Research Areas
Authored Publications
Google Publications
Other Publications
Sort By
Preview abstract
Early machine-learning systems were inspired by neural networks — now AI might allow neuroscientists to get to grips with the brain’s unique complexities.
View details
Multi-Layered Maps of Neuropil with Segmentation Guided Contrastive Learning
Sven Dorkenwald
Daniel R. Berger
Agnes L. Bodor
Forrest Collman
Casey M. Schneider-Mizell
Nuno Maçarico da Costa
Jeff W. Lichtman
Nature Methods (2023)
Preview abstract
Maps of the nervous system that identify individual cells along with their type, subcellular components and connectivity have the potential to elucidate fundamental organizational principles of neural circuits. Nanometer-resolution imaging of brain tissue provides the necessary raw data, but inferring cellular and subcellular annotation layers is challenging. We present segmentation-guided contrastive learning of representations (SegCLR), a self-supervised machine learning technique that produces representations of cells directly from 3D imagery and segmentations. When applied to volumes of human and mouse cortex, SegCLR enables accurate classification of cellular subcompartments and achieves performance equivalent to a supervised approach while requiring 400-fold fewer labeled examples. SegCLR also enables inference of cell types from fragments as small as 10 μm, which enhances the utility of volumes in which many neurites are truncated at boundaries. Finally, SegCLR enables exploration of layer 5 pyramidal cell subtypes and automated large-scale analysis of synaptic partners in mouse visual cortex.
View details
Structured sampling of olfactory input by the fly mushroom body
Zhihao Zheng
Feng Li
Corey Fisher
Iqbal J. Ali
Nadiya Sharifi
Steven Calle-Schuler
Joseph Hsu
Najla Masoodpanah
Lucia Kmecova
Tom Kazimiers
Eric Perlman
Matthew Nichols
Davi Bock
Current Biology, vol. 32 (2022), pp. 3334-3349
Preview abstract
Associative memory formation and recall in the fruit fly Drosophila melanogaster is subserved by the mushroom body (MB). Upon arrival in the MB, sensory information undergoes a profound transformation from broadly tuned and stereotyped odorant responses in the olfactory projection neuron (PN) layer to narrowly tuned and nonstereotyped responses in the Kenyon cells (KCs). Theory and experiment suggest that this transformation is implemented by random connectivity between KCs and PNs. However, this hypothesis has been challenging to test, given the difficulty of mapping synaptic connections between large numbers of brain-spanning neurons. Here, we used a recent whole-brain electron microscopy volume of the adult fruit fly to map PN-to-KC connectivity at synaptic resolution. The PN-KC connectome revealed unexpected structure, with preponderantly food-responsive PN types converging at above-chance levels on downstream KCs. Axons of the overconvergent PN types tended to arborize near one another in the MB main calyx, making local KC dendrites more likely to receive input from those types. Overconvergent PN types preferentially co-arborize and connect with dendrites of αβ and α′β′ KC subtypes. Computational simulation of the observed network showed degraded discrimination performance compared with a random network, except when all signal flowed through the overconvergent, primarily food-responsive PN types. Additional theory and experiment will be needed to fully characterize the impact of the observed non-random network structure on associative memory formation and recall.
View details
SyConn2: dense synaptic connectivity inference for volume electron microscopy
Philipp J. Schubert
Sven Dorkenwald
Jonathan Klimesch
Fabian Svara
Andrei Mancu
Hashir Ahmad
Michale S. Fee
Joergen Kornfeld
Nature Methods, vol. 19 (2022), 1367–1370
Preview abstract
The ability to acquire ever larger datasets of brain tissue using volume electron microscopy leads to an increasing demand for the automated extraction of connectomic information. We introduce SyConn2, an open-source connectome analysis toolkit, which works with both on-site high-performance compute environments and rentable cloud computing clusters. SyConn2 was tested on connectomic datasets with more than 10 million synapses, provides a web-based visualization interface and makes these data amenable to complex anatomical and neuronal connectivity queries.
