CN117953005A - Simulation-based rapid wide-field calcium imaging neuron extraction method and system - Google Patents
Simulation-based rapid wide-field calcium imaging neuron extraction method and system Download PDFInfo
- Publication number
- CN117953005A CN117953005A CN202410152547.7A CN202410152547A CN117953005A CN 117953005 A CN117953005 A CN 117953005A CN 202410152547 A CN202410152547 A CN 202410152547A CN 117953005 A CN117953005 A CN 117953005A
- Authority
- CN
- China
- Prior art keywords
- background
- neuron
- wide
- field
- field image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 210000002569 neuron Anatomy 0.000 title claims abstract description 131
- OYPRJOBELJOOCE-UHFFFAOYSA-N Calcium Chemical compound [Ca] OYPRJOBELJOOCE-UHFFFAOYSA-N 0.000 title claims abstract description 40
- 229910052791 calcium Inorganic materials 0.000 title claims abstract description 40
- 239000011575 calcium Substances 0.000 title claims abstract description 40
- 238000004088 simulation Methods 0.000 title claims abstract description 40
- 238000000605 extraction Methods 0.000 title claims abstract description 38
- 238000003384 imaging method Methods 0.000 title claims abstract description 38
- 230000011218 segmentation Effects 0.000 claims abstract description 43
- 238000000034 method Methods 0.000 claims description 31
- 230000003287 optical effect Effects 0.000 claims description 25
- 230000006870 function Effects 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 2
- 230000002123 temporal effect Effects 0.000 abstract description 5
- 210000001153 interneuron Anatomy 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 5
- 230000001537 neural effect Effects 0.000 description 5
- 238000011109 contamination Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 210000005013 brain tissue Anatomy 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000000149 argon plasma sintering Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003925 brain function Effects 0.000 description 1
- 238000004587 chromatography analysis Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000001054 cortical effect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 210000005171 mammalian brain Anatomy 0.000 description 1
- 238000000386 microscopy Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000340 multi-photon laser scanning microscopy Methods 0.000 description 1
- 238000000399 optical microscopy Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/09—Supervised learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10056—Microscopic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a rapid wide-field calcium imaging neuron extraction method and system based on simulation, which belong to the technical field of wide-field calcium imaging. The background network is trained by taking the background image as input and the non-background image as supervision. After the background removing network converges, clean neuron signals are conveniently extracted from the scattering background. The neuron segmentation network receives the output from the background network and performs high-speed neuron segmentation. Neurons are further partitioned into independent regions according to their temporal connectivity. Since there is no inter-neuron crosstalk, the time trace of each neuron can be directly obtained. The ultra-fast and high-precision neuron extraction function is realized.
Description
Technical Field
The invention relates to the technical field of wide-field calcium imaging, in particular to a simulation-based rapid wide-field calcium imaging neuron extraction method and system.
Background
Calcium imaging helps to study brain function in a variety of behavioral tasks by means of genetically encoded calcium indicators and optical microscopy instruments.
On the one hand, serial acquisition methods such as a two-photon laser scanning microscope and the like have stronger optical chromatography capability and robustness to scattering, but have lower time resolution on a millimeter-level view field. While some methods may increase the frame rate of a two-photon laser scanning microscope, high power illumination may result in thermally induced damage. With respect to spatial dimensions, two-photon laser scanning microscopy has achieved a field of view of about 5mm in diameter, but this typically requires time subsampling of calcium dynamics.
On the other hand, the parallel scheme of a wide-field microscope and the like is combined with the increasing range of array sensors, and a practical tool is provided for neuroscientists. With optimized optical setup and computational tools, a wide field microscope is able to record large populations of neurons of tens of mammalian brain regions in a pixel size of 0.8 μm in a10 x 8mm 2 field of view, while recording millions of neurons.
