Skip to content

Reads any type of medical image. DICOM, 4D, DWI, fMRI, etc.

License

Notifications You must be signed in to change notification settings

juanprietob/med-img-reader

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

36 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

med-img-reader

Read or write a 3D or 2D image with one or multiple components in a variety of formats. DICOM, NIFTI, NRRD, JPG, PNG, MHD Viewer example

Installation


npm install med-img-reader

Usage example

In NODE environment

Read a single image file


const MedImgReader = require('med-img-reader');

//Image with one or multiple components in any format
var inputImage = '/path/to/input{.nrrd,.nii.gz,.jpg,.png,.dcm}'
var outputImage = '/path/to/output{.nrrd,.nii.gz,.jpg,.png,.dcm}'


const medImgReader = new MedImgReader();
medImgReader.SetFilename(inputImage);
medImgReader.ReadImage();
var image = medImgReader.GetOutput();
console.log("Image:", image);


const medImgWriter = new MedImgReader();

medImgWriter.SetInput(image);
medImgWriter.SetFilename(outputImage);
medImgWriter.WriteImage();

This is an example of an object return by the reader, which is compatible with itk.js


image = {
    imageType: {
      dimension: 2,
      componentType: 'uint16_t',
      pixelType: 1,
      components: 1
    },
    name: 'Image',
    origin: [ 0, 0 ],
    spacing: [ 0.148489, 0.148489 ],
    direction: { rows: 2, columns: 2, data: [ 1, 0, 0, 1 ] },
    size: [ 256, 256 ],
    data: Uint16Array [...]
}

Convert the image to a Tensor from tensorflow tfjs


const tf = require('@tensorflow/tfjs-node');//Or tfjs in browser or tfjs-node-gpu if in linux

tf.tensor(
	Float32Array.from(image.data), 
	[...[...image.size].reverse(), image.imageType.components]
));

Read a DICOM series


var inputDirectory = '/path/to/series/directory'
var outputImage = 'out.nrrd';

const medImgReader = new MedImgReader();
medImgReader.SetDirectory(inputDirectory);
medImgReader.ReadDICOMDirectory();

var image = medImgReader.GetOutput();

const medImgWriter = new MedImgReader();
medImgWriter.SetInput(inputImage);
medImgWriter.SetFilename(outputImage);
medImgWriter.WriteImage();

In browser environment

React component

If you are going to use this in the browser, the build time will be long so be patient. This library is compiled using emscripten and it bundles a file system with the FS library.

Here is an example for a React component:


import React, { Component } from 'react'

const axios = require('axios');
const MedImgReader = require('med-img-reader');

export default class ExampleComponent extends Component {

  constructor(){
    super();

    this.state = {
      itkImage: {}
    }

    const self = this;
    var medImgReader = new MedImgReader();

    var filename = '/brain.png';

    axios({
      method: 'get',
      url: filename,
      responseType: 'blob'
    })
    .then(function(brain){
      var blob = brain.data;
      return blob.arrayBuffer()
      .then((arr)=>{
        medImgReader.WriteFile(filename, arr);//We add the file to the FS filesystem
        medImgReader.SetFilename(filename);//Set the file name 
        medImgReader.ReadImage();
        return medImgReader.GetOutput();
      })
      
    })
    .then((itkImage)=>{
      self.setState({...self.state, itkImage})
    })
  }

  render() {

    const {
      itkImage
    } = this.state;

    var copyImg = {...itkImage, data: []};

    return (
      <div className={styles.test}>
        {JSON.stringify(copyImg)}
      </div>
    )
  }
}

Display the image using vtk.js

Example is here react-med-img-viewer

About

Reads any type of medical image. DICOM, 4D, DWI, fMRI, etc.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published