03_Image_analysis.pot

# SOME DESCRIPTIVE TITLE.
# Copyright (C) 2024
# This file is distributed under the same license as the Python package.
# FIRST AUTHOR <EMAIL@ADDRESS>, YEAR.
#
#, fuzzy
msgid ""
msgstr ""
"Project-Id-Version: Python \n"
"Report-Msgid-Bugs-To: \n"
"POT-Creation-Date: 2026-04-11 14:17+0000\n"
"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
"Language-Team: LANGUAGE <LL@li.org>\n"
"MIME-Version: 1.0\n"
"Content-Type: text/plain; charset=UTF-8\n"
"Content-Transfer-Encoding: 8bit\n"

#: ../../03_Image_analysis/Classification.md:1
msgid "Object classification"
msgstr ""

#: ../../03_Image_analysis/Classification.md:3
msgid "What is classification?"
msgstr ""

#: ../../03_Image_analysis/Classification.md:4
msgid "Simply put, phenotypic classification is about categorizing objects into different groups based on their features (aka measurements)."
msgstr ""

#: ../../03_Image_analysis/Classification.md:0
#: ../../03_Image_analysis/Colocalization.md:0
#: ../../03_Image_analysis/Intensity.md:0
#: ../../03_Image_analysis/Shape.md:0
#: ../../03_Image_analysis/Tracking.md:0
msgid "📏 How do I measure it?"
msgstr ""

#: ../../03_Image_analysis/Classification.md:10
msgid "Phenotypic classification can be performed a few different ways. One way to break this down is by unsupervised vs. supervised classification."
msgstr ""

#: ../../03_Image_analysis/Classification.md:12
msgid "In **supervised classification**, a human also provides information on what the different groups of objects should look like by providing representative examples of each group in a training dataset. The computer then learns how to assign objects to groups based on their measurements by testing models against the ground truth training dataset."
msgstr ""

#: ../../03_Image_analysis/Classification.md:14
msgid "For example, you could classify cells based on a visual phenotype and train a machine learning classifier to derive which measurement ranges are associated with different classes. This is supervised classification because a person is providing instruction of how many classes there should be and examples of what each class should look like for the computer to learn from. An example of this could be annotating a subset of cells that are in different stages of mitosis and training a classifier to use your labels to find other cells in those stages."
msgstr ""

#: ../../03_Image_analysis/Classification.md:16
msgid "In **unsupervised classification**, you group objects based on their measurements, but without any top-down human-defined guidance into how many groups there are or what the groups should look like."
msgstr ""

#: ../../03_Image_analysis/Classification.md:18
msgid "For example, you could measure hundreds or thousands of features of cells from many treatments, as is typical in large-scale cell profiling experiments. Next you could let the computer cluster the cells into some number of different groups based on having similar measurements. This is a form of unsupervised clustering, where you observe what groups emerge from a computer considering their measurements only, and not class labels we impose as researchers. These sorts of clustering experiments can provide novel results but may also be harder to interpret; see this protocol{cite}`GarciaFossaCruz2023` for more information."
msgstr ""

#: ../../03_Image_analysis/Classification.md:0
#: ../../03_Image_analysis/Colocalization.md:0
#: ../../03_Image_analysis/Data_management.md:0
#: ../../03_Image_analysis/Intensity.md:0
#: ../../03_Image_analysis/Shape.md:0
#: ../../03_Image_analysis/Tracking.md:0
msgid "<span style=\"color: red\">⚠️</span> Where can things go wrong?"
msgstr ""

#: ../../03_Image_analysis/Classification.md:23
msgid "**Valid measurements are still important** Classification can be simple or complex, but results always depend on the validity of your measurements. For this reason, all the caveats of earlier measurement sections also apply here."
msgstr ""

#: ../../03_Image_analysis/Classification.md:24
msgid "**Machines are lazy** Machine learning classifiers aren’t necessarily going to learn the biologically relevant features that distinguish objects from distinct groups. Confounding features, or features that vary with your phenotype but are not biologically related to it, can limit the usefulness of your classifier and lead to incorrect conclusions. For instance, if clinicians often put rulers next to malignant looking moles and not next to benign moles and try to train a machine learning classifier to distinguish malignant vs. benign, the model might learn to classify images with rulers as malignant without tapping into any of the relevant features of the moles. This is a [real example](https://pubmed.ncbi.nlm.nih.gov/30244720/) {cite}`Narla2018-qh`. If possible, examining which features your model is relying on to classify objects can be a way to check for this. It’s also important to standardize how you capture images of your different classes of objects and include a large enough training set with images with lots of variation. You wouldn’t want all your positive cells to come from samples you imaged in March and all your negative cells from samples you imaged in January, for example."
msgstr ""

#: ../../03_Image_analysis/Classification.md:25
msgid "**Violating model assumptions** If using a machine learning classifier, different models come with different baked-in assumptions. If you’re starting out, it can be difficult to know which to pick. There are interactive tools such as [CellProfiler Analyst](https://academic.oup.com/bioinformatics/article/32/20/3210/2196630) {cite}`Stirling2021-ov` and [Piximi](https://www.piximi.app/) that make training a classifier easier, especially if you don’t know how to code."
msgstr ""

#: ../../03_Image_analysis/Classification.md:26
msgid "**Messy boundaries** Most methods of supervised classification, where the user assigns objects to a score or to a bin, ultimately treat each bin as a totally separate entity; biology is rarely so neat. For example, a supervised classifier for cell cycle phase must assign a cell to one phase, but in fact progression through the cell cycle is not a perfectly switch-like process, as can be visualized by measurements of individual cells (colored by their class given by a human observer). More sophisticated methods may be needed to classify more continuous phenotypes"
msgstr ""

#: ../../03_Image_analysis/Classification.md:0
msgid "a continuous distribution of cell cycle states"
msgstr ""

#: ../../03_Image_analysis/Classification.md:35
msgid "**Strict division into supervised classes can be tricky for continuous biological processes**. Adapted from Eulenberg, P., Köhler, N., Blasi, T. _et al_. Reconstructing cell cycle and disease progression using deep learning. _Nat Commun_ **8**, 463 (2017) {cite}`Eulenberg2017-ax`"
msgstr ""

