usuyama / pydata-medical-image

A Deep Learning talk+tutorial for medical image processing

PyData Seattle 2017: Medical Image Analysis

This repository is for storing scripts/notebooks for "Medical image processing using Microsoft Deep Learning framework (CNTK)" (https://pydata.org/seattle2017/schedule/presentation/94/).

The presentation slide is available here.

The talk provides 2 examples:

• 2D-CNN: Diabetic Retinopathy Classification
• 3D-CNN: Lung Nodule Detection

Machine Environment

It is recommended to prepare a machine with a GPU if you wish to train the 2D/3D deep learning models in the examples. If you don't have one at hand, Azure Data Science VM with N-series VM is a great option. As of July 2017, you can choose from Tesla M60 and K80.

pydata-medical-image's Issues

get_class_weights(current_epoch)

Hello I was wondering if this is part of CNTK or is a separate class you have written ?

trained model weights

Hi,

could you please publish the trained model weights?

Thanks
Yael