1
|
Ajioka T, Nakai N, Yamashita O, Takumi T. End-to-end deep learning approach to mouse behavior classification from cortex-wide calcium imaging. PLoS Comput Biol 2024; 20:e1011074. [PMID: 38478563 PMCID: PMC10986998 DOI: 10.1371/journal.pcbi.1011074] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/04/2023] [Revised: 04/02/2024] [Accepted: 03/04/2024] [Indexed: 04/04/2024] Open
Abstract
Deep learning is a powerful tool for neural decoding, broadly applied to systems neuroscience and clinical studies. Interpretable and transparent models that can explain neural decoding for intended behaviors are crucial to identifying essential features of deep learning decoders in brain activity. In this study, we examine the performance of deep learning to classify mouse behavioral states from mesoscopic cortex-wide calcium imaging data. Our convolutional neural network (CNN)-based end-to-end decoder combined with recurrent neural network (RNN) classifies the behavioral states with high accuracy and robustness to individual differences on temporal scales of sub-seconds. Using the CNN-RNN decoder, we identify that the forelimb and hindlimb areas in the somatosensory cortex significantly contribute to behavioral classification. Our findings imply that the end-to-end approach has the potential to be an interpretable deep learning method with unbiased visualization of critical brain regions.
Collapse
Affiliation(s)
- Takehiro Ajioka
- Department of Physiology and Cell Biology, Kobe University School of Medicine, Chuo, Kobe, Japan
| | - Nobuhiro Nakai
- Department of Physiology and Cell Biology, Kobe University School of Medicine, Chuo, Kobe, Japan
| | - Okito Yamashita
- Department of Computational Brain Imaging, ATR Neural Information Analysis Laboratories, Seika, Kyoto, Japan
| | - Toru Takumi
- Department of Physiology and Cell Biology, Kobe University School of Medicine, Chuo, Kobe, Japan
- RIKEN Center for Biosystems Dynamics Research, Chuo, Kobe, Japan
| |
Collapse
|
2
|
Mo Q, Zhang T, Wu J, Wang L, Luo J. Identification of thrombopoiesis inducer based on a hybrid deep neural network model. Thromb Res 2023; 226:36-50. [PMID: 37119555 DOI: 10.1016/j.thromres.2023.04.011] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/08/2022] [Revised: 03/13/2023] [Accepted: 04/11/2023] [Indexed: 05/01/2023]
Abstract
Thrombocytopenia is a common haematological problem worldwide. Currently, there are no relatively safe and effective agents for the treatment of thrombocytopenia. To address this challenge, we propose a computational method that enables the discovery of novel drug candidates with haematopoietic activities. Based on different types of molecular representations, three deep learning (DL) algorithms, namely recurrent neural networks (RNNs), deep neural networks (DNNs), and hybrid neural networks (RNNs+DNNs), were used to develop classification models to distinguish between active and inactive compounds. The evaluation results illustrated that the hybrid DL model exhibited the best prediction performance, with an accuracy of 97.8 % and Matthews correlation coefficient of 0.958 on the test dataset. Subsequently, we performed drug discovery screening based on the hybrid DL model and identified a compound from the FDA-approved drug library that was structurally divergent from conventional drugs and showed a potential therapeutic action against thrombocytopenia. The novel drug candidate wedelolactone significantly promoted megakaryocyte differentiation in vitro and increased platelet levels and megakaryocyte differentiation in irradiated mice with no systemic toxicity. Overall, our work demonstrates how artificial intelligence can be used to discover novel drugs against thrombocytopenia.
Collapse
Affiliation(s)
- Qi Mo
- Department of Pharmacology, School of Pharmacy, Southwest Medical University, Luzhou 646000, China
| | - Ting Zhang
- Department of Pharmacology, School of Pharmacy, Southwest Medical University, Luzhou 646000, China
| | - Jianming Wu
- Basic Medical College, Southwest Medical University, Luzhou 646000, China.
| | - Long Wang
- Department of Pharmacology, School of Pharmacy, Southwest Medical University, Luzhou 646000, China.
| | - Jiesi Luo
- Basic Medical College, Southwest Medical University, Luzhou 646000, China; State Key Laboratory of Southwestern Chinese Medicine Resources, Chengdu University of Traditional Chinese Medicine, Chengdu 610075, China.