However, cross-talk and background contamination due to scattering present challenges for wide field microscopy. Neurons that are far from the focal plane will produce a blurred background signal as the wide field microscope illuminates and detects the entire volume of the sample. Light scattering in opaque tissue further worsens the fluorescence signal from the focal plane and distorts the information about neuronal location and activity. To reduce these effects, researchers often have to sacrifice imaging speed and even sample health.
Some conventional algorithms may separate the wide-field neuron signal from background contamination. Constrained non-negative matrix factorization (CNMF-E) methods model strong background signals with a priori knowledge. However, optimizing the background model requires complex parameter adjustments and is computationally intensive, preventing its application to cortical scale neuronal processing. On-line processing using a lightweight version of the algorithm may partially alleviate the speed problem, but may sacrifice performance. Other methods without explicit modeling of the fluctuating background can achieve higher processing speeds, but are often at risk of residual background contamination. Thus, existing calcium record analysis algorithms are far from optimal in terms of high speed and excellent performance.
The neural network breaks through in the image processing tasks of neurons such as image enhancement, neuron segmentation, spike inference and the like. By proper training, deep learning based two-photon imaging neuron activity inference can achieve faster speeds without affecting performance. However, due to the lack of paired wide-field and background-free data for training, there is little interest in the application of deep learning to background removal in wide-field neuron recordings. The method of converting the background model into a trainable convolutional layer alleviates the need for paired data, but requires retraining each sample and performance compromises compared to other neuron extraction methods.
Disclosure of Invention
In view of the shortcomings in the prior art, the invention provides a simulation-based rapid wide-field calcium imaging neuron extraction method and system for solving at least some of the technical problems, which are convenient for realizing ultra-rapid and high-precision wide-field calcium imaging neuron extraction.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
In a first aspect, an embodiment of the present invention provides a method for extracting a fast wide-field calcium imaging neuron based on simulation, the method comprising: training phase and application phase, wherein:
The training phase comprises:
given optical system parameters, generating a background-free wide-field image, a background-carrying wide-field image and a neuron footprint by using a digital twin neuroanatomy and optical microscopic simulation method;
taking the background-carrying wide field image as an input of a background removing network, taking the background-free wide field image as a supervision of the background removing network, and training the background removing network; after convergence of the background-removing network training, outputting a purified background-free wide-field image;
Taking the purified background-free wide-field image as an input of a neuron segmentation network, taking the neuron footprint as a supervision of the neuron segmentation network, and training the neuron segmentation network;
the application stage comprises the following steps:
inputting the real shot background-carrying wide-field image into a trained background removing network to obtain a purified background-free wide-field image; inputting the purified background-free wide-field image into a trained neuron segmentation network, outputting predicted neuron footprints, and realizing neuron extraction.
Optionally, the optical system parameters include: numerical aperture and magnification.
Optionally, the background removing network includes: 14 three-dimensional convolutional layers, each followed by Relu activation functions.
Optionally, the neuron segmentation network includes: 14 two-dimensional convolutional layers, each followed by Relu activation functions.
Optionally, the neuron segmentation network extracts the spatial footprint and the time signal of calcium from the purified background-free image, performs neuron segmentation, and divides the neurons into independent areas according to the time-space connectivity of the neurons to obtain the time traces of the neurons.
Optionally, spatially overlapping neurons are decomposed by a local non-negative matrix factorization algorithm to eliminate active crosstalk.
Alternatively, a real-time taken background wide field image is obtained by a wide field fluorescence microscope.
