Import HoverNet and ProvGigaPath features with imageFeatureTCGA

imageFeatureTCGA

library(imageFeatureTCGA)
library(SummarizedExperiment)
library(dplyr)

Overview

imageFeatureTCGA provides convenient access to histopathology-derived data from TCGA through two complementary pipelines:

  • HoVerNet → cell segmentation and classification (Graham et al. 2019)
  • ProvGigaPath → slide- and tile-level embeddings (Xu et al. 2024)

These datasets can be imported directly into R as Bioconductor objects, facilitating downstream integration with TCGA omics and clinical data.

Installation

if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install("waldronlab/imageFeatureTCGA")

Data structure and technical details

The datasets accessible through imageFeatureTCGA originate from whole-slide histopathology images processed by deep learning pipelines. They are distributed as precomputed features to avoid the computational cost of running segmentation and embedding models locally.

HoVerNet outputs

HoVerNet provides nuclei segmentation and classification results at the single-cell level. Each detected nucleus is represented by:

  • spatial coordinates (x, y) in pixel units relative to the slide
  • predicted cell type labels
  • class probabilities
  • polygon contours describing nuclear boundaries (when available)

When imported as a SpatialExperiment or SpatialFeatureExperiment, the data are structured as follows:

  • columns represent individual nuclei
  • colData stores cell-level metadata (coordinates, cell types)
  • assays may contain quantitative features (e.g., probabilities)
  • metadata may include segmentation contours and image information

These objects enable spatial analyses and integration with other Bioconductor workflows for spatial transcriptomics and imaging data.

ProvGigaPath embeddings

ProvGigaPath is a foundation model trained on large-scale pathology image tiles that produces high-dimensional embeddings summarizing visual and morphological features.

Two levels of embeddings are provided:

Slide-level embeddings

Slide-level embeddings summarize the entire whole-slide image into a single feature vector.

  • one row per slide
  • embedding dimension corresponds to the encoder output size
  • suitable for slide-level prediction or clustering tasks

Tile-level embeddings

Tile-level embeddings provide localized representations of tissue regions.

Each tile entry includes:

  • spatial coordinates (tile_x, tile_y) corresponding to the tile position on the slide
  • a high-dimensional embedding vector
  • optional metadata describing tile extraction parameters

These embeddings enable spatial analyses of tissue heterogeneity and can be integrated with cell-level data from HoVerNet using complementary packages such as imageTCGAutils.

Available Data

Use the following function to download the catalog of available files:

getCatalog()
## Warning in file.rename(file, dest): cannot rename file
## '/tmp/RtmpoAhpI3/file219f693e65e5/api/records/17981132/files/store_cancerdatasci_catalog.tsv/content'
## to
## '/github/home/.cache/R/BiocFileCache/api/records/17981132/files/store_cancerdatasci_catalog.tsv/content',
## reason 'Invalid cross-device link'
## # A tibble: 54,253 × 25
##    pipeline format  filename    fullpath fnsansext tcga_barcode Case.ID TSS.Code
##    <chr>    <chr>   <chr>       <chr>    <chr>     <chr>        <chr>   <chr>   
##  1 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  2 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  3 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  4 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  5 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  6 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  7 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  8 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  9 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
## 10 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
## # ℹ 54,243 more rows
## # ℹ 17 more variables: File.ID <chr>, File.Name <chr>, Data.Category <chr>,
## #   Data.Type <chr>, Project.ID <chr>, Sample.ID <chr>, Sample.Type <chr>,
## #   Source.Site <chr>, Study.Name <chr>, BCR <chr>, city <chr>, state <chr>,
## #   country <chr>, bcr_patient_uuid <chr>, lat <dbl>, lon <dbl>, level <chr>

Formats

  • HoVerNet data is available in JSON, GeoJSON, thumb and H5AD formats.
  • ProvGigaPath data is available in CSV format.

Note that the thumb format refers to the png thumbnails of the whole-slide images.