| |
Collapse
|
3
|
Chen Z, Blair GJ, Cao C, Zhou J, Aharoni D, Golshani P, Blair HT, Cong J. FPGA-Based In-Vivo Calcium Image Decoding for Closed-Loop Feedback Applications. IEEE TRANSACTIONS ON BIOMEDICAL CIRCUITS AND SYSTEMS 2023; 17:169-179. [PMID: 37071510 PMCID: PMC10414190 DOI: 10.1109/tbcas.2023.3268130] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/11/2023]
Abstract
Miniaturized calcium imaging is an emerging neural recording technique that has been widely used for monitoring neural activity on a large scale at a specific brain region of rats or mice. Most existing calcium-image analysis pipelines operate offline. This results in long processing latency, making it difficult to realize closed-loop feedback stimulation for brain research. In recent work, we have proposed an FPGA-based real-time calcium image processing pipeline for closed-loop feedback applications. It can perform real-time calcium image motion correction, enhancement, fast trace extraction, and real-time decoding from extracted traces. Here, we extend this work by proposing a variety of neural network based methods for real-time decoding and evaluate the tradeoff among these decoding methods and accelerator designs. We introduce the implementation of the neural network based decoders on the FPGA, and show their speedup against the implementation on the ARM processor. Our FPGA implementation enables the real-time calcium image decoding with sub-ms processing latency for closed-loop feedback applications.
Collapse
|
4
|
Supekar OD, Sias A, Hansen SR, Martinez G, Peet GC, Peng X, Bright VM, Hughes EG, Restrepo D, Shepherd DP, Welle CG, Gopinath JT, Gibson EA. Miniature structured illumination microscope for in vivo 3D imaging of brain structures with optical sectioning. BIOMEDICAL OPTICS EXPRESS 2022; 13:2530-2541. [PMID: 35519247 PMCID: PMC9045919 DOI: 10.1364/boe.449533] [Citation(s) in RCA: 13] [Impact Index Per Article: 6.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/24/2021] [Revised: 02/26/2022] [Accepted: 03/05/2022] [Indexed: 05/25/2023]
Abstract
We present a high-resolution miniature, light-weight fluorescence microscope with electrowetting lens and onboard CMOS for high resolution volumetric imaging and structured illumination for rejection of out-of-focus and scattered light. The miniature microscope (SIMscope3D) delivers structured light using a coherent fiber bundle to obtain optical sectioning with an axial resolution of 18 µm. Volumetric imaging of eGFP labeled cells in fixed mouse brain tissue at depths up to 260 µm is demonstrated. The functionality of SIMscope3D to provide background free 3D imaging is shown by recording time series of microglia dynamics in awake mice at depths up to 120 µm in the brain.
Collapse
Affiliation(s)
- Omkar D Supekar
- Department of Electrical, Energy and Computer Engineering, University of Colorado Boulder, CO 80309, USA
- Department of Mechanical Engineering, University of Colorado Boulder, CO 80309, USA
| | - Andrew Sias
- Department of Bioengineering, University of Colorado Anschutz Medical Campus, CO 80045, USA
| | - Sean R Hansen
- Department of Bioengineering, University of Colorado Anschutz Medical Campus, CO 80045, USA
| | - Gabriel Martinez
- Department of Bioengineering, University of Colorado Anschutz Medical Campus, CO 80045, USA
| | - Graham C Peet
- Department of Cell and Developmental Biology, University of Colorado Anschutz Medical Campus, CO 80045, USA
| | - Xiaoyu Peng
- Department of Physiology and Biophysics, University of Colorado Anschutz Medical Campus, CO 80045, USA
| | - Victor M Bright
- Department of Mechanical Engineering, University of Colorado Boulder, CO 80309, USA
| | - Ethan G Hughes
- Department of Cell and Developmental Biology, University of Colorado Anschutz Medical Campus, CO 80045, USA
| | - Diego Restrepo
- Department of Cell and Developmental Biology, University of Colorado Anschutz Medical Campus, CO 80045, USA
| | | | - Cristin G Welle
- Department of Physiology and Biophysics, University of Colorado Anschutz Medical Campus, CO 80045, USA
- Department of Neurosurgery, University of Colorado Anschutz Medical Campus, CO 80045, USA
| | - Juliet T Gopinath
- Department of Electrical, Energy and Computer Engineering, University of Colorado Boulder, CO 80309, USA
- Department of Physics, University of Colorado Boulder, CO 80309, USA
| | - Emily A Gibson
- Department of Bioengineering, University of Colorado Anschutz Medical Campus, CO 80045, USA
| |
Collapse
|
5
|
