With one in eight women (about 12%) in US being projected to develop invasive breast cancer in her lifetime, it is clearly a healthcare related challenge against human race. Detecting the presence and type of the tumour earlier is the key to save majority of life threatening situations from arising.
Using Convolutional Neural Network, which are highly suitable for applications like image recognition, can be used in determining the type of tumour based on its ultrasonic image. To explore and showcase how this technique can be used, I conducted a small experiment using dataset provided on this page. Dataset contains 250 ultrasonic grayscale images of tumours out of which 100 are of benign and 150 are malignant.
- Project Notebook: https://nbviewer.jupyter.org/github/patelatharva/Classify_Breast_Cancer_Type_From_Image/blob/master/tumor_type_detection.ipynb
- Blog post on Medium: https://towardsdatascience.com/how-to-fight-breast-cancer-with-deep-learning-ab28e42e4250?source=friends_link&sk=e8ebb63462826331b3b3263d8b8ae09e