/bone-age-prediction

RSNA Bone Age Challenge 16Bit Solution

Primary LanguagePython

File Structure

.
+-- flow_dataframe.py	
+-- train.py
+-- rsna-bone-age      <-------------- downloaded dataset
|   +-- boneage-test-dataset.csv
|   +-- boneage-train-dataset.csv
|   +-- boneage-train-dataset
|   |   +-- boneage-train-dataset
|   |   |   +-- 1377.png
|   |   |   +-- 1378.png
|   |   |   +-- ...
|   +-- boneage-test-dataset
|   |   +-- boneage-test-dataset
|   |   |   +-- 4360.png
|   |   |   +-- 4361.png
|   |   |   +-- ...

Dataset

The dataset is release to RSNA Pediatric Bone Age Machine Learning Challenge. That is consisting of 14 236 hand radiographs (12 611 training set, 1425 validation set, 200 test set)

Model Structure (16Bit)

drawing

Prerequisites

Usage

  1. Clone this repository.

  2. Download images of 2017 RSNA Bone Age Challenge Dataset from this kaggle page and decompress them to the directory. Or download with kaggle-api

    kaggle datasets download -d kmader/rsna-bone-age

  3. Setting up your own parameters and run

    python train.py

to-do's

  • Adding test.
  • Add model's accurancy table.
  • Rewrite with pytorch.