Show simple item record

dc.contributor.authorLee, Kisuk, Ph. D. Massachusetts Institute of Technology.en_US
dc.contributor.otherMassachusetts Institute of Technology. Department of Brain and Cognitive Sciences.en_US
dc.date.accessioned2021-10-22T20:27:20Z
dc.date.available2021-10-22T20:27:20Z
dc.date.copyright2019en_US
dc.date.issued2019en_US
dc.identifier.urihttps://hdl.handle.net/1721.1/133076
dc.descriptionThis electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.en_US
dc.descriptionThesis: Ph. D. in Computation, Massachusetts Institute of Technology, Department of Brain and Cognitive Sciences, June, 2019en_US
dc.descriptionCataloged from student-submitted PDF version of thesis.en_US
dc.descriptionIncludes bibliographical references (pages 155-167).en_US
dc.description.abstractNeuronal connectivity can be reconstructed from a 3D electron microscopy (EM) image of a brain volume. A challenging and important subproblem is the segmentation of the image into neurons. For the past decade, convolutional networks have been used for 3D reconstruction of neurons from EM brain images. In this thesis, we develop a set of deep learning algorithms based on convolutional nets for automated reconstruction of neurons, with particular focus on highly anisotropic images of brain tissue acquired by serial section EM (ssEM). In the first part of the thesis, we propose a recursively trained hybrid 2D-3D convolutional net architecture, and demonstrate the feasibility of exploiting 3D context to further improve boundary detection accuracy despite the high anisotropy of ssEM images. In the following parts, we propose two techniques for training convolutional nets that can substantially improve boundary detection accuracy. First, we introduce novel forms of training data augmentation based on simulation of known types of image defects such as misalignments, missing sections, and out-of-focus sections. Second, we add the auxiliary task of predicting affinities between nonneighboring voxels, reflecting the structured nature of neuronal boundary detection. We demonstrate the effectiveness of the proposed techniques on large-scale ssEM images acquired from the mouse primary visual cortex. Lastly, we take a radical departure from simple boundary detection by exploring an alternative approach to object-centered representation, that is, learning dense voxel embeddings via deep metric learning. Convolutional nets are trained to generate dense voxel embeddings by assigning similar vectors to voxels within the same objects and well-separated vectors to voxels from different objects. Our proposed method achieves state-of-the-art accuracy with substantial improvements on very thin objects.en_US
dc.description.statementofresponsibilityby Kisuk Lee.en_US
dc.format.extent167 pagesen_US
dc.language.isoengen_US
dc.publisherMassachusetts Institute of Technologyen_US
dc.rightsMIT theses may be protected by copyright. Please reuse MIT thesis content according to the MIT Libraries Permissions Policy, which is available through the URL provided.en_US
dc.rights.urihttp://dspace.mit.edu/handle/1721.1/7582en_US
dc.subjectBrain and Cognitive Sciences.en_US
dc.titleReconstructing neurons from serial section electron microscopy imagesen_US
dc.typeThesisen_US
dc.description.degreePh. D. in Computationen_US
dc.contributor.departmentMassachusetts Institute of Technology. Department of Brain and Cognitive Sciencesen_US
dc.identifier.oclc1264708701en_US
dc.description.collectionPh.D.inComputation Massachusetts Institute of Technology, Department of Brain and Cognitive Sciencesen_US
dspace.imported2021-10-22T20:27:20Zen_US
mit.thesis.degreeDoctoralen_US
mit.thesis.departmentBrainen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record