In a second aspect, an embodiment of the present invention further provides a system for extracting a fast wide-field calcium imaging neuron based on simulation, where the method for extracting a fast wide-field calcium imaging neuron based on simulation is applied to implement fast wide-field calcium imaging neuron extraction, and the system includes:
The simulation module is used for giving optical system parameters, and generating a background-free wide-field image, a background-carrying wide-field image and a neuron footprint by using a digital twin neuroanatomy and optical microscopic simulation method;
The network training module is used for taking the background wide field image as the input of the background removing network, taking the background-free wide field image as the supervision of the background removing network, and training the background removing network; after convergence of the background-removing network training, outputting a purified background-free wide-field image; the method is also used for taking the purified background-free wide-field image as an input of a neuron segmentation network, taking the neuron footprint as a supervision of the neuron segmentation network, and training the neuron segmentation network;
The neuron extraction module is used for inputting the real shot wide field image with the background into a trained background removing network to obtain a purified wide field image without the background; inputting the purified background-free wide-field image into a trained neuron segmentation network, outputting predicted neuron footprints, and realizing neuron extraction.
Compared with the prior art, the invention has at least the following beneficial effects:
The invention provides a rapid wide-field calcium imaging neuron extraction method and system based on simulation, which can generate brain tissue images with background and without background according to specific optical system parameters by using digital twin neuroanatomy and optical microscopic simulation technology, thereby avoiding the requirements on common wide-field images and high-quality matched images without background. By taking the background-image as input and the background-free image as supervision, the de-background network (RB-Net) of the present invention can be trained. After RB-Net convergence, clean neuronal signals are conveniently extracted from the scattering background. Further, a neuron segmentation network (NS-Net) receives the output of the RB-Net and performs high-speed neuron segmentation. Neurons are further partitioned into independent regions according to their temporal connectivity. Since there is no inter-neuron crosstalk, the time trace of each neuron can be directly obtained. In the invention, RB-Net and NS-Net are combined to form a complete background removal and neuron segmentation framework, thereby realizing a rapid and high-precision neuron extraction function.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention.
Fig. 1 is a schematic diagram of a training simulation-based fast wide-field calcium imaging neuron extraction network according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a data processing flow of a simulation-based rapid wide-field calcium imaging neuron extraction method according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. Moreover, various numbers and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1 and 2, the present invention provides a simulation-based rapid wide-field calcium imaging neuron extraction method, the training phase of which includes:
given optical system parameters, generating a background-free wide-field image, a background-carrying wide-field image and a neuron footprint by using a digital twin neuroanatomy and optical microscopic simulation method;
Taking the wide field image with the background as the input of the background removing network, taking the wide field image without the background as the supervision of the background removing network, and training the background removing network; after convergence of the background-removing network training, outputting a purified background-free wide-field image;
Taking the purified background-free wide-field image as input of a neuron segmentation network, taking the neuron footprint as supervision of the neuron segmentation network, and training the neuron segmentation network;
the application phase of the method comprises the following steps:
inputting the real shot background-carrying wide-field image into a trained background removing network to obtain a purified background-free wide-field image; inputting the purified background-free wide-field image into a trained neuron segmentation network, outputting predicted neuron footprints, and realizing neuron extraction.
The following describes in detail specific embodiments of the process according to the invention:
in a specific embodiment, the invention forms a complete background removal and neuron segmentation framework based on digital twin neuro-anatomy and optical micro-simulation technology and combination of RB-Net and NS-Net, and can generate brain tissue images with background and without background according to specific optical system parameters (preferably, numerical aperture range is 0.1-1.0 and magnification is 5 times to 20 times), thereby avoiding the requirement of common wide-field images and high-quality non-background paired images. The background-free network (RB-Net) of the present invention can be trained by taking the background-free image as an input to the neural network. After RB-Net convergence, clean neuronal signals can be extracted from the scattering background. Further, the present invention uses a lightweight neuronal extraction network (NS-Net) to extract the spatial footprint and temporal signal of calcium from the clean de-background image described above. The NS-Net receives the RB-Net output and performs high-speed neuron segmentation. Neurons are further partitioned into independent regions according to their temporal connectivity. Since there is no inter-neuron crosstalk, the time trace of each neuron can be directly obtained. Spatially overlapping neurons may be further decomposed by a local non-Negative Matrix Factorization (NMF) algorithm to eliminate active crosstalk.