Livezey JA, Glaser JI. Deep learning approaches for neural decoding across architectures and recording modalities. Brief Bioinform 2020; 22:1577-1591. [PMID: 33372958 DOI: 10.1093/bib/bbaa355] [Citation(s) in RCA: 9] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/18/2020] [Revised: 10/31/2020] [Accepted: 11/04/2020] [Indexed: 12/19/2022] Open
Abstract
Decoding behavior, perception or cognitive state directly from neural signals is critical for brain-computer interface research and an important tool for systems neuroscience. In the last decade, deep learning has become the state-of-the-art method in many machine learning tasks ranging from speech recognition to image segmentation. The success of deep networks in other domains has led to a new wave of applications in neuroscience. In this article, we review deep learning approaches to neural decoding. We describe the architectures used for extracting useful features from neural recording modalities ranging from spikes to functional magnetic resonance imaging. Furthermore, we explore how deep learning has been leveraged to predict common outputs including movement, speech and vision, with a focus on how pretrained deep networks can be incorporated as priors for complex decoding targets like acoustic speech or images. Deep learning has been shown to be a useful tool for improving the accuracy and flexibility of neural decoding across a wide range of tasks, and we point out areas for future scientific development.
Collapse
Affiliation(s)
- Jesse A Livezey
- Neural Systems and Data Science Laboratory at the Lawrence Berkeley National Laboratory. He obtained his PhD in Physics from the University of California, Berkeley
| | - Joshua I Glaser
- Center for Theoretical Neuroscience and Department of Statistics at Columbia University. He obtained his PhD in Neuroscience from Northwestern University
| |
Collapse
|
6
|
Leopold AV, Shcherbakova DM, Verkhusha VV. Fluorescent Biosensors for Neurotransmission and Neuromodulation: Engineering and Applications. Front Cell Neurosci 2019; 13:474. [PMID: 31708747 PMCID: PMC6819510 DOI: 10.3389/fncel.2019.00474] [Citation(s) in RCA: 64] [Impact Index Per Article: 12.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/15/2019] [Accepted: 10/08/2019] [Indexed: 12/21/2022] Open
Abstract
Understanding how neuronal activity patterns in the brain correlate with complex behavior is one of the primary goals of modern neuroscience. Chemical transmission is the major way of communication between neurons, however, traditional methods of detection of neurotransmitter and neuromodulator transients in mammalian brain lack spatiotemporal precision. Modern fluorescent biosensors for neurotransmitters and neuromodulators allow monitoring chemical transmission in vivo with millisecond precision and single cell resolution. Changes in the fluorescent biosensor brightness occur upon neurotransmitter binding and can be detected using fiber photometry, stationary microscopy and miniaturized head-mounted microscopes. Biosensors can be expressed in the animal brain using adeno-associated viral vectors, and their cell-specific expression can be achieved with Cre-recombinase expressing animals. Although initially fluorescent biosensors for chemical transmission were represented by glutamate biosensors, nowadays biosensors for GABA, acetylcholine, glycine, norepinephrine, and dopamine are available as well. In this review, we overview functioning principles of existing intensiometric and ratiometric biosensors and provide brief insight into the variety of neurotransmitter-binding proteins from bacteria, plants, and eukaryotes including G-protein coupled receptors, which may serve as neurotransmitter-binding scaffolds. We next describe a workflow for development of neurotransmitter and neuromodulator biosensors. We then discuss advanced setups for functional imaging of neurotransmitter transients in the brain of awake freely moving animals. We conclude by providing application examples of biosensors for the studies of complex behavior with the single-neuron precision.
Collapse
Affiliation(s)
- Anna V Leopold
- Medicum, Faculty of Medicine, University of Helsinki, Helsinki, Finland
| | - Daria M Shcherbakova
- Department of Anatomy and Structural Biology, Gruss-Lipper Biophotonics Center, Albert Einstein College of Medicine, Bronx, NY, United States
| | - Vladislav V Verkhusha
- Medicum, Faculty of Medicine, University of Helsinki, Helsinki, Finland.,Department of Anatomy and Structural Biology, Gruss-Lipper Biophotonics Center, Albert Einstein College of Medicine, Bronx, NY, United States
| |
Collapse
|