Multi-atlas learner fusion: An efficient segmentation approach for large-scale data

• Published in: Medical image analysis Vol. 26; no. 1; pp. 82 - 91
• Main Authors: Asman, Andrew J., Huo, Yuankai, Plassard, Andrew J., Landman, Bennett A.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.12.2015
• Subjects: AdaBoost; Algorithms; Brain - anatomy & histology; Brain - physiology; Connectome - methods; Databases, Factual; Humans; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Information Storage and Retrieval - methods; Machine learning; Magnetic Resonance Imaging - methods; Multi-atlas learner fusion; Multi-atlas segmentation; Multimodal Imaging - methods; Pattern Recognition, Automated - methods; Reproducibility of Results; Sensitivity and Specificity; Subtraction Technique; Multi-atlas segmentation; Multi-atlas learner fusion; AdaBoost; Machine learning
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2015.08.010

•We build the multi-atlas learner fusion (MLF) framework for mapping weak initial segmentations to the more accurate multi-atlas segmentation.•The MLF framework cuts the runtime from 36 h down to 3–8 min.•We demonstrate significant increases in the reproducibility of intra-subject segmentations.•We show the large-scale data model significantly improve the segmentation over the small-scale model under the MLF framework.•The MLF framework has comparable performance as state-of-the-art multi-atlas segmentation algorithms without using non-local information. We propose multi-atlas learner fusion (MLF), a framework for rapidly and accurately replicating the highly accurate, yet computationally expensive, multi-atlas segmentation framework based on fusing local learners. In the largest whole-brain multi-atlas study yet reported, multi-atlas segmentations are estimated for a training set of 3464 MR brain images. Using these multi-atlas estimates we (1) estimate a low-dimensional representation for selecting locally appropriate example images, and (2) build AdaBoost learners that map a weak initial segmentation to the multi-atlas segmentation result. Thus, to segment a new target image we project the image into the low-dimensional space, construct a weak initial segmentation, and fuse the trained, locally selected, learners. The MLF framework cuts the runtime on a modern computer from 36 h down to 3–8 min – a 270× speedup – by completely bypassing the need for deformable atlas-target registrations. Additionally, we (1) describe a technique for optimizing the weak initial segmentation and the AdaBoost learning parameters, (2) quantify the ability to replicate the multi-atlas result with mean accuracies approaching the multi-atlas intra-subject reproducibility on a testing set of 380 images, (3) demonstrate significant increases in the reproducibility of intra-subject segmentations when compared to a state-of-the-art multi-atlas framework on a separate reproducibility dataset, (4) show that under the MLF framework the large-scale data model significantly improve the segmentation over the small-scale model under the MLF framework, and (5) indicate that the MLF framework has comparable performance as state-of-the-art multi-atlas segmentation algorithms without using non-local information. [Display omitted]