The training flow is shown in figure 1. The method comprises the following specific steps: (1) Given optical system parameters, a background-free wide-field image (truth value), a background-free wide-field image, and a neuron footprint (truth value) are generated using digital twin neuro-anatomy and optical micro-simulation techniques. (2) The background-free wide field image (true value) is used as the supervision of the background removing network, and the background removing network is trained. After convergence of the background-free network training, a background-free wide-field image (prediction), i.e., a purified background-free wide-field image, can be output. (3) The background-free wide-field image (prediction) is used as an input of the neuron segmentation network, and the neuron footprints (true values) are used as supervision of the neuron segmentation network to train the neuron segmentation network. After convergence of the neuron segmentation network training, the neuron footprints (predictions) may be output.
In a specific embodiment, a simulation-based rapid wide-field calcium imaging neuron extraction method is specifically applied, and the data processing flow is shown in fig. 2. The method comprises the following specific steps: (1) And (5) performing real shooting by using a wide-field fluorescence microscope to obtain a wide-field image with a background. (2) And inputting the wide-field image with the background into a trained background removing network to obtain the wide-field image with the background removed. (3) And inputting the wide-field image with the background removed into a neuron segmentation network to obtain neuron footprints.
From the above description of embodiments, those skilled in the art will appreciate that the present invention provides a fast wide-field calcium imaging neuron extraction method based on simulation, which has the following advantages: (1) The rapid neuron extraction method based on simulation is combined with wide-field imaging, so that rapid large-field calcium activity analysis is realized. The wide-field imaging has a higher temporal resolution and a simpler optical system structure than the two-photon imaging. (2) The digital twin neuro-anatomy and optical micro-simulation technology avoids the acquisition of paired data with high signal-to-back ratio and also ensures the accuracy of the trained neural network. (3) RB-Net can produce high contrast output without background contamination.
Further, the invention also provides a simulation-based rapid wide-field calcium imaging neuron extraction system, which is applied to the simulation-based rapid wide-field calcium imaging neuron extraction method in the embodiment, and realizes rapid wide-field calcium imaging neuron extraction, and the system comprises:
The simulation module is used for giving optical system parameters, and generating a background-free wide-field image, a background-carrying wide-field image and a neuron footprint by using a digital twin neuroanatomy and optical microscopic simulation method;
The network training module is used for taking the background wide field image as the input of the background removing network, taking the background-free wide field image as the supervision of the background removing network, and training the background removing network; after convergence of the background-removing network training, outputting a purified background-free wide-field image; the method is also used for taking the purified background-free wide-field image as an input of a neuron segmentation network, taking the neuron footprint as a supervision of the neuron segmentation network, and training the neuron segmentation network;
The neuron extraction module is used for inputting the real shot wide field image with the background into a trained background removing network to obtain a purified wide field image without the background; inputting the purified background-free wide-field image into a trained neuron segmentation network, outputting predicted neuron footprints, and realizing neuron extraction.
The system provided by the embodiment of the present invention has the same implementation principle and technical effects as those of the foregoing method embodiment, and for brevity description, the corresponding contents in the foregoing method embodiment may be referred to for the parts of the system embodiment that are not mentioned, and will not be described herein again.
In addition, embodiments of the present invention also provide a storage medium having stored thereon one or more programs readable by a computing device, the one or more programs including instructions, which when executed by the computing device, cause the computing device to perform a simulation-based rapid wide-field calcium imaging neuron extraction method of the above embodiments.
In an embodiment of the present invention, the storage medium may be, for example, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the storage medium include: portable computer disks, hard disks, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static Random Access Memory (SRAM), portable compact disk read-only memory (CD-ROM), digital Versatile Disks (DVD), memory sticks, floppy disks, mechanical coding devices, and any suitable combination of the foregoing.
It will be appreciated by those skilled in the art that embodiments of the invention may be provided as a method, system, or computer program product, or the like. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects.
