Melanoma-Detection

Skin Cancer Detection using custom CNN (Shallow CNN) in Tensorflow

Table of Contents

General Information

To build a CNN based model which can accurately detect melanoma. Melanoma is a type of cancer that can be deadly if not detected early. It accounts for 75% of skin cancer deaths. A solution that can evaluate images and alert dermatologists about the presence of melanoma has the potential to reduce a lot of manual effort needed in diagnosis.

Conclusions

  • Real world data is expected to have class imbalance issues i.e. some classes with very small sample sizes.
  • This can be a challenge in training high performance models (Accuracy and Generalization wise)
  • Using appropriate data augmentation techniques it is possible to improve the model performance.

Technologies Used

  • Python 3.8.13
  • Tensoflow 2.9.1
  • Kera 2.9.0
  • Augmentor 0.2.10
  • Numpy 1.23.1
  • Pandas 1.4.2
  • NVIDIA Driver 515.48.07 (GTX 1070)
  • CUDA 11.7

Acknowledgements

  • The dataset consists of 2357 images of malignant and benign oncological diseases, which were formed from the International Skin Imaging Collaboration (ISIC)

Contact

Created by [@jaskirat8] - feel free to contact me!