Our approach to tackle Kaggle Ultrasound Nerve Segmentation competition

More info on this Kaggle competition can be found on https://www.kaggle.com/c/ultrasound-nerve-segmentation.

This deep neural network achieved ~0.68 score on the leaderboard (rank: 76/923)

The architecture was inspired by U-Net: Convolutional Networks for Biomedical Image Segmentation. We used the code provided by Marco Jocic as a starting point.

---

Overview

Data

Provided data is processed by data.py script. This script just loads the images and saves them into NumPy binary format files .npy for faster loading later.

Prepare the data

In order to extract raw images and save them to .npy files, you should first prepare its structure. Make sure that raw dir is located in the root of this project. Also, the tree of raw dir must be like:

-raw
 |
 ---- train
 |    |
 |    ---- 1_1.tif
 |    |
 |    ---- …
 |
 ---- test
      |
      ---- 1.tif
      |
      ---- …

• Now run python data.py.

Running this script will create train and test images and save them to .npy files.

Pre-processing

The images are resized to 64 x 96. Data augmentations such as random rotation, horizontal flip and vertical flip are added at runtime by using ImageDataGenerator module. This is modified version of keras ImageDataGenerator https://www.kaggle.com/hexietufts/ultrasound-nerve-segmentation/easy-to-use-keras-imagedatagenerator/code, as the original version does not augment the masks.

Output images (masks) are scaled to [0, 1] interval.

Model

You can find it here

Train the model and generate masks for test images

• Run python train.py to train the model.

Check out train_predict() to modify the number of iterations (epochs), batch size, etc.

After this script finishes, in imgs_mask_test.npy masks for corresponding images in imgs_test.npy should be generated. I suggest you examine these masks for getting further insight of your model's performance.

Generate submission

• Run python submission.py to generate the submission file submission.csv for the generated masks.

Check out function submission() and run_length_enc()

Post-processing

• Run ultrasound_postprocessing.ipynb to process the generated submission.csv to remove masks with width less than 60 pixels.