Skip to content

HeshanDissanayake/SIIM_ISIC_melanoma_classification

Repository files navigation

Melanoma classification

problem description

Skin cancer is the most prevalent type of cancer. Melanoma, specifically, is responsible for 75% of skin cancer deaths, despite being the least common skin cancer. The American Cancer Society estimates over 100,000 new melanoma cases will be diagnosed in 2020. It's also expected that almost 7,000 people will die from the disease. As with other cancers, early and accurate detection—potentially aided by data science—can make treatment more effective.

Used methods

  • Convolutional neural nerwork.
  • Convolutional Auto Encoder.
  • Transfer learning.(EfficientNetB5).

Some results

Auto encoder results.

In the following images, it shows the results that was generated trough the Auto encoder. This auto encode was tested to reduce the dimentionality by generating a latent space.

[Left - Input] [Mid - Output] [Right - Latent space]

image image

Visualization of the Convolution layers.

image

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published