HoVerNet data

getCatalog("hovernet")
## # A tibble: 33,177 × 25
##    pipeline format  filename    fullpath fnsansext tcga_barcode Case.ID TSS.Code
##    <chr>    <chr>   <chr>       <chr>    <chr>     <chr>        <chr>   <chr>   
##  1 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  2 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  3 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  4 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  5 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  6 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  7 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  8 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  9 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
## 10 hovernet geojson TCGA-02-00… hoverne… TCGA-02-… TCGA-02-000… TCGA-0… 02      
## # ℹ 33,167 more rows
## # ℹ 17 more variables: File.ID <chr>, File.Name <chr>, Data.Category <chr>,
## #   Data.Type <chr>, Project.ID <chr>, Sample.ID <chr>, Sample.Type <chr>,
## #   Source.Site <chr>, Study.Name <chr>, BCR <chr>, city <chr>, state <chr>,
## #   country <chr>, bcr_patient_uuid <chr>, lat <dbl>, lon <dbl>, level <chr>

ProvGigaPath data

getCatalog("provgigapath")
## # A tibble: 21,076 × 25
##    pipeline     format filename fullpath fnsansext tcga_barcode Case.ID TSS.Code
##    <chr>        <chr>  <chr>    <chr>    <chr>     <chr>        <chr>   <chr>   
##  1 provgigapath csv    TCGA-02… provgig… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  2 provgigapath csv    TCGA-02… provgig… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  3 provgigapath csv    TCGA-02… provgig… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  4 provgigapath csv    TCGA-02… provgig… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  5 provgigapath csv    TCGA-02… provgig… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  6 provgigapath csv    TCGA-02… provgig… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  7 provgigapath csv    TCGA-02… provgig… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  8 provgigapath csv    TCGA-02… provgig… TCGA-02-… TCGA-02-000… TCGA-0… 02      
##  9 provgigapath csv    TCGA-02… provgig… TCGA-02-… TCGA-02-000… TCGA-0… 02      
## 10 provgigapath csv    TCGA-02… provgig… TCGA-02-… TCGA-02-000… TCGA-0… 02      
## # ℹ 21,066 more rows
## # ℹ 17 more variables: File.ID <chr>, File.Name <chr>, Data.Category <chr>,
## #   Data.Type <chr>, Project.ID <chr>, Sample.ID <chr>, Sample.Type <chr>,
## #   Source.Site <chr>, Study.Name <chr>, BCR <chr>, city <chr>, state <chr>,
## #   country <chr>, bcr_patient_uuid <chr>, lat <dbl>, lon <dbl>, level <chr>

Importing HoVerNet data

You can import HoVerNet segmentation results as either a SpatialExperiment or SpatialFeatureExperiment. Here we selectively import a file based on its filename, but you can also filter by other metadata fields such as Project.ID, pipeline, format, etc.

hspe <- getCatalog("hovernet") |>
    dplyr::filter(
        filename == paste(
            "TCGA-VG-A8LO-01A-01-DX1",
            "B39A4D64-82A1-4A04-8AB6-918F3058B83B",
            "json",
            "gz",
            sep = "."
        )
    ) |>
    getFileURLs() |>
    HoverNet(outClass = "SpatialExperiment") |>
    import()
## Warning in file.rename(file, dest): cannot rename file
## '/tmp/RtmpoAhpI3/file219f1e2f85bf/waldronlab-image-features/hovernet/json/TCGA-VG-A8LO-01A-01-DX1.B39A4D64-82A1-4A04-8AB6-918F3058B83B.json.gz'
## to
## '/github/home/.cache/R/BiocFileCache/waldronlab-image-features/hovernet/json/TCGA-VG-A8LO-01A-01-DX1.B39A4D64-82A1-4A04-8AB6-918F3058B83B.json.gz',
## reason 'Invalid cross-device link'
hspe
## class: SpatialExperiment 
## dim: 0 67081 
## metadata(1): type_map
## assays(1): counts
## rownames: NULL
## rowData names(0):
## colnames: NULL
## colData names(10): cell_id x ... B sample_id
## reducedDimNames(0):
## mainExpName: NULL
## altExpNames(0):
## spatialCoords names(2) : x y
## imgData names(0):

Each cell is represented with:

