Improving MRI brain image classification with anatomical regional kernels

TY - CHAP

T1 - Improving MRI brain image classification with anatomical regional kernels

AU - Young, Jonathan

AU - Mendelson, Alex

AU - Cardoso, M. Jorge

AU - Modat, Marc

AU - Ashburner, John

AU - Ourselin, Sebastien

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Classification of brain images is frequently done using kernel based methods, such as the support vector machine. These lend themselves to improvement via multiple kernel learning, where a number of different kernels are linearly combined to integrate different sources of information and increase accuracy. Previous applications made use of a small number of kernels representing different image modalities or kernel functions. Here, the kernels instead represent 83 anatomically meaningful brain regions. To find the optimal combination of kernels and perform classification, we use a Gaussian Process framework to infer the maximum likelihood weights. The resulting formulation successfully combines voxel level features with prior anatomical knowledge. This gives an improvement in classification accuracy of MRI images of Alzheimer’s disease patients and healthy controls from the ADNI database to almost 88%, compared to less than 86% using a single kernel representing the whole brain. Moreover, interpretability of the classifier is also improved, as the optimal kernel weights are sparse and give an indication of the importance of each brain region in separating the two groups.

AB - Classification of brain images is frequently done using kernel based methods, such as the support vector machine. These lend themselves to improvement via multiple kernel learning, where a number of different kernels are linearly combined to integrate different sources of information and increase accuracy. Previous applications made use of a small number of kernels representing different image modalities or kernel functions. Here, the kernels instead represent 83 anatomically meaningful brain regions. To find the optimal combination of kernels and perform classification, we use a Gaussian Process framework to infer the maximum likelihood weights. The resulting formulation successfully combines voxel level features with prior anatomical knowledge. This gives an improvement in classification accuracy of MRI images of Alzheimer’s disease patients and healthy controls from the ADNI database to almost 88%, compared to less than 86% using a single kernel representing the whole brain. Moreover, interpretability of the classifier is also improved, as the optimal kernel weights are sparse and give an indication of the importance of each brain region in separating the two groups.

KW - Alzheimer’s disease

KW - Classification

KW - Gaussian processes

KW - Interpretability

KW - MRI

KW - Multi-kernel learning

UR - http://www.scopus.com/inward/record.url?scp=84955262892&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-27929-9_5

DO - 10.1007/978-3-319-27929-9_5

M3 - Conference paper

AN - SCOPUS:84955262892

SN - 9783319279282

VL - 9487

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 45

EP - 53

BT - Machine Learning Meets Medical Imaging - 1st International Workshop, MLMMI 2015 Held in Conjunction with ICML 2015, Revised Selected Papers

PB - Springer Verlag

T2 - 1st International Workshop on Machine Learning Meets Medical Imaging, MLMMI 2015

Y2 - 11 July 2015 through 11 July 2015

ER -