Skip to main content

A CLI tool designed to help source data for skin lesion research.

Project description


Maintenance Build Status Coverage Status Codacy Badge License: MIT PyPI version PyPI pyversions Downloads

A Skin Lesion Acquisition (SLA) CLI tool designed to help source data for skin lesion research.


While working on an academic project in the domain of automatic skin lesion detection it became clear that there was no easy way to track down datasets cited highly in the literature

This is what motivated the creation of SDA-CLI.

SDA-CLI is targeted toward academic and medical researchers looking to source lesion dataset quickly to accelerate their research efforts.

Features at a Glance


  • Dataset summaries and label distribution.
  • Console-based dashboards.
  • Full support for downloading datasets + metadata from the ISIC Archive API.
  • Full support for downloading PH2 dataset.
  • Full support for downloading PAD-UFES-20 dataset.
  • Full support for downloading MEDNODE dataset.


  • Matplotlib integration for data distribution visualisation.
  • Dataset downloading (public datasets only).
  • Metadata extraction on applicable datasets.
  • Data background information sources and links.
  • Preprocessing of datasets for binary classification.

Datasets Available

The table below shows the dataset currently available to acquire via the tool.

Dataset Available
Altlas of Dermoscopy
BCN 20000
BCN 20000 Challenge
Brisbane ISIC Challenge 2020
Dermoscopedia (CC BY)
DermIS ⚠️
DermQuest ⚠️
ISIC 2020 Challenge MSKCC Contribution
ISIC 2020 Vienna Part 1
ISIC 2020 Vienna Part 2
JID Editorial Images 2018
MClass (Dermoscopy) ⚠️
MClass (Dermoscopy) ⚠️
PAD-UFES-20 ✅ ️
PH2 ✅ ️
Sydney MIA SMDC 2020 ISIC Challenge Contribution
  • ✅ - Fully implemented
  • ⚠️- Work in progress.
  • ❌ - Private dataset

Loading Configurations

There are three ways to load configuration files when using the CLI interface.

  1. The first method is using an explicit YAML configuration file like so:

    sla-cli -f/--config-file <FILE_PATH> <COMMAND> ...

  1. The second method is referencing the SLA_CLI_CONFIG_FILE environment variable. Once the variable is set, it will auto-reference the environment variables value, which should be a path to the configuration file you wish to load into the tool.

  1. The third and final method of loading a configuration with the tool is creating a ".sla_cli_config.yml" file in the directory you plan to run the tool in. This method is helpful if you wish to check-in your configuration to SCM.


The following sub sections discuss the how to use the tool.

The following conventions are used to describe tool usage.

<NAME>                 ---> Required argument.

[NAME: DEFAULT_VALUE]  ---> Optional argument showing default value. 

If unsure of how to use a command, use -h/--help on any command to get context on what commands are available and what they do.


The ls command is to gain quick insight into what data is available.

sla-cli ls [regex: '.*']              # Shows a list of dataset names available.
sla-cli ls -v totals [regex: '.*']    # Shows a list of dataset names and the number of images it contains.
sla-cli ls -v all [regex: '.*']       # Shows a list of dataset names and a full breakdown of all image label distribution

A sample of the sla-cli ls -v all output is shown below:

img.png Sample output of 'sla-cli ls -v all' command.

Project details

Download files

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

Source Distribution

sla-cli-0.2.1.tar.gz (39.2 kB view hashes)

Uploaded Source

Supported by

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