#: ../../03_Image_analysis/Classification.md:0
#: ../../03_Image_analysis/Colocalization.md:0
#: ../../03_Image_analysis/Data_management.md:0
#: ../../03_Image_analysis/Intensity.md:0
#: ../../03_Image_analysis/Shape.md:0
#: ../../03_Image_analysis/Tracking.md:0
msgid "📚🤷‍♀️ Where can I learn more?"
msgstr ""

#: ../../03_Image_analysis/Classification.md:41
msgid "📄 [Data-analysis strategies for image-based cell profiling](https://www.nature.com/articles/nmeth.4397) {cite}`Caicedo2017-ks`"
msgstr ""

#: ../../03_Image_analysis/Classification.md:42
msgid "📄 [Scoring diverse cellular morphologies in image-based screens with iterative feedback and machine learning](https://www.pnas.org/doi/10.1073/pnas.0808843106) {cite}`Jones2009-zz`"
msgstr ""

#: ../../03_Image_analysis/Classification.md:43
msgid "🎥 [iBiology video series: Measurement and Phenotype Classification](https://www.youtube.com/watch?v=Odi9pIerT7I)"
msgstr ""

#: ../../03_Image_analysis/Classification.md:44
msgid "📄 [Interpreting Image-based Profiles using Similarity Clustering and Single-Cell Visualization](https://doi.org/10.1002/cpz1.713) {cite}`GarciaFossaCruz2023`"
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:1
msgid "Colocalization"
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:4
msgid "What is colocalization?"
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:6
msgid "Colocalization is when two or more different labels (e.g., eGFP and mCherry) spatially overlap in your image (also called co-occurrence). Another component of colocalization is that the fluorescent labels often correlate in intensity (i.e., pixels with brighter eGFP also have brighter mCherry). It is very important to measure colocalization quantitatively–**do not just trust your eyes!**"
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:8
msgid "It is also important to recognize that co-occurrence does not _necessarily_ imply interaction."
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:11
msgid "As an example, it is possible for two people to work in the same building and never interact."
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:21
msgid "degrees of colocalization"
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:21
msgid "**Colocalization is about intensity and spatial overlap of labels**"
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:28
msgid "There are two main branches for how to look at colocalization: Object-based and correlation-based."
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:30
msgid "**Object-based colocalization** is appropriate when you want to be able to say something about a fraction of objects being positive for multiple labels (e.g., 99% of eGFP+ cells were also mCherry+). Here’s a sample workflow:"
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:51
msgid "**Correlation-based colocalization** is more appropriate when the labels you’re measuring are not found in discrete objects or when you predict that the signal in the labels should correlate. Correlation-based colocalization is simpler to measure as it does not require any {term}`segmentation` of objects. Pearson correlation coefficients are readily measurable in most image analysis softwares (e.g., FIJI, CellProfiler, etc.)."
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:55
msgid "**Not having adequate controls** (e.g., single label controls where only one fluorescent label is present). Noise, uneven illumination, and other technical artifacts can cause correlation between two channels"
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:56
msgid "**Not correcting for shift between channels** It’s common to have some degree of shift between different imaging channels due to differences in optics (e.g., different filter cubes). Not correcting for this shift (e.g., by measuring it and applying the corresponding correction to the channels) can limit your ability to detect colocalization."
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:57
msgid "**Bleedthrough** Sometimes signal from one fluorescent channel bleeds into another, which can falsely increase your detected colocalization. This happens most commonly with fluorophores that are similar in spectra (e.g., GFP and YFP). This is caused by fluorophore in one channel being weakly excited by light used to excite a different fluorophore and the resulting emitted fluorescence makes it through the emission filter (e.g., your green objects show up in the yellow channel as well). Assess this with single label controls."
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:58
msgid "**Improper interpretation of your colocalization metric** The details of the metric chosen here are very important - they vary widely in terms of their inclusion of background pixels, their sensitivity to signal-to-noise, etc. Consultation with an expert and the inclusion of proper controls can help you be assured that your measurement is truly what you think it is."
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:63
msgid "🔢 [Theoretical background on colocalization](https://svi.nl/ColocalizationTheory)"
msgstr ""

#: ../../03_Image_analysis/Colocalization.md:64
msgid "📄 [Image co-localization - co-occurrence versus correlation](https://journals.biologists.com/jcs/article/131/3/jcs211847/77151/Image-co-localization-co-occurrence-versus) {cite}`Aaron2018-qi`"
msgstr ""

#: ../../03_Image_analysis/Common_types_of_analysis.md:1
#: ../../03_Image_analysis/Software.md:1
msgid "Default Caption"
msgstr ""

#: ../../03_Image_analysis/Common_types_of_analysis.md:1
msgid "Common types of analysis"
msgstr ""

#: ../../03_Image_analysis/Common_types_of_analysis.md:3
msgid "In this section,  we detail information and tips about some common, very useful types of image analysis. We first describe two aspects of morphological analysis. **Morphology** is the study of the appearance, form, and structure of an object and morphological measurements include those based on object **shape** (e.g. cell area) as well as **intensity** (e.g. mean brightness of GFP). After describing shape-based and intensity-based analyses, we also present tips for measuring **colocalization**, **tracking objects**, and **classifying objects**."
msgstr ""

#: ../../03_Image_analysis/Data_management.md:1
msgid "Data management and sharing"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:3
msgid "What is it?"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:5
msgid "Both journals and scientific funders have placed more emphasis in recent years on the fact that it is critical to save both the raw data generated during the course of scientific discovery as well as the workflows used to process such data. While mandates to publicly deposit raw image data have recently gone into place in several countries, it can be difficult for researchers to know where to store images, code, and metadata associated with their bioimaging experiments."
msgstr ""

#: ../../03_Image_analysis/Data_management.md:0
msgid "🤔 What are my options?"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:9
msgid "There are many options for storing image data in online repositories. These services make it easy to share and reuse data. The best option will depend on the size of the dataset, the budget for storage, whether there is related non-image data, and how much metadata is available for the dataset. Some options are summarized below:"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:0
msgid "Comparison of various data respositories"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:18
msgid "**Options for storing bioimaging data** Figure by Beth Cimini (2023) [Source](https://doi.org/10.5281/zenodo.7628604)"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:24
msgid "**Not storing original versions of images**. It is critical that the raw image data be saved and stored. It is very important that these files are not compressed formats (e.g., '.jpeg') or modified from the original files on which analysis was performed. If files are modified, measurements will change and the analysis pipeline will not be reproducible to anyone else."
msgstr ""

