Skip to main content

Modular visualization and analysis dashboard creation for high-resolution microscopy images

Project description

fusion-tools

Modular visualization and analysis dashboard creation for high-resolution microscopy images.

Installation

$ pip install fusion-tools[interactive]

Usage

fusion-tools is intended to bring some of the functionality found in FUSION to developers working with whole slide images (WSIs) stored locally.

One such example would be the Visualization and SlideMap class:

from fusion_tools import Visualization
from fusion_tools.components import SlideMap

vis_session = Visualization(
    local_slides = [path_to_slide]
    components = [
        [
            SlideMap()
        ]
    ]
)

vis_session.start()

The Visualization class lets users construct custom layouts of different tools by passing a list containing rows, columns, and tabs. (e.g. [ [column in row 1], [ [ tab 1 in column 2 in row 1, tab 2 in column 2 in row 1] ], [ [ column 1 in row 2 ] ] ] ).

By passing a list of paths to locally-stored whole slide images (WSIs), fusion-tools automatically generates a LocalTileServer which is bundled in with the Visualization session to allow for high-resolution image viewing.

from fusion_tools import Visualization
from fusion_tools.components import SlideMap, OverlayOptions, PropertyViewer
from fusion_tools.utils.shapes import load_aperio


path_to_slide = '/path/to/wsi.svs'
path_to_annotations = '/path/to/aperio_annotations.xml'

annotations = load_aperio(path_to_annotations)

vis_session = Visualization(
    local_slides = [path_to_slide],
    local_annotations = [annotations],
    components = [
        [
            SlideMap(),
            [
                OverlayOptions(),
                PropertyViewer()
            ]
        ]
    ]
)

vis_session.start()

You can also access remote tile servers (either through DSATileServer or CustomTileServer) as well as annotations stored on a Digital Slide Archive instance.

from fusion_tools import Visualization
from fusion_tools.handler.dsa_handler import DSAHandler
from fusion_tools.components import SlideMap

# Grabbing first item from demo DSA instance
base_url = 'https://demo.kitware.com/histomicstk/api/v1'
item_id = '5bbdeed1e629140048d01bcb'

# Starting the DSAHandler to grab information:
dsa_handler = DSAHandler(
    girderApiUrl = base_url
)

# Checking how many annotations this item has:
#print('This item has the following annotations: ')
#print(dsa_handler.query_annotation_count(item=item_id).to_dict('records'))

vis_session = Visualization(
    tileservers = [dsa_handler.get_tile_server(item_id)],
    components = [
        [
            SlideMap()
        ]
    ]
)

vis_session.start()

You can also use some of segmentation components for adding labels and annotations to structures in the slide.

Creating annotations on top of structures

Applying labels to many structures at the same time

from fusion_tools import Visualization
from fusion_tools.handler.dsa_handler import DSAHandler
from fusion_tools.components import SlideMap, FeatureAnnotation, BulkLabels

# Grabbing first item from demo DSA instance
base_url = 'https://demo.kitware.com/histomicstk/api/v1'
item_id = '5bbdeed1e629140048d01bcb'

# Starting the DSAHandler to grab information:
dsa_handler = DSAHandler(
    girderApiUrl = base_url
)

# Checking how many annotations this item has:
#print('This item has the following annotations: ')
#print(dsa_handler.query_annotation_count(item=item_id).to_dict('records'))

vis_session = Visualization(
    tileservers = [dsa_handler.get_tile_server(item_id)],
    components = [
        [
            SlideMap(),
            [
                FeatureAnnotation(
                    storage_path = os.getcwd()+'\\tests\\Test_Annotations\\',
                    labels_format = 'json',
                    annotations_format = 'rgb'
                ),
                BulkLabels()
            ]
        ]
    ]
)

vis_session.start()

New in fusion-tools>2.0.0!

Now you can add multiple slides to a single visualization session and you can even view them side-by-side!

  • By default, components in the same row are linked, or they can interact with each other through callbacks. This can be updated using the "linkage" kwarg when initializing a Visualization session.
  • If two of the same types of components (e.g., two SlideMap components) are placed in the same row and "linkage" is set to "row", callbacks will not work. Beware!
from fusion_tools import Visualization
from fusion_tools.components import SlideMap, OverlayOptions, PropertyViewer, PropertyPlotter
from fusion_tools.handler.dsa_handler import DSAHandler

# Mixed types of slides and annotations
local_slide_list = ['slide1.tif','slide2.ome.tif','slide3.svs']
local_annotations_list = ['slide1_annotations.xml','slide2 annotations.json','annotations for slide3.h5ad']

dsa_handler = DSAHandler(
    girderApiUrl = 'http://example_dsa_address.com/api/v1'
)

dsa_items_list = [
    'item_uuid_1',
    'item_uuid_2'
]

dsa_tileservers = [dsa_handler.get_tile_server(i) for i in dsa_items_list]

# Setting linkage to "col" to enable side-by-side visualization
vis_sess = Visualization(
    local_slides = local_slide_list,
    local_annotations = local_annotations_list,
    tileservers = dsa_tileservers,
    linkage = 'col',
    components = [
        [
            [
                SlideMap(),
                OverlayOptions(),
                PropertyViewer(),
                PropertyPlotter()
            ],            
            [
                SlideMap(),
                OverlayOptions(),
                PropertyViewer(),
                PropertyPlotter()
            ]
        ]
    ],
    app_options={'port': 8050}
)

vis_sess.start()

Contributing

Open to contributions. Feel free to submit a PR or post feature requests in Issues

Open Projects:

  • Automated segmentation workflow for locally stored images (active-learning, SAM, etc.)
  • Monitoring long-running model training/other external processes
  • Import anndata spatial --omics dataset (UPDATE (v2.0.0): see utils.shapes.load_visium for example loading 10x Visium dataset)

License

fusion-tools was created by Samuel Border. It is licensed under the terms of the Apache 2.0 License

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

fusion_tools-3.3.0.tar.gz (208.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

fusion_tools-3.3.0-py3-none-any.whl (182.9 kB view details)

Uploaded Python 3

File details

Details for the file fusion_tools-3.3.0.tar.gz.

File metadata

  • Download URL: fusion_tools-3.3.0.tar.gz
  • Upload date:
  • Size: 208.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.11.3

File hashes

Hashes for fusion_tools-3.3.0.tar.gz
Algorithm Hash digest
SHA256 faa43d3e5750a3e953d3d48274886c1b54f082e00c0b52f8056530ce31be4e16
MD5 8d1fa80729197839ed11be6d68ce20cc
BLAKE2b-256 90b414c12680467fb62d41d900dc2390c7a396813a7af3c1c3844695a3fa2c9c

See more details on using hashes here.

File details

Details for the file fusion_tools-3.3.0-py3-none-any.whl.

File metadata

  • Download URL: fusion_tools-3.3.0-py3-none-any.whl
  • Upload date:
  • Size: 182.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.11.3

File hashes

Hashes for fusion_tools-3.3.0-py3-none-any.whl
Algorithm Hash digest
SHA256 51808c5af01772c50204875647fa94b4cced89db1101feb259841d5c9e0436d5
MD5 20cc9e87741f22a7a27da30481d9f095
BLAKE2b-256 a26c6835e1f2ba67e302b5ccbd4eecd0bf1a142978e9f4928067457aa49ac7aa

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page