  • x, y spatial coordinates
  • cell type and type probabilities
  • optional contours stored in metadata
colData(hspe)
## DataFrame with 67081 rows and 10 columns
##           cell_id         x         y      type type_prob       label         R
##       <character> <numeric> <numeric> <integer> <numeric> <character> <integer>
## 1               8   3894.64   15842.5         1 0.9696429      neopla       255
## 2               9   3920.24   15864.1         1 0.9077253      neopla       255
## 3              10   3857.38   15871.0         1 0.9575972      neopla       255
## 4              11   3892.01   15872.0         1 0.9658887      neopla       255
## 5              12   3943.50   15872.6         1 0.0384615      neopla       255
## ...           ...       ...       ...       ...       ...         ...       ...
## 67077       88901   86252.5   8284.12         3  0.768924      connec         0
## 67078       88902   86219.8   8291.11         2  0.993865      inflam         0
## 67079       88903   86052.2   8301.45         4  0.750000      necros       255
## 67080       88904   86189.5   8306.04         1  0.774737      neopla       255
## 67081       88905   85777.0   8381.20         2  0.996622      inflam         0
##               G         B   sample_id
##       <integer> <integer> <character>
## 1             0         0    sample01
## 2             0         0    sample01
## 3             0         0    sample01
## 4             0         0    sample01
## 5             0         0    sample01
## ...         ...       ...         ...
## 67077         0       255    sample01
## 67078       255         0    sample01
## 67079       255         0    sample01
## 67080         0         0    sample01
## 67081       255         0    sample01

Importing ProvGigaPath embeddings

Slide-level embeddings

ProvGigaPath embeddings summarize tile or slide-level image features. In this example, we import slide-level embeddings for a single file. Each row corresponds to a slide, with an embedding vector describing the image-derived features.

getCatalog("provgigapath") |>
    dplyr::filter(
        filename == paste(
            "TCGA-VG-A8LO-01A-01-DX1",
            "B39A4D64-82A1-4A04-8AB6-918F3058B83B",
            "csv",
            "gz",
            sep = "."
        ) &
        level == "slide_level"
    ) |>
    getFileURLs() |>
    ProvGiga() |>
    import()
## Warning in file.rename(file, dest): cannot rename file
## '/tmp/RtmpoAhpI3/file219f1f734cdb/waldronlab-image-features/provgigapath/slide_level/TCGA-VG-A8LO-01A-01-DX1.B39A4D64-82A1-4A04-8AB6-918F3058B83B.csv.gz'
## to
## '/github/home/.cache/R/BiocFileCache/waldronlab-image-features/provgigapath/slide_level/TCGA-VG-A8LO-01A-01-DX1.B39A4D64-82A1-4A04-8AB6-918F3058B83B.csv.gz',
## reason 'Invalid cross-device link'
## # A tibble: 1 × 771
##   slideName     tumorType fileName     V1    V2     V3     V4     V5    V6    V7
##   <chr>         <chr>     <chr>     <dbl> <dbl>  <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 TCGA-VG-A8LO… <NA>      TCGA-VG… -0.352 0.594 -0.403 -0.529 0.0304 0.206 -1.13
## # ℹ 761 more variables: V8 <dbl>, V9 <dbl>, V10 <dbl>, V11 <dbl>, V12 <dbl>,
## #   V13 <dbl>, V14 <dbl>, V15 <dbl>, V16 <dbl>, V17 <dbl>, V18 <dbl>,
## #   V19 <dbl>, V20 <dbl>, V21 <dbl>, V22 <dbl>, V23 <dbl>, V24 <dbl>,
## #   V25 <dbl>, V26 <dbl>, V27 <dbl>, V28 <dbl>, V29 <dbl>, V30 <dbl>,
## #   V31 <dbl>, V32 <dbl>, V33 <dbl>, V34 <dbl>, V35 <dbl>, V36 <dbl>,
## #   V37 <dbl>, V38 <dbl>, V39 <dbl>, V40 <dbl>, V41 <dbl>, V42 <dbl>,
## #   V43 <dbl>, V44 <dbl>, V45 <dbl>, V46 <dbl>, V47 <dbl>, V48 <dbl>, …

Tile-level embeddings

ProvGigaPath tile-level embeddings provide a more granular representation of image features at the tile level. Each row corresponds to a tile, with spatial coordinates (tile_x, tile_y) and an embedding vector describing the image-derived features for that tile. In this example, we filter the catalog to the tile-level file corresponding to the same slide as above.