#: ../../03_Image_analysis/Data_management.md:26
msgid "**Necessary imaging metadata is unavailable**. In order to properly calibrate measurement data, it's critical that information like pixel size (e.g., in microns), the microscope manufacturer and model, and acquisition settings are included alongside the data. If this isn't included, it will be very difficult to reproduce results or use the combine the data with other datasets."
msgstr ""

#: ../../03_Image_analysis/Data_management.md:32
msgid "🌐 [Zenodo](https://zenodo.org/)"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:33
msgid "🌐 [Figshare](https://figshare.com)"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:34
msgid "🌐 [Dryad](https://datadryad.org)"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:35
msgid "🌐 [Bioimage Archive](https://www.ebi.ac.uk/bioimage-archive/) {cite}`Hartley2022-mt`"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:36
msgid "🌐 [Image Data Resource (IDR)](https://idr.openmicroscopy.org/) {cite}`Williams2017-yy`"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:37
msgid "🌐 [BBBC](https://bbbc.broadinstitute.org/) {cite}`Ljosa2012-fr`"
msgstr ""

#: ../../03_Image_analysis/Data_management.md:38
msgid "🌐 [Cell Painting Gallery](https://registry.opendata.aws/cellpainting-gallery/)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:1
msgid "General Use Software"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:3
msgid "Tools on this page tend to be relatively multipurpose across a lot of kinds of analyses and/or images. For tools that specialize in certain analysis steps, see the [Specific Use Software](./SpecificUseSoftware.md) page."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:6
msgid "<img src=\"https://imagej.net/media/icons/imagej2.png\" alt=\"logo\" width=\"30px\"> ImageJ"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:9
msgid "[ImageJ](https://imagej.net/) {cite}`Schneider2012-gs,Schindelin2012-kk,Rueden2017-ku` is an imaging processing program that is capable of operating on a variety of images including multichannel, 3D and time series. It provides a variety of basic imaging processing operations, but it can be complemented with a variety of plugins for more complex tasks."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:0
#: ../../03_Image_analysis/SpecificUseSoftware.md:0
msgid "What are its disadvantages?"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:14
msgid "Imaging processing operations are done one at a time, while it has the capability of batch processing and creating macros it does require some understanding of coding."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:15
msgid "But while it can open large images, there is a size limit to the size it can handle based on the available memory. And even if it can open large images it can slow down performance."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:0
#: ../../03_Image_analysis/SpecificUseSoftware.md:0
msgid "How to download/install and learn more?"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:20
msgid "To download ImageJ or its “batteries-included” distribution Fiji go to"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:24
msgid "🌐 [ImageJ download](https://imagej.net/downloads)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:26
msgid "For documentation and tutorials on how to use ImageJ as well as a list of available plugins"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:30
msgid "🌐 [ImageJ basics](https://imagej.net/learn/)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:34
msgid "<img src=\"https://us1.discourse-cdn.com/flex015/uploads/imagej/original/2X/b/bcdcd5ba157e07e74dd1964ec81765e708455ed9.png\" alt=\"logo\" width=\"30px\"> CellProfiler"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:36
msgid "[CellProfiler](https://cellprofiler.org/) {cite}`Stirling2021-sg` is a software designed for biologists by biologists; it creates a bridge between image analysis and scientist with no need of computational expertise. It was designed with the idea of an image analysis pipeline in mind: it allows you to take a series of interoperable modules to design your own custom analysis pipeline that can be applied to one or thousands of images, making it suitable for high throughput image analysis."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:42
msgid "CellProfiler can’t handle large images, like whole tissue sections from histology experiments. The image size is currently limited by the available memory on your computer."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:43
msgid "While CellProfiler can perform analysis on 3D images the visualization is limited to a one z-plane at a time via a slider on the viewing window."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:44
msgid "Also several features of CellProfiler are only available for 2D images."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:49
msgid "To download CellProfiler"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:53
msgid "🌐 [CellProfiler download](https://cellprofiler.org/releases)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:55
#: ../../03_Image_analysis/GeneralUseSoftware.md:84
#: ../../03_Image_analysis/GeneralUseSoftware.md:113
#: ../../03_Image_analysis/GeneralUseSoftware.md:142
#: ../../03_Image_analysis/GeneralUseSoftware.md:172
#: ../../03_Image_analysis/SpecificUseSoftware.md:59
msgid "For documentation examples and  tutorials."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:59
msgid "🌐 [CellProfiler user manual](https://cellprofiler-manual.s3.amazonaws.com/CellProfiler-4.2.6/index.html)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:60
msgid "🌐 [Examples](https://cellprofiler.org/examples)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:61
msgid "🌐 [Tutorials](https://tutorials.cellprofiler.org/)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:62
msgid "🎥 [Video tutorials and workshops](https://www.youtube.com/playlist?list=PLXSm9cHbSZBBy7JkChB32_e3lURUcT3RL)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:66
msgid "<img src=\"https://us1.discourse-cdn.com/flex015/uploads/imagej/optimized/3X/6/0/6039b2daa4b6b1c32943f63f464cf3c477898bfe_2_750x750.png\" alt=\"logo\" width=\"30px\"> QuPath"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:68
msgid "[QuPath](https://qupath.github.io/) {cite}`Bankhead2017-kz` offers a wide set of image analysis tools that can be applied to whole slide images. For that reason it is widely utilized with pathology images, but it can be used with other images as well. QuPath also contains pixel classification tools and can integrate with ImageJ (e.g., for sending {term}`ROIs` between the programs, or for accessing ImageJ plugins)."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:73
msgid "To get the most out of QuPath (especially when analyzing many images), some scripting and knowledge of coding (or adapting other’s code) is necessary"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:78
msgid "To download QuPath"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:82
msgid "🌐 [QuPath download](https://qupath.github.io/)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:88
msgid "🌐 [QuPath user manual](https://qupath.readthedocs.io/en/stable/)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:89
msgid "🎥 [Video tutorials and workshops](https://www.youtube.com/c/qupath)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:93
msgid "<img src=\"https://icy.bioimageanalysis.org/wp-content/uploads/2018/07/logo_full_notext600px.png\" alt=\"logo\" width=\"30px\"> Icy"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:95
msgid "[Icy](https://icy.bioimageanalysis.org/) {cite}`De_Chaumont2012-pe` is an out of the box image analysis tools, it utilizes plugins to create visual image analysis protocols that can be shared with other users."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:102
msgid "Icy interoperability with other softwares is limited to ImageJ"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:107
msgid "To download Icy"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:111
msgid "🌐 [Icy download](https://icy.bioimageanalysis.org/download/)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:117
msgid "🌐 [Icy course and tutorial](https://icy.bioimageanalysis.org/trainings/)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:118
msgid "🎥 [Bioimage analysis with Icy ](https://www.youtube.com/watch?v=myal9BD6J-k)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:122
msgid "<img src=\"https://us1.discourse-cdn.com/flex015/uploads/imagej/optimized/3X/7/4/74273a1f9a663b52053d44c9767ed49193f2170f_2_787x750.png\" alt=\"logo\" width=\"30px\"> MIB (Microscopy Image Browser)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:124
msgid "[MIB](http://mib.helsinki.fi/index.html) {cite}`Belevich2016-vi` is a user-friendly software for image analysis of multidimensional datasets for both light and electron microscopy. It allows you to use the whole acquired data for its analysis and extraction of morphological features."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:131
msgid "It was created using MATLAB, a standalone packaged version exist, but they do not use the most up-to-date MATLAB releases"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:136
msgid "To download MIB"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:140
msgid "🌐 [MIB download](http://mib.helsinki.fi/downloads.html)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:146
msgid "🌐 [MIB user guide](http://mib.helsinki.fi/help/main2/im_browser_user_guide.html)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:147
msgid "🌐 [Tutorials](http://mib.helsinki.fi/tutorials.html)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:148
msgid "🎥 [Video tutorials](https://www.youtube.com/playlist?list=PLGkFvW985wz8cj8CWmXOFkXpvoX_HwXzj)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:152
msgid "<img src=\"https://us1.discourse-cdn.com/flex015/uploads/imagej/optimized/3X/7/7/775e83f70639e1cb7cb299d8681d272e18718089_2_750x750.png\" alt=\"logo\" width=\"30px\"> napari"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:154
msgid "[napari](https://napari.org/) {cite}`Sofroniew2022-nd` is being developed as a multi-dimensional image viewer that can be expanded via a variety of plugins to perform basic and complex image analysis tasks."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:161
msgid "napari is still in the development stages, but it is a very popular platform and already has a variety of plugins and use cases with tutorials available."
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:166
msgid "To download napari"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:170
msgid "🌐 [napari bundled app download ](https://napari.org/stable/tutorials/fundamentals/installation.html#install-as-a-bundled-app)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:176
msgid "🌐 [New to napari guide ](https://napari.org/stable/tutorials/fundamentals/getting_started.html)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:177
msgid "🌐 [Tutorials](https://napari.org/stable/tutorials/index.html)"
msgstr ""

