TY - JOUR
T1 - An atlas of classifiers—a machine learning paradigm for brain MRI segmentation
AU - Gordon, Shiri
AU - Kodner, Boris
AU - Goldfryd, Tal
AU - Sidorov, Michael
AU - Goldberger, Jacob
AU - Raviv, Tammy Riklin
N1 - Funding Information:
This study was partially supported by the Israel Science Foundation (1638/16, T.R.R) and by the the Israeli Ministry of Science & Technology (J.G.).
Publisher Copyright:
© 2021, International Federation for Medical and Biological Engineering.
PY - 2021/9/1
Y1 - 2021/9/1
N2 - We present the Atlas of Classifiers (AoC)—a conceptually novel framework for brain MRI segmentation. The AoC is a spatial map of voxel-wise multinomial logistic regression (LR) functions learned from the labeled data. Upon convergence, the resulting fixed LR weights, a few for each voxel, represent the training dataset. It can, therefore, be considered as a light-weight learning machine, which despite its low capacity does not underfit the problem. The AoC construction is independent of the actual intensities of the test images, providing the flexibility to train it on the available labeled data and use it for the segmentation of images from different datasets and modalities. In this sense, it does not overfit the training data, as well. The proposed method has been applied to numerous publicly available datasets for the segmentation of brain MRI tissues and is shown to be robust to noise and outreach commonly used methods. Promising results were also obtained for multi-modal, cross-modality MRI segmentation. Finally, we show how AoC trained on brain MRIs of healthy subjects can be exploited for lesion segmentation of multiple sclerosis patients. [Figure not available: see fulltext.]
AB - We present the Atlas of Classifiers (AoC)—a conceptually novel framework for brain MRI segmentation. The AoC is a spatial map of voxel-wise multinomial logistic regression (LR) functions learned from the labeled data. Upon convergence, the resulting fixed LR weights, a few for each voxel, represent the training dataset. It can, therefore, be considered as a light-weight learning machine, which despite its low capacity does not underfit the problem. The AoC construction is independent of the actual intensities of the test images, providing the flexibility to train it on the available labeled data and use it for the segmentation of images from different datasets and modalities. In this sense, it does not overfit the training data, as well. The proposed method has been applied to numerous publicly available datasets for the segmentation of brain MRI tissues and is shown to be robust to noise and outreach commonly used methods. Promising results were also obtained for multi-modal, cross-modality MRI segmentation. Finally, we show how AoC trained on brain MRIs of healthy subjects can be exploited for lesion segmentation of multiple sclerosis patients. [Figure not available: see fulltext.]
KW - Brain MRI
KW - Logistic regression classifiers
KW - MS-lesions
KW - Machine learning
KW - Segmentation
UR - http://www.scopus.com/inward/record.url?scp=85111398775&partnerID=8YFLogxK
U2 - 10.1007/s11517-021-02414-x
DO - 10.1007/s11517-021-02414-x
M3 - Article
C2 - 34313921
AN - SCOPUS:85111398775
SN - 0140-0118
VL - 59
SP - 1833
EP - 1849
JO - Medical and Biological Engineering and Computing
JF - Medical and Biological Engineering and Computing
IS - 9
ER -