View details
A connectomic study of a petascale fragment of human cerebral cortex
Alex Shapson-Coe
Daniel R. Berger
Yuelong Wu
Richard L. Schalek
Shuohong Wang
Neha Karlupia
Sven Dorkenwald
Evelina Sjostedt
Dongil Lee
Luke Bailey
Angerica Fitzmaurice
Rohin Kar
Benjamin Field
Hank Wu
Julian Wagner-Carena
David Aley
Joanna Lau
Zudi Lin
Donglai Wei
Hanspeter Pfister
Adi Peleg
Jeff W. Lichtman
bioRxiv (2021)
Preview abstract
We acquired a rapidly preserved human surgical sample from the temporal lobe of the cerebral cortex. We stained a 1 mm3 volume with heavy metals, embedded it in resin, cut more than 5000 slices at ∼30 nm and imaged these sections using a high-speed multibeam scanning electron microscope. We used computational methods to render the three-dimensional structure containing 57,216 cells, hundreds of millions of neurites and 133.7 million synaptic connections. The 1.4 petabyte electron microscopy volume, the segmented cells, cell parts, blood vessels, myelin, inhibitory and excitatory synapses, and 104 manually proofread cells are available to peruse online. Many interesting and unusual features were evident in this dataset. Glia outnumbered neurons 2:1 and oligodendrocytes were the most common cell type in the volume. Excitatory spiny neurons comprised 69% of the neuronal population, and excitatory synapses also were in the majority (76%). The synaptic drive onto spiny neurons was biased more strongly toward excitation (70%) than was the case for inhibitory interneurons (48%). Despite incompleteness of the automated segmentation caused by split and merge errors, we could automatically generate (and then validate) connections between most of the excitatory and inhibitory neuron types both within and between layers. In studying these neurons we found that deep layer excitatory cell types can be classified into new subsets, based on structural and connectivity differences, and that chandelier interneurons not only innervate excitatory neuron initial segments as previously described, but also each other’s initial segments. Furthermore, among the thousands of weak connections established on each neuron, there exist rarer highly powerful axonal inputs that establish multi-synaptic contacts (up to ∼20 synapses) with target neurons. Our analysis indicates that these strong inputs are specific, and allow small numbers of axons to have an outsized role in the activity of some of their postsynaptic partners.
View details
Denoising-based Image Compression for Connectomics
Alex Shapson-Coe
Richard L. Schalek
Jeff W. Lichtman
bioRxiv (2021)
Preview abstract
Connectomic reconstruction of neural circuits relies on nanometer resolution microscopy which produces on the order of a petabyte of imagery for each cubic millimeter of brain tissue. The cost of storing such data is a significant barrier to broadening the use of connectomic approaches and scaling to even larger volumes. We present an image compression approach that uses machine learning-based denoising and standard image codecs to compress raw electron microscopy imagery of neuropil up to 17-fold with negligible loss of reconstruction accuracy.
View details
Preview abstract
Recent advances in 3d electron microscopy are yielding ever larger reconstructions of brain tissue, encompassing thousands of individual neurons interconnected by millions of synapses. Interpreting reconstructions at this scale demands advances in the automated analysis of neuronal morphologies, for example by identifying morphological and functional subcompartments within neurons. We present a method that for the first time uses full 3d input (voxels) to automatically classify reconstructed neuron fragments as axon, dendrite, or somal subcompartments. Based on 3d convolutional neural networks, this method achieves a mean f1-score of 0.972, exceeding the previous state of the art of 0.955. The resulting predictions can support multiple analysis and proofreading applications. In particular, we leverage finely localized subcompartment predictions for automated detection and correction of merge errors in the volume reconstruction, successfully detecting 90.6% of inter-class merge errors with a false positive rate of only 2.7%.
