Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
213 changes: 213 additions & 0 deletions docs/source/for_non_devs.rst
Original file line number Diff line number Diff line change
@@ -0,0 +1,213 @@
Upload Images & Segmentations (No Coding Required)
====================================================

:bdg-success:`Beginner`

This guide is for people who need to upload medical images and segmentations to
Datamint, but don't write code. It walks through everything from a brand new
computer to your first successful upload, one step at a time.

.. note::

Already comfortable with Python, ``pip``, and a terminal? You'll move faster with
the regular :doc:`Quick Start guide <getting_started>` instead.

What you'll do
---------------

1. Set up a terminal on your computer (Mac or Windows).
2. Install Datamint (one command).
3. Get your API key and connect it.
4. Upload your images.
5. Upload segmentations alongside them.

A "terminal" is just a window where you type a line of text and press Enter,
instead of clicking buttons. It looks intimidating the first time, but every
command below can be copy-pasted exactly as written.

Step 1: Set up your computer
-----------------------------

.. tab-set::

.. tab-item:: Mac

1. Install Python: go to `python.org/downloads <https://www.python.org/downloads/>`_,
click the big **Download Python** button, open the downloaded file, and click
through the installer with the default options.
2. Open the **Terminal** app: press ``Cmd + Space``, type ``Terminal``, and press
Enter.

From here on, every command in this guide goes into that Terminal window.

.. tab-item:: Windows

Use **WSL** (Windows Subsystem for Linux). WSL gives you a small, reliable
Linux environment inside Windows, and it's what the rest of this guide
assumes.

1. Open the **Start Menu**, search for **PowerShell**, right-click it, and
choose **Run as administrator**.
2. In the blue window that opens, type:

.. code-block:: bash

wsl --install

3. Restart your computer when it asks you to.
4. After restarting, open the **Start Menu** again and search for **Ubuntu**.
Open it. The first time, it will ask you to create a username and password
for this Linux environment. Pick anything you'll remember, it does not
need to match your Windows login.

From here on, whenever this guide says "open a terminal," open **Ubuntu**
from the Start Menu. It behaves just like the Mac Terminal.

.. dropdown:: 'wsl' is not recognized / WSL won't install
:icon: tools

Your Windows version may be too old, or WSL may be disabled by your
organization's IT policy. See Microsoft's
`WSL install documentation <https://learn.microsoft.com/windows/wsl/install>`_,
or ask your IT administrator to enable it for you.

Step 2: Install Datamint
--------------------------

In your terminal (Mac Terminal, or Ubuntu on Windows), paste this and press Enter:

.. code-block:: bash

pip install -U datamint

Check it installed correctly:

.. code-block:: bash

python3 -c "import datamint; print(datamint.__version__)"

If that prints a version number (like ``2.18.4``), you're set.

.. dropdown:: 'python3' or 'pip' is not recognized / command not found
:icon: tools

- On Mac, make sure you finished the Python installer in Step 1, then close
and reopen the Terminal app.
- Some systems use ``python``/``pip`` instead of ``python3``/``pip3``,
try that instead.
- If ``pip`` alone doesn't work, try this command ``python3 -m pip install -U datamint``.

Step 3: Get and set your API key
------------------------------------

.. include:: setup_api_key.rst

Step 4: Upload your images
----------------------------

Find the folder on your computer that holds your images (DICOM, NIfTI, video, or
regular image files all work). Then run:

.. code-block:: bash

datamint upload /path/to/your/images

Replace ``/path/to/your/images`` with the real folder location. Typing a long path
by hand is error-prone, most terminal windows let you **drag the folder from your
file browser and drop it into the terminal**, and it fills in the path for you.

.. dropdown:: Dragging a folder into Ubuntu (WSL) pastes a Windows-style path
:icon: tools

Dragging a folder from Windows File Explorer into the Ubuntu terminal often
pastes something like ``C:\Users\you\Scans``. WSL needs it written as
``/mnt/c/Users/you/Scans`` instead (lowercase drive letter, forward slashes).
Rewrite the path this way before pressing Enter.