getCatalog("provgigapath") |>
    dplyr::filter(
        filename == paste(
            "TCGA-VG-A8LO-01A-01-DX1",
            "B39A4D64-82A1-4A04-8AB6-918F3058B83B",
            "csv",
            "gz",
            sep = "."
        ) &
        level == "tile_level"
    ) |>
    getFileURLs() |>
    ProvGiga() |>
    import()
## Warning in file.rename(file, dest): cannot rename file
## '/tmp/RtmpoAhpI3/file219f63d88f3c/waldronlab-image-features/provgigapath/tile_level/TCGA-VG-A8LO-01A-01-DX1.B39A4D64-82A1-4A04-8AB6-918F3058B83B.csv.gz'
## to
## '/github/home/.cache/R/BiocFileCache/waldronlab-image-features/provgigapath/tile_level/TCGA-VG-A8LO-01A-01-DX1.B39A4D64-82A1-4A04-8AB6-918F3058B83B.csv.gz',
## reason 'Invalid cross-device link'
## # A tibble: 211 × 1,543
##    slide_name    tile_id tile_name tile_x tile_y      `0`    `1`     `2`     `3`
##    <chr>           <dbl> <chr>      <dbl>  <dbl>    <dbl>  <dbl>   <dbl>   <dbl>
##  1 TCGA-VG-A8LO…       0 02155x_1…   2155  18612  0.177   -1.42   0.787   0.0380
##  2 TCGA-VG-A8LO…       1 03180x_1…   3180  15540  0.433   -1.33   1.11   -0.382 
##  3 TCGA-VG-A8LO…       2 03180x_1…   3180  16564  0.886   -1.71  -0.0505  0.317 
##  4 TCGA-VG-A8LO…       3 03180x_1…   3180  17588  0.183   -1.31  -0.599  -0.138 
##  5 TCGA-VG-A8LO…       4 03180x_1…   3180  18612  0.631   -0.588  0.652   0.548 
##  6 TCGA-VG-A8LO…       5 03180x_1…   3180  19636  0.700   -1.51   0.909   0.380 
##  7 TCGA-VG-A8LO…       6 04204x_1…   4204  14516  0.821   -2.47   1.05   -0.411 
##  8 TCGA-VG-A8LO…       7 04204x_1…   4204  15540  0.251   -0.958  0.259  -0.186 
##  9 TCGA-VG-A8LO…       8 04204x_1…   4204  16564 -0.0607  -0.917 -0.168   0.727 
## 10 TCGA-VG-A8LO…       9 04204x_1…   4204  17588 -0.00672 -1.05   0.460   0.600 
## # ℹ 201 more rows
## # ℹ 1,534 more variables: `4` <dbl>, `5` <dbl>, `6` <dbl>, `7` <dbl>,
## #   `8` <dbl>, `9` <dbl>, `10` <dbl>, `11` <dbl>, `12` <dbl>, `13` <dbl>,
## #   `14` <dbl>, `15` <dbl>, `16` <dbl>, `17` <dbl>, `18` <dbl>, `19` <dbl>,
## #   `20` <dbl>, `21` <dbl>, `22` <dbl>, `23` <dbl>, `24` <dbl>, `25` <dbl>,
## #   `26` <dbl>, `27` <dbl>, `28` <dbl>, `29` <dbl>, `30` <dbl>, `31` <dbl>,
## #   `32` <dbl>, `33` <dbl>, `34` <dbl>, `35` <dbl>, `36` <dbl>, `37` <dbl>, …

Importing multiple ProvGigaPath files

One can also import multiple files at once. Here we filter the catalog to the first three slide-level files for the TCGA-GBM project, and import them as a ProvGigaList. Each element of the list corresponds to a slide, with the same structure as described above for slide-level embeddings. Note that the ProvGigaList constructor can also accept a vector of file paths or URLs. The import method for ProvGigaList will then import each file in the list and return either a single SummarizedExperiment or a list of SummarizedExperiment objects based on the diversity of the data levels in the input files. In this example, the catalog is filtered to a slide-level subset, so the output is a single SummarizedExperiment object with three columns corresponding to the three slides.