#: ../../03_Image_analysis/GeneralUseSoftware.md:178
msgid "🌐 [Bioimage Analysis with Napari - A FocalPlane blog series](https://focalplane.biologists.com/category/blog-series/bio-image-analysis-with-napari/)"
msgstr ""

#: ../../03_Image_analysis/Intensity.md:1
msgid "Intensity measurements"
msgstr ""

#: ../../03_Image_analysis/Intensity.md:3
msgid "What are intensity measurements?"
msgstr ""

#: ../../03_Image_analysis/Intensity.md:4
msgid "Intensity refers to the brightness of signal for a fluorescent label. Using intensity measurements, we can infer a relative amount of fluorophore or stain. So for instance, if you have a protein tagged with a fluorophore, you can measure the intensity of that fluorophore to get a relative measure of how much protein is present in your sample. Intensity measurements include the following (non-exhaustive) and can be measured within an image, in a object like a cell, in subregions of an object:"
msgstr ""

#: ../../03_Image_analysis/Intensity.md:5
msgid "**Mean intensity**: the average intensity across all pixels"
msgstr ""

#: ../../03_Image_analysis/Intensity.md:6
msgid "**Integrated intensity**: the sum of pixel intensities, a proxy for the total amount of that marker in an object"
msgstr ""

#: ../../03_Image_analysis/Intensity.md:7
msgid "**Texture measurements**: the smoothness of the intensities"
msgstr ""

#: ../../03_Image_analysis/Intensity.md:13
msgid "Intensity is relatively straightforward to measure, but can be quite tricky to do _correctly_ (see below). We strongly suggest you contact an image analysis expert before proceeding with this type of analysis because there are so many places things can go wrong. In general, you want to measure on either raw images, or illumination-corrected images, but in general with minimal {term}`image processing`. Illumination-correction is a form of {term}`image processing` to compensate for the uneven pattern of illumination produced by most light sources where the middle of the field of illumination is brighter than the edges. Then intensity measurements can be made in any standard image analysis software, either across the whole image or in identified objects. See below for an example workflow:"
msgstr ""

#: ../../03_Image_analysis/Intensity.md:35
msgid "To understand saturation another way, imagine you’re trying to measure average male height with a 2 meter tape measure. If our sample contains men that are taller than 2 meters, we can’t tell _how much taller_ than 6 feet they are; they’re like saturated pixels that exceed the intensity we can detect. This saturation of our measurement tool means we can’t accurately report average height."
msgstr ""

#: ../../03_Image_analysis/Intensity.md:38
msgid "**Saturation** Saturated pixels are so bright their intensity values max out our detector (camera). If you have saturated pixels in the cells you’re trying to measure, you really can’t do most intensity measurements. This is because for saturated pixels, you don’t know how bright they really are, just that they’re brighter than you can detect. There are some intensity measurements that are robust to some saturation. For example, the median intensity of an image won’t be affected by saturation unless you have >½ the image saturated. But measurements like mean intensity will be affected by saturation."
msgstr ""

#: ../../03_Image_analysis/Intensity.md:40
msgid "**Inadequate controls** In most cases, the exact intensity measures you get don’t mean anything biologically in isolation. It’s only by comparison of conditions that we can generate some biological insight. A control condition is therefore **very** important to compare to your experimental condition."
msgstr ""

#: ../../03_Image_analysis/Intensity.md:42
msgid "**Not matching imaging conditions across experimental conditions** Because intensity measures are affected by exposure time, light source intensity, and other factors, it’s very important to match imaging settings across your samples. Relatedly, you should make sure you don’t separate imaging your experimental and control conditions to different days if this can be in any way avoided. Fluorophores can become dimmer over time in samples, which complicates interpretation if different sample types were imaged on different days."
msgstr ""

#: ../../03_Image_analysis/Intensity.md:48
msgid "🎓 [Neubias training resource on intensity measures](https://neubias.github.io/training-resources/measure_intensities/index.html)"
msgstr ""

#: ../../03_Image_analysis/Introduction.md:1
#: ../../03_Image_analysis/Software.md:3
#: ../../03_Image_analysis/_notinyet_Image_processing.md:3
#: ../../03_Image_analysis/_notinyet_Image_segmentation.md:3
msgid "Introduction"
msgstr ""

#: ../../03_Image_analysis/Introduction.md:3
msgid "Microscopy images are inherently quantitative, which makes them a very powerful data source. As a biologist, image analysis allows you to translate these numbers into insights that answer biological questions. For our purposes, **image analysis** is the process of measuring aspects of biological phenomena captured in microscopy images. Microscopy images are already _inherently quantitative_ in that they are matrices (i.e., grids) of numbers. However, image analysis is the process of turning these raw numbers into biologically interpretable measurements. Image analysis typically involves a series of steps that can be collected into a pipeline or analysis workflow. A simple example workflow is shown below:"
msgstr ""

#: ../../03_Image_analysis/Introduction.md:26
msgid "The specifics of your workflow depend on your biological question. Below we present a few common types of analysis for fluorescence microscopy experiments. For each, we’ll explain key ideas to understand before you begin, common pitfalls, and links to a few key resources to learn more. We encourage you to think about your analysis strategy even before beginning sample preparation. While not always possible, speaking with an image analysis expert{cite}`Cimini2024-uh` in your local core facility or asking a question on the [image.sc](https://image.sc) forum _before you begin_ can save you a ton of time and headache when it comes to designing an image analysis strategy. Some general concepts are also covered in a recent short review{cite}`Cimini2024-sz`."
msgstr ""

#: ../../03_Image_analysis/Introduction.md:28
msgid "Note that \"automated\" and \"quantified\" are not at all the same as \"accurate\" \"unbiased\" - see {cite}`Jost2019-nx` and {cite}`Lee2024-zu` for further treatment of these issues."
msgstr ""

#: ../../03_Image_analysis/Resources.md:1
msgid "Resources for learning more"
msgstr ""

#: ../../03_Image_analysis/Resources.md:7
msgid "**Resource Name**"
msgstr ""

#: ../../03_Image_analysis/Resources.md:8
msgid "**Link**"
msgstr ""

#: ../../03_Image_analysis/Resources.md:9
msgid "**Brief description**"
msgstr ""

#: ../../03_Image_analysis/Resources.md:10
msgid "🌐 Image.sc {cite}`Rueden2019-qp`"
msgstr ""

#: ../../03_Image_analysis/Resources.md:11
msgid "[link](https://forum.image.sc/)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:12
msgid "Discussion forum for bioimage analysis software"
msgstr ""

#: ../../03_Image_analysis/Resources.md:13
msgid "🌐 GloBIAS Bioimage Analyst Database"
msgstr ""

#: ../../03_Image_analysis/Resources.md:14
msgid "[link](https://www.globias.org/resources/bioimage-analyst-db)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:15
msgid "Find the friendly local bioimage analyst near you!"
msgstr ""

#: ../../03_Image_analysis/Resources.md:16
msgid "🌐 Peter Bankhead’s Intro to Bioimage Analysis"
msgstr ""

#: ../../03_Image_analysis/Resources.md:17
msgid "[link](https://bioimagebook.github.io/)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:18
msgid "Guide for absolute beginners to image analysis, including embedded questions/answers, exercises with Python and ImageJ, and videos to check understanding"
msgstr ""

#: ../../03_Image_analysis/Resources.md:19
msgid "📖 Bioimage Data Analysis {cite}`Miura2016-wq`"
msgstr ""

#: ../../03_Image_analysis/Resources.md:20
msgid "[link](https://analyticalscience.wiley.com/do/10.1002/was.00050003/full/bioimagedataanalysis.pdf)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:21
msgid "A free online textbook introducing various topics by Bioimage Analysis experts, edited by Kota Miura"
msgstr ""

#: ../../03_Image_analysis/Resources.md:22
msgid "📄 Reproducible image handling and analysis {cite}`Miura2021-mb`"
msgstr ""

#: ../../03_Image_analysis/Resources.md:23
msgid "[link](https://www.embopress.org/doi/full/10.15252/embj.2020105889)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:24
msgid "An article reviewing major pitfalls in image handling and how to avoid them and create reproducible analysis workflows"
msgstr ""

#: ../../03_Image_analysis/Resources.md:25
msgid "📄 Made to measure: an introduction to quantification in microscopy data"
msgstr ""

#: ../../03_Image_analysis/Resources.md:26
msgid "[link](https://arxiv.org/abs/2302.01657#) {cite}`Culley2023-dj`"
msgstr ""

#: ../../03_Image_analysis/Resources.md:27
msgid "An article describing several common classes of measurements made in microscopy data, as well as technical factors that may affect the results"
msgstr ""

#: ../../03_Image_analysis/Resources.md:28
msgid "🌐 Metrics Reloaded {cite}`Maier-Hein2024-cp,Reinke2024-aj,reinke2023common`"
msgstr ""

#: ../../03_Image_analysis/Resources.md:29
msgid "[link](https://metrics-reloaded.dkfz.de/)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:30
msgid "An interactive guide to choosing the right metric for your scientific question"
msgstr ""

#: ../../03_Image_analysis/Resources.md:31
msgid "📄 A Hitchhiker's guide through the bio-image analysis software universe {cite}`Haase2022-ad`"
msgstr ""

#: ../../03_Image_analysis/Resources.md:32
msgid "[link](https://febs.onlinelibrary.wiley.com/doi/full/10.1002/1873-3468.14451)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:33
msgid "An article that gives guidance and a glossary of available image analysis software and packages"
msgstr ""

#: ../../03_Image_analysis/Resources.md:34
msgid "🌐 BioImage Informatics Index"
msgstr ""

#: ../../03_Image_analysis/Resources.md:35
msgid "[link](https://biii.eu/)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:36
msgid "Repository platform for searching bioimage analysis tools and workflows based on the problem, method or software of choice"
msgstr ""

#: ../../03_Image_analysis/Resources.md:37
msgid "🎥 iBiology Bioimage Analysis video series"
msgstr ""

#: ../../03_Image_analysis/Resources.md:38
msgid "[link](https://youtu.be/1xo4vi6Ub4I)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:39
msgid "Video series that introduces Bioimage analysis, including overviews of {term}`image processing`, {term}`segmentation`, tracking, making and interpreting measurements, tips and pitfalls"
msgstr ""

#: ../../03_Image_analysis/Resources.md:40
msgid "🌐 Bioimage ANalysis Desktop (BAND)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:41
msgid "[link](https://band.embl.de)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:42
msgid "Access to virtual desktops allowing access to bioimage analysis software from a browser"
msgstr ""

#: ../../03_Image_analysis/Resources.md:43
msgid "🌐 Galaxy Imaging Node"
msgstr ""

#: ../../03_Image_analysis/Resources.md:44
msgid "[link](https://imaging.usegalaxy.eu/)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:45
msgid "A Galaxy node prepopulated with a number of open-source image analysis tools and workflows, making it easy to create and share reproducible FAIR workflows"
msgstr ""

#: ../../03_Image_analysis/Resources.md:46
msgid "🌐 Image Analysis Training Resources"
msgstr ""

#: ../../03_Image_analysis/Resources.md:47
msgid "[link](https://neubias.github.io/training-resources/index.html)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:48
msgid "Image analysis traning resources collected by NEUBIAS, styled based on The Carpentries"
msgstr ""

#: ../../03_Image_analysis/Resources.md:49
msgid "🌐 Bioimage Analysis AwesomeList"
msgstr ""

#: ../../03_Image_analysis/Resources.md:50
msgid "[link](https://github.com/hallvaaw/awesome-biological-image-analysis)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:51
msgid "A curated [AwesomeList](https://github.com/sindresorhus/awesome) of resources related to bioimage analysis"
msgstr ""

#: ../../03_Image_analysis/Resources.md:52
msgid "🌐 Bioimage Analysis Notebooks"
msgstr ""

#: ../../03_Image_analysis/Resources.md:53
msgid "[link](https://haesleinhuepf.github.io/BioImageAnalysisNotebooks/)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:54
msgid "A collection of Python Jupyter notebooks for BioImageAnalysis, GPU-accelerated image processing, bio-image data science and more"
msgstr ""

#: ../../03_Image_analysis/Resources.md:55
msgid "📄 Believing is seeing - the deceptive influence of bias in quantitative microscopy{cite}`Lee2024-zu`"
msgstr ""

#: ../../03_Image_analysis/Resources.md:56
msgid "[link](https://doi.org/10.1242/jcs.261567)"
msgstr ""

#: ../../03_Image_analysis/Resources.md:57
msgid "Thoughtful review on technological and especially psychological factors that can lead to biased microscopy results"
msgstr ""

#: ../../03_Image_analysis/Shape.md:1
msgid "Size and Shape measurements"
msgstr ""

#: ../../03_Image_analysis/Shape.md:3
msgid "What are size measurements?"
msgstr ""

#: ../../03_Image_analysis/Shape.md:4
msgid "Size measurements describe the dimensions of objects in your image. Common size measurements include:"
msgstr ""

#: ../../03_Image_analysis/Shape.md:6
msgid "**Area**: the 2D space an object takes up in the image or the 3D surface area of an object"
msgstr ""

#: ../../03_Image_analysis/Shape.md:7
msgid "**Volume**: the 3D space an object takes up in a 3D image"
msgstr ""

#: ../../03_Image_analysis/Shape.md:8
msgid "**Perimeter**: the distance around the edge of an object"
msgstr ""

#: ../../03_Image_analysis/Shape.md:10
msgid "What are shape measurements?"
msgstr ""

#: ../../03_Image_analysis/Shape.md:11
msgid "Shape measurements describe the 2D or 3D form of objects in our sample. Common shape measurements include:"
msgstr ""

#: ../../03_Image_analysis/Shape.md:13
msgid "**Circularity**: How round vs. elongated an object is. Formally defined as $circularity = 4pi*{area}/{perimeter}^2$ where 1 is a perfect circle and circularity <1 is a more elongated polygon."
msgstr ""

#: ../../03_Image_analysis/Shape.md:14
msgid "**Solidity**: how dense vs. wispy/holey an object is. Formally defined as $solidity = area/convex area$ where _convex area_ is akin to the area inside a shape formed by stretching a rubber band around the object."
msgstr ""

#: ../../03_Image_analysis/Shape.md:20
msgid "After segmenting an image to locate the pixels belonging to different objects, morphology can be measured readily in many image analysis softwares, like FIJI and CellProfiler. For example, in {term}`Fiji`, after identifying your objects as {term}`ROIs`, be sure to **Analyze > Set Measurements…** and select “Shape Descriptors” then simply measure your {term}`ROIs` with **Analyze > Measure**. In CellProfiler, this is accomplished using the module MeasureObjectSizeShape."
msgstr ""

#: ../../03_Image_analysis/Shape.md:25
msgid "**Not understanding the limitations of your images**. All biological structures are 3D, but we often analyze 2D images. Often this is still very useful! But the larger and more complex your objects (e.g., neurons in a tissue section), the more limited a 2D view becomes."
msgstr ""

#: ../../03_Image_analysis/Shape.md:26
msgid "**Failing to use calibrated units**. Be sure to properly calibrate your images and present final measurements in microns (or similar units). If measuring 3D images, be sure to take into account the z-step, which is likely larger than the xy pixel size."
msgstr ""

#: ../../03_Image_analysis/Shape.md:31
msgid "📄 [Current Methods and Pipelines for Image-Based Quantitation of Nuclear Shape and Nuclear Envelope Abnormalities](https://www.mdpi.com/2073-4409/11/3/347/htm) {cite}`Janssen2022-bm`"
msgstr ""

#: ../../03_Image_analysis/Shape.md:32
msgid "🌐 [Description of morphological measurements made by CellProfiler](https://cellprofiler-manual.s3.amazonaws.com/CellProfiler-4.2.4/modules/measurement.html#id20)"
msgstr ""

#: ../../03_Image_analysis/Shape.md:33
msgid "🎓 [Plain language description of various morphological measures by Michael Wirth](http://www.cyto.purdue.edu/cdroms/micro2/content/education/wirth10.pdf)"
msgstr ""

#: ../../03_Image_analysis/Software.md:1
msgid "Open source software"
msgstr ""