It is to be noticed that the term 'comprising', does not exclude the presence of elements or steps other than those listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A simulation-based rapid wide-field calcium imaging neuron extraction method, which is characterized by comprising the following steps: training phase and application phase, wherein:
The training phase comprises:
given optical system parameters, generating a background-free wide-field image, a background-carrying wide-field image and a neuron footprint by using a digital twin neuroanatomy and optical microscopic simulation method;
taking the background-carrying wide field image as an input of a background removing network, taking the background-free wide field image as a supervision of the background removing network, and training the background removing network; after convergence of the background-removing network training, outputting a purified background-free wide-field image;
Taking the purified background-free wide-field image as an input of a neuron segmentation network, taking the neuron footprint as a supervision of the neuron segmentation network, and training the neuron segmentation network;
the application stage comprises the following steps:
inputting the real shot background-carrying wide-field image into a trained background removing network to obtain a purified background-free wide-field image; inputting the purified background-free wide-field image into a trained neuron segmentation network, outputting predicted neuron footprints, and realizing neuron extraction.
2. A method of fast wide-field calcium imaging neuron extraction based on simulation according to claim 1, wherein the optical system parameters comprise: numerical aperture and magnification.
3. A method of simulation-based rapid wide-field calcium imaging neuron extraction according to claim 1, wherein the de-background network comprises: 14 three-dimensional convolutional layers, each followed by Relu activation functions.
4. A method of fast wide-field calcium imaging neuron extraction based on simulation according to claim 1, wherein said neuron segmentation network comprises: 14 two-dimensional convolutional layers, each followed by Relu activation functions.
5. The method for extracting the fast wide-field calcium imaging neurons based on simulation according to claim 1, wherein the neuron segmentation network extracts a spatial footprint and a time signal of calcium from the purified background-free image, carries out neuron segmentation, and the neurons are segmented into independent areas according to the time-space connectivity of the neurons, so as to obtain the time trace of each neuron.
6. The method for extracting neurons in a fast wide-field calcium imaging system based on simulation as claimed in claim 5, wherein spatially overlapping neurons are decomposed by a local non-negative matrix factorization algorithm to eliminate active crosstalk.
7. The method for extracting the simulation-based rapid wide-field calcium imaging neuron according to claim 1, wherein the real-time photographed wide-field image with the background is obtained through a wide-field fluorescence microscope.
8. A simulation-based rapid broad-field calcium imaging neuron extraction system, wherein a simulation-based rapid broad-field calcium imaging neuron extraction method according to any one of claims 1 to 7 is applied to realize rapid broad-field calcium imaging neuron extraction, and the system comprises:
The simulation module is used for giving optical system parameters, and generating a background-free wide-field image, a background-carrying wide-field image and a neuron footprint by using a digital twin neuroanatomy and optical microscopic simulation method;
The network training module is used for taking the background wide field image as the input of the background removing network, taking the background-free wide field image as the supervision of the background removing network, and training the background removing network; after convergence of the background-removing network training, outputting a purified background-free wide-field image; the method is also used for taking the purified background-free wide-field image as an input of a neuron segmentation network, taking the neuron footprint as a supervision of the neuron segmentation network, and training the neuron segmentation network;