View details
A connectome and analysis of the adult Drosophila central brain
Louis K Scheffer
C Shan Xu
Zhiyuan Lu
Shin-ya Takemura
Kenneth J Hayworth
Gary B Huang
Kazunori Shinomiya
Stuart Berg
Jody Clements
Philip M Hubbard
William T Katz
Lowell Umayam
Ting Zhao
David Ackerman
John Bogovic
Tom Dolafi
Dagmar Kainmueller
Takashi Kawase
Khaled A Khairy
Larry Lindsey
Nicole Neubarth
Donald J Olbris
Hideo Otsuna
Eric T Trautman
Masayoshi Ito
Alexander S Bates
Jens Goldammer
Tanya Wolff
Robert Svirskas
Philipp Schlegel
Erika Neace
Christopher J Knecht
Chelsea X Alvarado
Dennis A Bailey
Samantha Ballinger
Jolanta A Borycz
Brandon S Canino
Natasha Cheatham
Michael Cook
Marisa Dreher
Octave Duclos
Bryon Eubanks
Kelli Fairbanks
Samantha Finley
Nora Forknall
Audrey Francis
Gary Patrick Hopkins
Emily M Joyce
SungJin Kim
Nicole A Kirk
Julie Kovalyak
Shirley Lauchie
Alanna Lohff
Charli Maldonado
Emily A Manley
Sari McLin
Caroline Mooney
Miatta Ndama
Omotara Ogundeyi
Nneoma Okeoma
Christopher Ordish
Nicholas Padilla
Christopher M Patrick
Tyler Paterson
Elliott E Phillips
Emily M Phillips
Neha Rampally
Caitlin Ribeiro
Madelaine K Robertson
Jon Thomson Rymer
Sean M Ryan
Megan Sammons
Anne K Scott
Ashley L Scott
Aya Shinomiya
Claire Smith
Kelsey Smith
Natalie L Smith
Margaret A Sobeski
Alia Suleiman
Jackie Swift
Satoko Takemura
Iris Talebi
Dorota Tarnogorska
Emily Tenshaw
Temour Tokhi
John J Walsh
Tansy Yang
Jane Anne Horne
Feng Li
Ruchi Parekh
Patricia K Rivlin
Vivek Jayaraman
Marta Costa
Gregory SXE Jefferis
Kei Ito
Stephan Saalfeld
Reed George
Ian Meinertzhagen
Gerald M Rubin
Harald F Hess
Stephen M Plaza
eLife, vol. 9 (2020)
Preview abstract
The neural circuits responsible for animal behavior remain largely unknown. We summarize new methods and present the circuitry of a large fraction of the brain of the fruit fly Drosophila melanogaster. Improved methods include new procedures to prepare, image, align, segment, find synapses in, and proofread such large data sets. We define cell types, refine computational compartments, and provide an exhaustive atlas of cell examples and types, many of them novel. We provide detailed circuits consisting of neurons and their chemical synapses for most of the central brain. We make the data public and simplify access, reducing the effort needed to answer circuit questions, and provide procedures linking the neurons defined by our analysis with genetic reagents. Biologically, we examine distributions of connection strengths, neural motifs on different scales, electrical consequences of compartmentalization, and evidence that maximizing packing density is an important criterion in the evolution of the fly's brain.
View details
The Mind of a Mouse
Larry F. Abbott
Davi D. Bock
Edward M. Callaway
Winfried Denk
Catherine Dulac,
Adrienne L. Fairhall
Ila Fiete
Kristen M. Harris
Moritz Helmstaedter
Narayanan Kasthuri
Yann LeCun
Jeff W. Lichtman
Peter B. Littlewood
Liqun Luo
John H.R. Maunsell
R. Clay Reid
Bruce R. Rosen
Gerald M. Rubin
Terrence J. Sejnowski
H. Sebastian Seung
Karel Svoboda
David W. Tank
Doris Tsao
David C. Van Essen
Cell, vol. 182 (2020)
Preview abstract
Large scientific projects in genomics and astronomy are influential not because they answer any single ques- tion but because they enable investigation of continuously arising new questions from the same data-rich sources. Advances in automated mapping of the brain’s synaptic connections (connectomics) suggest that the complicated circuits underlying brain function are ripe for analysis. We discuss benefits of mapping a mouse brain at the level of synapses.
View details
An anatomical substrate of credit assignment in reinforcement learning
Jorgen Kornfeld
Michale S. Fee
Philipp Schubert
Winfried Denk
bioRxiv (2020)
Preview abstract
How is experience used to improve performance? In both biological and artificial systems,
the optimization of parameters that affect behavior requires a process that determines
whether a parameter affects the outcome and then modifies the parameter accordingly.
Central to the recent bloom of artificial intelligence has been the error-backpropagation
algorithm(Rumelhart, Hinton, and Williams 1986) , which computationally retraces the signal
from the output to each synapse (weight) and allows a large number of parameters to be
optimized in parallel at high learning rates. Biological systems, however, lack an obvious
mechanism to retrace the signal path. Here we show, by combining high-throughput volume
electron microscopy (Denk and Horstmann 2004) and automated connectomic
analysis(Januszewski et al. 2018; Dorkenwald et al. 2017; Schubert et al. 2019) , that the
synaptic architecture of songbird basal ganglia supports a form of local credit assessment
proposed in a model of songbird reinforcement learning (M. S. Fee and Goldberg 2011). We
show that three of this model’s major predictions hold true: first, cortical axons that encode
exploratory motor variability terminate predominantly on dendritic shafts of spiny neurons.
