TY - JOUR
T1 - Learning and combining image neighborhoods using random forests for neonatal brain disease classification
AU - Zimmer, Veronika A.
AU - Glocker, Ben
AU - Hahner, Nadine
AU - Eixarch, Elisenda
AU - Sanroma, Gerard
AU - Gratacós, Eduard
AU - Rueckert, Daniel
AU - González Ballester, Miguel Ángel
AU - Piella, Gemma
N1 - Publisher Copyright:
© 2017 Elsevier B.V.
PY - 2017/12
Y1 - 2017/12
N2 - It is challenging to characterize and classify normal and abnormal brain development during early childhood. To reduce the complexity of heterogeneous data population, manifold learning techniques are increasingly applied, which find a low-dimensional representation of the data, while preserving all relevant information. The neighborhood definition used for constructing manifold representations of the population is crucial for preserving the similarity structure and it is highly application dependent. The recently proposed neighborhood approximation forests learn a neighborhood structure in a dataset based on a user-defined distance. We propose a framework to learn multiple pairwise distances in a population of brain images and to combine them in an unsupervised manner optimally in a manifold learning step. Unlike other methods that only use a univariate distance measure, our method allows for a natural combination of multiple distances from heterogeneous sources. As a result, it yields a representation of the population that preserves the multiple distances. Furthermore, our method also selects the most predictive features associated with the distances. We evaluate our method in neonatal magnetic resonance images of three groups (term controls, patients affected by intrauterine growth restriction and mild isolated ventriculomegaly). We show that combining multiple distances related to the condition improves the overall characterization and classification of the three clinical groups compared to the use of single distances and classical unsupervised manifold learning.
AB - It is challenging to characterize and classify normal and abnormal brain development during early childhood. To reduce the complexity of heterogeneous data population, manifold learning techniques are increasingly applied, which find a low-dimensional representation of the data, while preserving all relevant information. The neighborhood definition used for constructing manifold representations of the population is crucial for preserving the similarity structure and it is highly application dependent. The recently proposed neighborhood approximation forests learn a neighborhood structure in a dataset based on a user-defined distance. We propose a framework to learn multiple pairwise distances in a population of brain images and to combine them in an unsupervised manner optimally in a manifold learning step. Unlike other methods that only use a univariate distance measure, our method allows for a natural combination of multiple distances from heterogeneous sources. As a result, it yields a representation of the population that preserves the multiple distances. Furthermore, our method also selects the most predictive features associated with the distances. We evaluate our method in neonatal magnetic resonance images of three groups (term controls, patients affected by intrauterine growth restriction and mild isolated ventriculomegaly). We show that combining multiple distances related to the condition improves the overall characterization and classification of the three clinical groups compared to the use of single distances and classical unsupervised manifold learning.
KW - Brain development
KW - Manifold learning
KW - Neighborhood approximation forest
KW - Random forest
KW - Similarity measure
UR - http://www.scopus.com/inward/record.url?scp=85027845423&partnerID=8YFLogxK
U2 - 10.1016/j.media.2017.08.004
DO - 10.1016/j.media.2017.08.004
M3 - Article
C2 - 28818743
AN - SCOPUS:85027845423
SN - 1361-8415
VL - 42
SP - 189
EP - 199
JO - Medical Image Analysis
JF - Medical Image Analysis
ER -