 View on TensorFlow.org
|
 Run in Google Colab
|
 View source on GitHub
|
 Download notebook
|
Overview
This tutorial shows how to use tfio.image.decode_dicom_image in TensorFlow IO to decode DICOM files with TensorFlow.
Setup and Usage
Download DICOM image
The DICOM image used in this tutorial is from the NIH Chest X-ray dataset.
The NIH Chest X-ray dataset consists of 100,000 de-identified images of chest x-rays in PNG format, provided by NIH Clinical Center and could be downloaded through this link.
Google Cloud also provides a DICOM version of the images, available in Cloud Storage.
In this tutorial, you will download a sample file of the dataset from the GitHub repo
- Xiaosong Wang, Yifan Peng, Le Lu, Zhiyong Lu, Mohammadhadi Bagheri, Ronald Summers, ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases, IEEE CVPR, pp. 3462-3471, 2017
curl -OL https://github.com/tensorflow/io/raw/master/docs/tutorials/dicom/dicom_00000001_000.dcmls -l dicom_00000001_000.dcm
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 164 0 164 0 0 602 0 --:--:-- --:--:-- --:--:-- 602
100 1024k 100 1024k 0 0 1679k 0 --:--:-- --:--:-- --:--:-- 12.0M
-rw-rw-r-- 1 kbuilder kokoro 1049332 Oct 27 16:24 dicom_00000001_000.dcm
Install required Packages, and restart runtime
try:
# Use the Colab's preinstalled TensorFlow 2.x
%tensorflow_version 2.x
except:
pass
pip install -q tensorflow-io
Decode DICOM image
import matplotlib.pyplot as plt
import numpy as np
import tensorflow as tf
import tensorflow_io as tfio
image_bytes = tf.io.read_file('dicom_00000001_000.dcm')
image = tfio.image.decode_dicom_image(image_bytes, dtype=tf.uint16)
skipped = tfio.image.decode_dicom_image(image_bytes, on_error='skip', dtype=tf.uint8)
lossy_image = tfio.image.decode_dicom_image(image_bytes, scale='auto', on_error='lossy', dtype=tf.uint8)
fig, axes = plt.subplots(1,2, figsize=(10,10))
axes[0].imshow(np.squeeze(image.numpy()), cmap='gray')
axes[0].set_title('image')
axes[1].imshow(np.squeeze(lossy_image.numpy()), cmap='gray')
axes[1].set_title('lossy image');
Text(0.5, 1.0, 'lossy image')