The neuron extraction module is used for inputting the real shot wide field image with the background into a trained background removing network to obtain a purified wide field image without the background; inputting the purified background-free wide-field image into a trained neuron segmentation network, outputting predicted neuron footprints, and realizing neuron extraction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410152547.7A CN117953005A (en) | 2024-02-03 | 2024-02-03 | Simulation-based rapid wide-field calcium imaging neuron extraction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410152547.7A CN117953005A (en) | 2024-02-03 | 2024-02-03 | Simulation-based rapid wide-field calcium imaging neuron extraction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117953005A true CN117953005A (en) | 2024-04-30 |
Family
ID=90794250
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410152547.7A Pending CN117953005A (en) | 2024-02-03 | 2024-02-03 | Simulation-based rapid wide-field calcium imaging neuron extraction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117953005A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5710830A (en) * | 1994-10-05 | 1998-01-20 | United Parcel Service | Method of and apparatus for segmenting foreground and background information for optical character recognition of labels employing single layer recurrent neural network |
| CN113383225A (en) * | 2018-12-26 | 2021-09-10 | 加利福尼亚大学董事会 | System and method for propagating two-dimensional fluorescence waves onto a surface using deep learning |
| CN113433130A (en) * | 2021-07-12 | 2021-09-24 | 清华大学 | Method and device for generating confocal imaging by wide-field imaging |
-
2024
- 2024-02-03 CN CN202410152547.7A patent/CN117953005A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5710830A (en) * | 1994-10-05 | 1998-01-20 | United Parcel Service | Method of and apparatus for segmenting foreground and background information for optical character recognition of labels employing single layer recurrent neural network |
| CN113383225A (en) * | 2018-12-26 | 2021-09-10 | 加利福尼亚大学董事会 | System and method for propagating two-dimensional fluorescence waves onto a surface using deep learning |
| CN113433130A (en) * | 2021-07-12 | 2021-09-24 | 清华大学 | Method and device for generating confocal imaging by wide-field imaging |
Non-Patent Citations (1)
| Title |
|---|
| YUANLONG ZHANG等: "Rapid detection of neurons in widefield calcium imaging datasets after training with synthetic data", 《NATURE METHODS》, 1 April 2023 (2023-04-01), pages 1 - 28 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shen et al. | End-to-end deep image reconstruction from human brain activity | |
| Li et al. | 3D neuron reconstruction in tangled neuronal image with deep networks | |
| Kloster et al. | Deep learning-based diatom taxonomy on virtual slides | |
| Jiang et al. | Cascaded subpatch networks for effective CNNs | |
| Chaudhary et al. | Fast, efficient, and accurate neuro-imaging denoising via supervised deep learning | |
| Halvardsson et al. | Interpretation of swedish sign language using convolutional neural networks and transfer learning | |
| Liu et al. | Automatic detection and segmentation of mitochondria from SEM images using deep neural network | |
| Li et al. | Robust blood cell image segmentation method based on neural ordinary differential equations | |
| CN117689928A (en) | Unmanned aerial vehicle detection method for improving yolov5 | |
| Baykal et al. | Modern convolutional object detectors for nuclei detection on pleural effusion cytology images | |
| Wiesner et al. | On generative modeling of cell shape using 3D GANs | |
| Xu et al. | AutoSegNet: An automated neural network for image segmentation | |
| Yang et al. | Research and Application of Visual Object Recognition System Based on Deep Learning and Neural Morphological Computation | |
| Irfan et al. | A novel feature extraction model to enhance underwater image classification | |
| Song et al. | Light-field microscopy for the optical imaging of neuronal activity: When model-based methods meet data-driven approaches | |
| Vizcaino et al. | Real-time light field 3D microscopy via sparsity-driven learned deconvolution | |
| Zhang et al. | CE-RetinaNet: A channel enhancement method for infrared wildlife detection in UAV images | |
| Javaid et al. | A Novel Action Transformer Network for Hybrid Multimodal Sign Language Recognition. | |
| Thirunavukkarasu et al. | Intracranial hemorrhage detection using deep convolutional neural network | |
| CN117953005A (en) | Simulation-based rapid wide-field calcium imaging neuron extraction method and system | |
| Ong et al. | Enhanced symbol recognition based on advanced data augmentation for engineering diagrams | |
| Feng et al. | Plantorganelle Hunter is an effective deep-learning-based method for plant organelle phenotyping in electron microscopy | |
| Zhang et al. | FCE-Net: a fast image contrast enhancement method based on deep learning for biomedical optical images | |
| Mayerich et al. | Deep Learning | |
| Mayerich et al. | Deep Learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |