Image Learner - HAM10000 1.0
Visit source
Download RO-Crate
Run on GalaxyAssociated Tutorial
This workflows is part of the tutorial GLEAM Image Learner - Validating Skin Lesion Classification on HAM10000, available in the GTN
Features
- Includes Galaxy Workflow Tests
- Includes a Galaxy Workflow Report
- Uses Galaxy Workflow Comments
Thanks to...
Workflow Author(s): Khai Dang, Paulo Cilas Morais Lyra Junior, Junhao Qiu, Jeremy Goecks
Tutorial Author(s): Khai Van Dang, Paulo Cilas Morais Lyra Junior, Junhao Qiu, Alyssa Pybus, Jeremy Goecks
Inputs
| ID | Name | Description | Type |
|---|---|---|---|
| selected_HAM10000_img_96_size.zip | #main/selected_HAM10000_img_96_size.zip | - Total dataset: 1,400 images - Generated 200 images per class - Resized all images to 96×96 pixels - Standardized format as PNG for consistent processing |
|
| selected_HAM10000_img_metadata_aug.csv | #main/selected_HAM10000_img_metadata_aug.csv | | Column | Description | | `lesion_id` | Lesion identifier used to group original and augmented images from the same lesion. | | `image_id` | Image identifier from the source dataset (shared by original and flipped versions). | | `dx` | Diagnosis label (target class). | | `dx_type` | Diagnosis confirmation method (for example, `histo`). | | `age` | Patient age in years. | | `sex` | Patient sex (`male`/`female`/`unknown`). | | `localization` | Anatomical site of the lesion. | | `image_path` | Image filename within the image ZIP. | |
|
Steps
| ID | Name | Description |
|---|---|---|
| 2 | Image Learner | toolshed.g2.bx.psu.edu/repos/goeckslab/image_learner/image_learner/0.1.5 |
Version History
master
Creators and Submitter
Views: 612 Downloads: 95 Runs: 1
Created: 2nd Feb 2026 at 13:25
AttributionsNone
Related items
Teams: Galaxy Training Network
Organizations: Galaxy
I'm a bot managed by @hexylena to upload workflows from the Galaxy Training Network to the WorkflowHub. If you have any issues with my behaviour please let her know by filing an issue.
Galaxy is an open, web-based platform for accessible, reproducible, and transparent computational biological research.
- Accessible: Users can easily run tools without writing code or using the CLI; all via a user-friendly web interface.
- Reproducible: Galaxy captures all the metadata from an analysis, making it completely reproducible.
- Transparent: Users share and publish analyses via interactive pages that can enhance analyses with user annotations.
- Scalable: Galaxy can run ...
Teams: Galaxy Training Network, usegalaxy-eu, Intergalactic Workflow Commission (IWC)
Web page: https://galaxyproject.org/
The Galaxy Training Network (GTN) is a collection of hands-on tutorials that are designed to be interactive and are built around Galaxy.
These tutorials can be used for learning and teaching how to use Galaxy for general data analysis, as well as a wide array of hands-on tutorials covering specific domains such as assembly, RNA-Seq analysis, deep learning, climate analysis, and more!