The command will list what it found and ask for confirmation before uploading
anything:

.. code-block:: console

(...)
Do you want to proceed with the upload? (y/n):

Type ``y`` and press Enter. When it finishes, log into the
`Datamint platform <https://app.datamint.io/>`_ and you should see your files.

By default, uploaded files land in your **inbox** for review before they're fully
published. A few optional flags help keep things organized:

.. code-block:: bash

# Group files under a named channel
datamint upload /path/to/your/images --channel "Knee MRIs"

# Add one or more tags
datamint upload /path/to/your/images --tag "baseline"

# Skip the inbox and publish immediately
datamint upload /path/to/your/images --publish

You can combine all three in a single command.

Step 5: Upload segmentations along with your images
------------------------------------------------------

A segmentation is a mask file that highlights a region on an image (for example,
the outline of a bone or a tumor). If you have segmentation files that match your
images, upload both together with ``--segmentation_path``:

.. code-block:: bash

datamint upload /path/to/your/images --segmentation_path /path/to/your/segmentations --publish

The two folders need to mirror each other. If an image is in a subfolder, its
matching segmentation file should be in the same subfolder position under the
segmentations folder.

After the upload, you'll see a short summary confirming how many images got a
matching segmentation:

.. code-block:: console

(...)
Number of images with an associated segmentation: 4 (100%)
Number of segmentations with associated name: 4 (100%)

.. tip::

Need to name your segmentation labels, or link them to an AI model? Those are
covered in the full :doc:`Command-line tools <command_line_tools>` reference,
under ``--segmentation_names`` and ``--ai-model``.

Troubleshooting
-----------------

.. dropdown:: API authentication errors
:icon: tools

Verify your API key is set correctly:

.. code-block:: bash

datamint config

.. dropdown:: SSL certificate errors
:icon: tools

See the dedicated :doc:`SSL Troubleshooting guide <ssl_troubleshooting>`.

.. dropdown:: My upload finished, but I don't see my files
:icon: tools

By default, files land in your **inbox**, not the main published list. Check
the inbox on the `Datamint platform <https://app.datamint.io/>`_, or re-run
your upload with ``--publish`` to skip the inbox next time.

What's next
-------------

Once uploading feels comfortable, the full :doc:`Command-line tools <command_line_tools>`
reference covers every option available, including advanced segmentation labeling
and associating uploads with a deployed AI model.
30 changes: 27 additions & 3 deletions docs/source/index.rst
Original file line number Diff line number Diff line change
Expand Up @@ -27,8 +27,8 @@ Datamint
- :bdg-warning:`Intermediate` assumes SDK familiarity, introduces ML/dataset concepts
- :bdg-danger:`Advanced` full training pipelines, custom models, 3D data, multi-step workflows

User Guide
----------
For Developers
--------------

.. grid:: 1 2 3 3
:gutter: 2
Expand Down Expand Up @@ -95,7 +95,7 @@ User Guide

.. toctree::
:maxdepth: 2
:caption: User Guide
:caption: For Developers
:hidden:

getting_started
Expand All @@ -107,6 +107,30 @@ User Guide
datamint_vs_raw_pytorch
ssl_troubleshooting

For Non-Developers
-------------------

Just need to upload images and segmentations, no coding? Start here.

.. grid:: 1 2 3 3
:gutter: 2

.. grid-item-card:: Upload Images & Segmentations
:link: for_non_devs
:link-type: doc

:bdg-success:`Beginner`

No coding required. Set up your computer, get an API key, and upload your files
from a terminal, step by step.

.. toctree::
:maxdepth: 2
:caption: For Non-Developers
:hidden:

for_non_devs

Quick Start
-----------

Expand Down
Loading