pgl <- getCatalog("provgigapath") |>
    dplyr::filter(level == "slide_level", Project.ID == "TCGA-GBM") |>
    dplyr::slice(1:3) |>
    getFileURLs() |>
    ProvGigaList() |>
    import()
## Warning in file.rename(file, dest): cannot rename file
## '/tmp/RtmpoAhpI3/file219f19a0fd9d/waldronlab-image-features/provgigapath/slide_level/TCGA-02-0001-01Z-00-DX1.83fce43e-42ac-4dcd-b156-2908e75f2e47.csv.gz'
## to
## '/github/home/.cache/R/BiocFileCache/waldronlab-image-features/provgigapath/slide_level/TCGA-02-0001-01Z-00-DX1.83fce43e-42ac-4dcd-b156-2908e75f2e47.csv.gz',
## reason 'Invalid cross-device link'
## Warning in file.rename(file, dest): cannot rename file
## '/tmp/RtmpoAhpI3/file219f19a0fd9d/waldronlab-image-features/provgigapath/slide_level/TCGA-02-0001-01Z-00-DX2.b521a862-280c-4251-ab54-5636f20605d0.csv.gz'
## to
## '/github/home/.cache/R/BiocFileCache/waldronlab-image-features/provgigapath/slide_level/TCGA-02-0001-01Z-00-DX2.b521a862-280c-4251-ab54-5636f20605d0.csv.gz',
## reason 'Invalid cross-device link'
## Warning in file.rename(file, dest): cannot rename file
## '/tmp/RtmpoAhpI3/file219f19a0fd9d/waldronlab-image-features/provgigapath/slide_level/TCGA-02-0001-01Z-00-DX3.2836ce55-491f-4d86-99b1-668946927af8.csv.gz'
## to
## '/github/home/.cache/R/BiocFileCache/waldronlab-image-features/provgigapath/slide_level/TCGA-02-0001-01Z-00-DX3.2836ce55-491f-4d86-99b1-668946927af8.csv.gz',
## reason 'Invalid cross-device link'
pgl
## class: SummarizedExperiment 
## dim: 768 3 
## metadata(5): slideName tumorType fileName patientIds sampleIds
## assays(1): embeddings
## rownames: NULL
## rowData names(0):
## colnames(3): TCGA-02-0001-01Z-00-DX1 TCGA-02-0001-01Z-00-DX2
##   TCGA-02-0001-01Z-00-DX3
## colData names(0):

Mixed level imports

The ProvGigaList constructor can also accept a mix of slide- and tile-level files. In this case, the import method will return a list of SummarizedExperiment objects, one for each data level. Here we filter the catalog to include both slide- and tile-level files for the same slide, and import them together.

pgl_mixed <- getCatalog("provgigapath") |>
    dplyr::filter(
        filename %in% c(
            paste(
                "TCGA-VG-A8LO-01A-01-DX1",
                "B39A4D64-82A1-4A04-8AB6-918F3058B83B",
                "csv",
                "gz",
                sep = "."
            )
        ) &
        level %in% c("slide_level", "tile_level")
    ) |>
    getFileURLs() |>
    ProvGigaList() |>
    import()
## Warning in .as(from): NAs introduced by coercion
pgl_mixed
## $slide_level
## class: SummarizedExperiment 
## dim: 768 1 
## metadata(5): slideName tumorType fileName patientIds sampleIds
## assays(1): embeddings
## rownames: NULL
## rowData names(0):
## colnames(1): TCGA-VG-A8LO-01A-01-DX1
## colData names(0):
## 
## $tile_level
## class: SummarizedExperiment 
## dim: 1538 1 
## metadata(3): metadata sampleIds patientIds
## assays(1): tiles
## rownames: NULL
## rowData names(0):
## colnames(1): TCGA-VG-A8LO-01A-01-DX1
## colData names(0):

See also

You can explore the full documentation through the MOFA and Point Pattern Analysis vignettes in the imageTCGA manuscript repository.

Note. More vignettes will be added as new feature types and workflows become available.

Shiny App: imageTCGA

The imageTCGA Shiny application provides an interactive interface for exploring TCGA Diagnostic Image Database metadata.

