people annually in the U.S. Neuropsychological tests are used to both assess the patient condition and to monitor patient progress. This work aims to use features extracted from MR images taken shortly after injury to predict the performance of MTBI patients on neuropsychological tests one year after injury. Successful prediction can enable early patient stratification and proper treatment planning. The goal of this project is to determine the most effective features and corresponding prediction method for such prediction. The main challenge is that we have only limited training data, from which we need to develop the prediction method that can be expected to provide accurate prediction results for unseen data. In the machine learning literature, feature selection is usually done to minimize the cross validation error, which still has a chance for overfitting if the available data set is small and noisy. We propose a novel criterion for feature selection, which considers both the cross validation error and the prediction model variance, to further reduce the chance for overfitting. The algorithm is applied to a data set of 15 MTBI patients. The proposed method was able to determine a subset of MR image features for predicting each neuropsychological test to yield both small prediction error and prediction variance.

UR - http://www.scopus.com/inward/record.url?scp=84897687510&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897687510&partnerID=8YFLogxK

U2 - 10.1109/SPMB.2013.6736783

DO - 10.1109/SPMB.2013.6736783

M3 - Conference contribution

SN - 9781479930074

BT - 2013 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2013

PB - IEEE Computer Society

ER -