#: ../../03_Image_analysis/Software.md:5
msgid "When it comes time to select a software program for your image analysis, there are many options, some more general and others highly specialized to specific image modalities or types of experiments. In general, a good place to start exploring is by examining papers in your field and seeing what others have used to analyze similar experiments to your own. It’s important to note that there isn’t one correct answer to \"Which program should I use?\" Depending on your biological question, your images, and your own comfort with coding, there are many options available."
msgstr ""

#: ../../03_Image_analysis/Software.md:7
msgid "Below, we summarize the use-cases and limitations of some of the most common, free and open-source software for image analysis, this is a small list and more extensive ones exist like the [A Hitchhiker's guide through the bio-image analysis software universe](https://febs.onlinelibrary.wiley.com/doi/full/10.1002/1873-3468.14451) {cite}`Haase2022-ad` and the [BioImage Informatics Index](https://biii.eu) {cite}`Paul-Gilloteaux2021-vw`."
msgstr ""

#: ../../03_Image_analysis/Software.md:9
msgid "Whatever software you choose, be sure to include a detailed description of your analysis in your methods, with pipeline or workflow files if possible, so others can reproduce your work. Also be sure to cite the analysis software you use! This helps developers of the software get grant funding and helps others find useful tools."
msgstr ""

#: ../../03_Image_analysis/Software.md:17
#: ../../03_Image_analysis/Software.md:26
#: ../../03_Image_analysis/Software.md:35
#: ../../03_Image_analysis/Software.md:44
#: ../../03_Image_analysis/Software.md:53
#: ../../03_Image_analysis/Software.md:62
#: ../../03_Image_analysis/Software.md:71
#: ../../03_Image_analysis/Software.md:80
#: ../../03_Image_analysis/Software.md:89
msgid "card-img-top"
msgstr ""

#: ../../03_Image_analysis/Software.md:0
msgid "ImageJ"
msgstr ""

#: ../../03_Image_analysis/Software.md:21
msgid "[ImageJ](https://imagej.net/) is an imaging processing program that is capable of operating on a variety of images including multichannel, 3D and time series. It provides basic imaging processing operations and has a variety of plugins for more complex tasks. [Read more...](content/imagej)"
msgstr ""

#: ../../03_Image_analysis/Software.md:0
msgid "CellProfiler"
msgstr ""

#: ../../03_Image_analysis/Software.md:30
msgid "[CellProfiler](https://cellprofiler.org/) was designed with the idea of an image analysis pipeline in mind; it allows you to take a series of interoperable modules to design your own custom analysis pipeline that can be applied to one or thousands of images, making it suitable for high throughput image analysis. [Read more...](content/cellprofiler)"
msgstr ""

#: ../../03_Image_analysis/Software.md:0
msgid "QuPath"
msgstr ""

#: ../../03_Image_analysis/Software.md:39
msgid "[QuPath](https://qupath.github.io/) offers a wide set of image analysis tools that can be applied to whole slide images like pathology images, but it can be used with other images as well. QuPath also contains pixel classification tools and can integrate with ImageJ. [Read more...](content/qupath)"
msgstr ""

#: ../../03_Image_analysis/Software.md:0
msgid "Icy"
msgstr ""

#: ../../03_Image_analysis/Software.md:48
msgid "[Icy](https://icy.bioimageanalysis.org/) is an out of the box image analysis tools, it utilizes plugins to create visual image analysis protocols that can be shared with other users. [Read more...](content/icy)"
msgstr ""

#: ../../03_Image_analysis/Software.md:0
msgid "MIB"
msgstr ""

#: ../../03_Image_analysis/Software.md:57
msgid "[MIB](http://mib.helsinki.fi/index.html) is a user-friendly software for image analysis of multidimensional datasets for both light and electron microscopy. It allows you to use the whole acquired data for its analysis and extraction of morphological features. [Read more...](content/mib)"
msgstr ""

#: ../../03_Image_analysis/Software.md:0
msgid "napari"
msgstr ""

#: ../../03_Image_analysis/Software.md:66
msgid "[napari](https://napari.org/) is being developed as a multi-dimensional image viewer that can be expanded via a variety of plugins to perform basic and complex image analysis tasks. [Read more...](content/napari)"
msgstr ""

#: ../../03_Image_analysis/Software.md:0
msgid "Cellpose"
msgstr ""

#: ../../03_Image_analysis/Software.md:75
msgid "[Cellpose](https://www.cellpose.org/) is a {term}`segmentation` algorithm, it provides a graphical user interface that allows users to use trained models or train their own using their images and annotations. [Read more...](content/cellpose)"
msgstr ""

#: ../../03_Image_analysis/Software.md:0
msgid "ilastik"
msgstr ""

#: ../../03_Image_analysis/Software.md:84
msgid "[ilastik](https://www.ilastik.org/) is a tool for interactive image classification, {term}`segmentation` and analysis. It leverages machine-learning algorithms to perform pixel and object-level classification. Using it requires no experience in {term}`image processing`. [Read more...](content/ilastik)"
msgstr ""

#: ../../03_Image_analysis/Software.md:0
msgid "Piximi"
msgstr ""