Second, cortical axons that encode timing seek out spines, which enable calcium-based
coincidence detection (R. Yuste and Denk 1995) and appear to be capable of creating and
storing eligibility traces (Yagishita et al. 2014). Third, synapse pairs that presynaptically share
a cortical timing axon and post-synaptically a medium spiny dendrite are substantially more
similar in size than expected, indicating a history of Hebbian plasticity (Bartol et al. 2015;
Kasthuri et al. 2015) . Combined with numerical simulations these data provide strong
evidence for a model of basal ganglia learning with a biologically plausible credit assignment
mechanism.
View details
Automated Reconstruction of a Serial-Section EM Drosophila Brain with Flood-Filling Networks and Local Realignment
Larry Lindsey
Zhihao Zheng
Alexander Shakeel Bates
István Taisz
Matthew Nichols
Feng Li
Eric Perlman
Gregory S.X.E. Jefferis
Davi Bock
bioRxiv (2019)
Preview abstract
Reconstruction of neural circuitry at single-synapse resolution is an attractive target for improving understanding of the nervous system in health and disease. Serial section transmission electron microscopy (ssTEM) is among the most prolific imaging methods employed in pursuit of such reconstructions. We demonstrate how Flood-Filling Networks (FFNs) can be used to computationally segment a forty-teravoxel whole-brain Drosophila ssTEM volume. To compensate for data irregularities and imperfect global alignment, FFNs were combined with procedures that locally re-align serial sections as well as dynamically adjust and synthesize image content. The proposed approach produced a largely merger-free segmentation of the entire ssTEM Drosophila brain, which we make freely available. As compared to manual tracing using an efficient skeletonization strategy, the segmentation enabled circuit reconstruction and analysis workflows that were an order of magnitude faster.
View details
Preview abstract
Algorithmic reconstruction of neurons from volume electron microscopy data traditionally requires training machine learning models on dataset-specific ground truth annotations that are expensive and tedious to acquire. We enhanced the training procedure of an unsupervised image-to-image translation method with additional components derived from an automated neuron segmentation approach. We show that this method, Segmentation-Enhanced CycleGAN, enables near perfect reconstruction accuracy on a benchmark connectomics segmentation dataset despite operating in a “zero-shot” setting in which only volumetric labels from a different volume imaging method were used.
View details
High-Precision Automated Reconstruction of Neurons with Flood-Filling Networks
Jörgen Kornfeld
Larry Lindsey
Winfried Denk
Nature Methods (2018)
Preview abstract
Reconstruction of neural circuits from volume electron microscopy data requires the tracing of cells in their entirety, including all their neurites. Automated approaches have been developed for tracing, but their error rates are too high to generate reliable circuit diagrams without extensive human proofreading. We present flood-filling networks, a method for automated segmentation that, similar to most previous efforts, uses convolutional neural networks, but contains in addition a recurrent pathway that allows the iterative optimization and extension of individual neuronal processes. We used flood-filling networks to trace neurons in a dataset obtained by serial block-face electron microscopy of a zebra finch brain. Using our method, we achieved a mean error-free neurite path length of 1.1 mm, and we observed only four mergers in a test set with a path length of 97 mm. The performance of flood-filling networks was an order of magnitude better than that of previous approaches applied to this dataset, although with substantially increased computational costs.
View details
Solving large Multicut problems for connectomics via domain decomposition
Constantin Pape
Thorsten Beier
Davi Bock
Anna Kreshuk
ICCV Bioimage Computing Workshop (2017)
Preview abstract
In this contribution we demonstrate how a Multicut-based segmentation pipeline can be scaled up to datasets of hundreds of Gigabytes in size. Such datasets are prevalent in connectomics, where neuron segmentation needs to be performed across very large electron microscopy image volumes. We show the advantages of a hierarchical block-wise scheme over local stitching strategies and evaluate the performance of different Multicut solvers for the segmentation of the blocks in the hierarchy. We validate the accuracy of our algorithm on a small fully annotated dataset (5×5×5 μm) and demonstrate no significant loss in segmentation quality compared to solving the Multicut problem globally. We evaluate the scalability of the algorithm on a 95×60×60 μm image volume and show that solving the Multicut problem is no longer the bottleneck of the segmentation pipeline.
