Skip to content

Latest commit

 

History

History
39 lines (32 loc) · 1.04 KB

README.md

File metadata and controls

39 lines (32 loc) · 1.04 KB

Breast-Cancer-Detection-from-Ultrasound-pictures-using-CNN

CNN to classify ultrasounds of cancerous and non cancerous cells

Topic

Topic In this notebook I will be dealing with ultrasound pictures classification. The task is to be able to tell from scan pictures whether a tumor is benign or cancerous. I will be using a convolutional network to classify the images and will try to achieve an accuracy above 80%.

Objective

  • Detect the existance of breast cancer from ultrasound images

Summary

  • Importing Libraries
  • The Dataset
  • Creating the transforms and data loaders
  • Some Dataviz
  • Creating the classifier
  • Training and validation
  • Testing the model
  • Evaluation metrics
  • Trying out the model
  • Conclusion

Libraries

  • Pandas
  • Numpy
  • Sklearn
  • Torch vision
  • CV2
  • Glob

Evaluation metrics

  • Confusion matrix
  • Classification report
  • Accuracy

Data source

https://www.kaggle.com/datasets/vuppalaadithyasairam/ultrasound-breast-images-for-breast-cancer?fbclid=IwAR20PikUw-IiuE0hNEi3i_h79qaTtSfmqq-nJhVh_9L1op_DSO8BXuRfMpI