Click here to explore the shiny app: imageTCGA

Session Info

Click here for Session Info 
sessionInfo()
## R version 4.6.0 (2026-04-24)
## Platform: x86_64-pc-linux-gnu
## Running under: Ubuntu 24.04.4 LTS
## 
## Matrix products: default
## BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3 
## LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.26.so;  LAPACK version 3.12.0
## 
## locale:
##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
##  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
##  [9] LC_ADDRESS=C               LC_TELEPHONE=C            
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       
## 
## time zone: Etc/UTC
## tzcode source: system (glibc)
## 
## attached base packages:
## [1] stats4    stats     graphics  grDevices utils     datasets  methods  
## [8] base     
## 
## other attached packages:
##  [1] dplyr_1.2.1                 SummarizedExperiment_1.42.0
##  [3] Biobase_2.72.0              GenomicRanges_1.64.0       
##  [5] Seqinfo_1.2.0               IRanges_2.46.0             
##  [7] S4Vectors_0.50.0            BiocGenerics_0.58.0        
##  [9] generics_0.1.4              MatrixGenerics_1.24.0      
## [11] matrixStats_1.5.0           imageFeatureTCGA_1.0.0     
## [13] BiocStyle_2.40.0           
## 
## loaded via a namespace (and not attached):
##  [1] tidyselect_1.2.1            blob_1.3.0                 
##  [3] bitops_1.0-9                Biostrings_2.80.0          
##  [5] filelock_1.0.3              BumpyMatrix_1.20.0         
##  [7] RCurl_1.98-1.18             fastmap_1.2.0              
##  [9] SingleCellExperiment_1.34.0 BiocFileCache_3.2.0        
## [11] GenomicAlignments_1.48.0    XML_3.99-0.23              
## [13] digest_0.6.39               lifecycle_1.0.5            
## [15] KEGGREST_1.52.0             RSQLite_2.4.6              
## [17] magrittr_2.0.5              compiler_4.6.0             
## [19] rlang_1.2.0                 sass_0.4.10                
## [21] tools_4.6.0                 utf8_1.2.6                 
## [23] yaml_2.3.12                 rtracklayer_1.72.0         
## [25] knitr_1.51                  S4Arrays_1.12.0            
## [27] bit_4.6.0                   curl_7.1.0                 
## [29] DelayedArray_0.38.1         xml2_1.5.2                 
## [31] BiocParallel_1.46.0         abind_1.4-8                
## [33] withr_3.0.2                 purrr_1.2.2                
## [35] sys_3.4.3                   grid_4.6.0                 
## [37] MultiAssayExperiment_1.38.0 cli_3.6.6                  
## [39] rmarkdown_2.31              crayon_1.5.3               
## [41] httr_1.4.8                  tzdb_0.5.0                 
## [43] rjson_0.2.23                BiocBaseUtils_1.14.0       
## [45] GenomicDataCommons_1.36.0   DBI_1.3.0                  
## [47] cachem_1.1.0                stringr_1.6.0              
## [49] rvest_1.0.5                 parallel_4.6.0             
## [51] AnnotationDbi_1.74.0        restfulr_0.0.16            
## [53] BiocManager_1.30.27         XVector_0.52.0             
## [55] vctrs_0.7.3                 Matrix_1.7-5               
## [57] jsonlite_2.0.0              hms_1.1.4                  
## [59] bit64_4.8.0                 TENxIO_1.14.0              
## [61] GenomicFeatures_1.64.0      maketools_1.3.2            
## [63] magick_2.9.1                jquerylib_0.1.4            
## [65] glue_1.8.1                  codetools_0.2-20           
## [67] stringi_1.8.7               GenomeInfoDb_1.48.0        
## [69] BiocIO_1.22.0               UCSC.utils_1.8.0           
## [71] tibble_3.3.1                pillar_1.11.1              
## [73] rappdirs_0.3.4              htmltools_0.5.9            
## [75] R6_2.6.1                    dbplyr_2.5.2               
## [77] httr2_1.2.2                 vroom_1.7.1                
## [79] evaluate_1.0.5              lattice_0.22-9             
## [81] readr_2.2.0                 cigarillo_1.2.0            
## [83] Rsamtools_2.28.0            png_0.1-9                  
## [85] SpatialExperiment_1.22.0    memoise_2.0.1              
## [87] bslib_0.10.0                rjsoncons_1.3.2            
## [89] Rcpp_1.1.1-1.1              SparseArray_1.12.2         
## [91] xfun_0.57                   TCGAutils_1.32.0           
## [93] buildtools_1.0.0            pkgconfig_2.0.3
Graham, Simon, Quoc Dang Vu, Shan E Ahmed Raza, et al. 2019. “Hover-Net: Simultaneous Segmentation and Classification of Nuclei in Multi-Tissue Histology Images.” Medical Image Analysis 58: 101563. https://doi.org/https://doi.org/10.1016/j.media.2019.101563.
Xu, Hanwen, Naoto Usuyama, Jaspreet Bagga, et al. 2024. “A Whole-Slide Foundation Model for Digital Pathology from Real-World Data.” Nature 630 (8015): 181–88. https://doi.org/10.1038/s41586-024-07441-w.