Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
/melPublic

Tools to help identify new and changing moles on the skin with the goal of early detection of melanoma skin cancer.

License

NotificationsYou must be signed in to change notification settings

aevri/mel

Repository files navigation

Work-in-progress tools to help identify new and changing moles on the skin withthe goal of early detection of melanoma skin cancer.

Tests

TheMayo Clinic says:

The first melanoma signs and symptoms often are:

  • A change in an existing mole
  • The development of a new pigmented or unusual-looking growth on your skin

TheUK National Health Service says:

The most common sign of melanoma is the appearance of a new mole or a changein an existing mole. This can happen anywhere on the body, but the back,legs, arms and face are most commonly affected.

To help you identify new and changing moles, the tools enable you to:

  1. Maintain a catalog of your moles
  2. Record a visual history of each mole

For the catalog to be useful, it must enable you to uniquely distinguish everymole from the other new and pre-existing moles. To this end you will be able toassign a name, identifying images and notes for each one.

For the visual history of each mole to be useful, it must enable you toidentify significant differences in moles over time. The tools will help you tonormalise the rotation and centering of your mole images so that they areeasier to compare.

Things you can use now

"Mel Zero" - no executables, just documentation of how to createand maintain a mole catalog manually using a smartphone, a laptop, and a USBmicroscope.

"Mel One" - the code in this repository provides WIP rough tools for maintaining a mole catalog using two complimentary techniques. Moles are assigned UUIDs so they map across techniques:

  • Roto-mapping: Detect new moles with themel rotomap * commands. Using the idea of 'photographing a rotating cylinder and cycling through the photographs' as a cheap way of presenting an interactive 3D model, treat parts of the body as if they were a cylinder and create photosets of them. Provide support for manually and automatically marking moles in the images and mapping known ones to a UUID using some basic deep learning. Also provide support for manually comparing moles, but this better done with microscope images.

  • Microscope imaging: Detect changes in moles with themel micro * commands. These help you quickly capture and manually compare images of known moles with a USB microscope.

License

Distributed under the Apache License (version 2.0); see theLICENSEfile at the top of the source tree for more information.

About

Tools to help identify new and changing moles on the skin with the goal of early detection of melanoma skin cancer.

Topics

Resources

License

Stars

Watchers

Forks


[8]ページ先頭

©2009-2025 Movatter.jp