#: ../../03_Image_analysis/Software.md:93
msgid "[Piximi](https://www.piximi.app/) is an application for annotation and classification that runs entirely from your browser and requires no installation and minimal setup.  [Read more...](content/piximi)"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:1
msgid "Specific Use Software"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:3
msgid "Tools on this page tend to be extremely good at certain tasks, but are less intended for a wide range of use cases. For tools with broader areas of focus, see the [General Use Software](./GeneralUseSoftware.md) page."
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:6
msgid "<img src=\"https://www.cellpose.org/static/images/cellpose_transparent.png\" alt=\"logo\" width=\"30px\"> Cellpose"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:8
msgid "[Cellpose](https://www.cellpose.org/) {cite}`Stringer2021-uq` is a {term}`segmentation` algorithm, it provides a graphical user interface that allows users to use trained models or train their own using their images and annotations."
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:0
msgid "What type of image analysis problem is it best at?"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:13
msgid "Object {term}`segmentation`, most trained models are for cell {term}`segmentation` but could be applied to segment other similar objects"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:20
msgid "Its use requires some computational knowledge."
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:21
msgid "Training a new model requires manual annotation correction that can be time consuming, but is likely less time consuming than other methods of training models."
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:28
msgid "🌐 [Installation instructions](https://cellpose.readthedocs.io/en/latest/installation.html)"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:29
msgid "🎥 [How to use tutorial](https://www.youtube.com/watch?v=5qANHWoubZU)"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:33
msgid "<img src=\"https://us1.discourse-cdn.com/flex015/uploads/imagej/original/3X/9/f/9f5be5e138c63bc6a50be0bb0027b8eef0194935.png\" alt=\"logo\" width=\"30px\"> ilastik"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:35
msgid "[ilastik](https://www.ilastik.org/) {cite}`Berg2019-no` is a tool for interactive image classification, {term}`segmentation` and analysis. It leverages machine-learning algorithms to perform pixel and object-level classification. Using it requires no experience in {term}`image processing`."
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:40
msgid "It can be used for both instance {term}`segmentation` and semantic {term}`segmentation`. It does also perform {term}`segmentation` and tracking, though with somewhat fewer tunable parameters than some other tools offer."
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:47
msgid "Sometimes loading or exporting images can require a bit of troubleshooting to get the dimensions correct."
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:48
msgid "ilastik is limited by your computer’s RAM so training a model with lots of features or working with very large images is likely to slow you down."
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:53
msgid "To download ilastik:"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:57
msgid "🌐 [ilastik download ](https://www.ilastik.org/download.html)"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:63
msgid "🌐 [User guide](https://www.ilastik.org/documentation/index.html#user-documentation)"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:67
msgid "<img src=\"https://us1.discourse-cdn.com/flex015/uploads/imagej/original/3X/3/f/3fe4d974194caabdb61a5574e24402db8484ab9b.png\" alt=\"logo\" width=\"30px\"> Piximi"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:70
msgid "[Piximi](https://www.piximi.app/) {cite}`Moser2024-vt` is an application for annotation and classification that runs entirely from your browser and requires no installation and minimal setup."
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:75
msgid "Piximi can do image classification using machine learning"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:82
msgid "It is still in the developing phases and some of its proposed features are not available yet"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:89
msgid "🌐 [Piximi website ](https://www.piximi.app/)"
msgstr ""

#: ../../03_Image_analysis/SpecificUseSoftware.md:90
msgid "🌐 [Piximi user guide ](https://documentation.piximi.app/intro.html)"
msgstr ""

#: ../../03_Image_analysis/Tracking.md:1
msgid "Object tracking"
msgstr ""

#: ../../03_Image_analysis/Tracking.md:3
msgid "What is tracking?"
msgstr ""

#: ../../03_Image_analysis/Tracking.md:4
msgid "Tracking, or _object tracking_, refers to the ability to estimate the location of objects in motion from one frame of video to the next. Practically, object tracking in microscopy involves identifying your objects in each frame of your video, then relating objects from frame to frame to be able to identify the same cell as it moves in space."
msgstr ""

#: ../../03_Image_analysis/Tracking.md:10
msgid "There are several options for tracking objects, like the [TrackMate](https://imagej.net/plugins/trackmate/) {cite}`Tinevez2017-fb` plugin in Fiji, but tracking is somewhat more complex to setup than the previous analyses. In addition to just identifying objects in the movie/time series, tracking also allows you to identify splitting (e.g. mitosis) and merging events. There are many measurements that come out of tracking, including spatial measures like where the objects move to, speed measurements, distance traveled (length of track), and rate of splitting events (e.g., mitotic events) and merging events."
msgstr ""

#: ../../03_Image_analysis/Tracking.md:15
msgid "**Poor {term}`segmentation`** If objects are dropping out from frame to frame, this makes it more difficult to track them over time. Accurate {term}`segmentation` is the foundation of good tracking results. This can become more difficult if your objects are also changing in shape or intensity (due to things such as bleaching) over the course of the video. It’s important to find a {term}`segmentation` strategy that can work well across your frames."
msgstr ""

#: ../../03_Image_analysis/Tracking.md:16
msgid "**Inadequate frame rate** If objects are highly dynamic but the images were not taken at a high frequency, tracking can be difficult because objects might have moved too much for the algorithm to relate them from one frame to the next. It is important to match the image acquisition frequency to how dynamic your cells or objects are."
msgstr ""

#: ../../03_Image_analysis/Tracking.md:22
msgid "📄 [Computerized Cell Tracking](https://www.sciencedirect.com/science/article/pii/S2468502X20300711)"
msgstr ""

#: ../../03_Image_analysis/Tracking.md:23
msgid "🌐 [TrackMate Manual](https://imagej.net/media/plugins/trackmate/trackmate-manual.pdf)"
msgstr ""

#: ../../03_Image_analysis/_notinyet_Common_pitfalls.md:1
msgid "Common pitfalls"
msgstr ""

#: ../../03_Image_analysis/_notinyet_Common_pitfalls.md:3
msgid "In addition to the mistakes we summarize for common image analysis tasks, here we present a general list of common mistakes for beginners to image analysis. For each, we explain why this is a problem and make suggestions for how to avoid these issues."
msgstr ""

#: ../../03_Image_analysis/_notinyet_Common_pitfalls.md:5
msgid "Changing your image bit depth"
msgstr ""

#: ../../03_Image_analysis/_notinyet_Common_pitfalls.md:7
msgid "Let's say you've opened your image in {term}`Fiji` and you want to process it in some way (e.g., maximum intensity projection, splitting channels, etc.) before saving it out to measure somewhere else. You go to `Image` > `Type` > `RGB (Color)` since you want a color image to import into your next analysis software and then save the resulting image. What's the problem with this? By making this change, even if the image looks exactly the same to your eyes, you've actually inadvertently changed the intensity values quite a bit! Let's take a look at an example. In Fiji there are several built-in example images. Let's open neuron.tif (you can follow along by going to `File` > `Open Samples...` > `Neuron (5 channels)`."
msgstr ""

#: ../../03_Image_analysis/_notinyet_Common_pitfalls.md:9
msgid "Here's how the image opens. There are 5 channels, but let's just look at the first one: ![image](https://user-images.githubusercontent.com/28116530/206793825-364998d4-6043-4b1d-8438-0a5b37b97232.png)"
msgstr ""

#: ../../03_Image_analysis/_notinyet_Common_pitfalls.md:9
msgid "image"
msgstr ""

#: ../../03_Image_analysis/_notinyet_Image_processing.md:1
msgid "Image Processing"
msgstr ""

#: ../../03_Image_analysis/_notinyet_Image_segmentation.md:1
msgid "Image Segmentation"
msgstr ""