View details
Preview abstract
For the past decade, convolutional networks have been used for 3D reconstruction
of neurons from electron microscopic (EM) brain images. Recent years have seen
great improvements in accuracy, as evidenced by submissions to the SNEMI3D
benchmark challenge. Here we report the first submission to surpass the estimate
of human accuracy provided by the SNEMI3D leaderboard. A variant of 3D UNet
is trained on a primary task of predicting affinities between nearest neighbor
voxels, and an auxiliary task of predicting long-range affinities. The training data is
augmented by simulated image defects. The nearest neighbor affinities are used to
create an oversegmentation, and then supervoxels are greedily agglomerated based
on mean affinity. The resulting SNEMI3D score exceeds the estimate of human
accuracy by a large margin. While one should be cautious about extrapolating
from the SNEMI3D benchmark to real-world accuracy of large-scale neural circuit
reconstruction, our submission inspires optimism that the goal of full automation
may be realizable in the future.
View details
Adversarial Image Alignment and Interpolation
arXiv, vol. abs/1707.00067 (2017)
Preview abstract
Volumetric (3d) images are acquired for many scientific and biomedical purposes using imaging methods such as serial section microscopy, CT scans, and MRI. A frequent step in the analysis and reconstruction of such data is the alignment and registration of images that were acquired in succession along a spatial or temporal dimension. For example, in serial section electron microscopy, individual 2d sections are imaged via electron microscopy and then must be aligned to one another in order to produce a coherent 3d volume. State of the art approaches find image correspondences derived from patch matching and invariant feature detectors, and then solve optimization problems that rigidly or elastically deform series of images into an aligned volume. Here we show how fully convolutional neural networks trained with an adversarial loss function can be used for two tasks: (1) synthesis of missing or damaged image data from adjacent sections, and (2) fine-scale alignment of block-face electron microscopy data. Finally, we show how these two capabilities can be combined in order to produce artificial isotropic volumes from anisotropic image volumes using a super-resolution adversarial alignment and interpolation approach.
View details
Combinatorial Energy Learning for Image Segmentation
Pieter Abbeel
Advances in Neural Information Processing Systems 29 (2016)
Preview abstract
We introduce a new machine learning approach for image segmentation that uses a neural network to model the conditional energy of a segmentation given an image. Our approach, combinatorial energy learning for image segmentation (CELIS) places a particular emphasis on modeling the inherent combinatorial nature of dense image segmentation problems. We propose efficient algorithms for learning deep neural networks to model the energy function, and for local optimization of this energy in the space of supervoxel agglomerations. We extensively evaluate our method on a publicly available 3-D microscopy dataset with 25 billion voxels of ground truth data. On an 11 billion voxel test set, we find that our method improves volumetric reconstruction accuracy by more than 20% as compared to two state-of-the-art baseline methods: graph-based segmentation of the output of a 3-D convolutional neural network trained to predict boundaries, as well as a random forest classifier trained to agglomerate supervoxels that were generated by a 3-D convolutional neural network.
View details
Flood-Filling Networks
Jörgen Kornfeld
Winfried Denk
CoRR, vol. abs/1611.00421 (2016)
Preview abstract
State-of-the-art image segmentation algorithms generally consist of at least two successive and distinct computations: a boundary detection process that uses local image information to classify image locations as boundaries between objects, followed by a pixel grouping step such as watershed or connected components that clusters pixels into segments. Prior work has varied the complexity and approach employed in these two steps, including the incorporation of multi-layer neural networks to perform boundary prediction, and the use of global optimizations during pixel clustering. We propose a unified and end-to-end trainable machine learning approach, flood-filling networks, in which a recurrent 3d convolutional network directly produces individual segments from a raw image. The proposed approach robustly segments images with an unknown and variable number of objects as well as highly variable object sizes. We demonstrate the approach on a challenging 3d image segmentation task, connectomic reconstruction from volume electron microscopy data, on which flood-filling neural networks substantially improve accuracy over other state-of-the-art methods. The proposed approach can replace complex multi-step segmentation pipelines with a single neural network that is learned end-to-end.
View details
No Results Found
