Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Dec;35(6):1623-1633.
doi: 10.1007/s10278-022-00660-5. Epub 2022 Jun 29.

AnatomySketch: An Extensible Open-Source Software Platform for Medical Image Analysis Algorithm Development

Mingrui Zhuang #  1 , Zhonghua Chen #  1   2 , Hongkai Wang  3   4 , Hong Tang  1 , Jiang He  1 , Bobo Qin  1 , Yuxin Yang  1 , Xiaoxian Jin  1 , Mengzhu Yu  1 , Baitao Jin  1 , Taijing Li  1 , Lauri Kettunen  2
Affiliations

AnatomySketch: An Extensible Open-Source Software Platform for Medical Image Analysis Algorithm Development

Mingrui Zhuang et al. J Digit Imaging. 2022 Dec.

Abstract

The development of medical image analysis algorithm is a complex process including the multiple sub-steps of model training, data visualization, human-computer interaction and graphical user interface (GUI) construction. To accelerate the development process, algorithm developers need a software tool to assist with all the sub-steps so that they can focus on the core function implementation. Especially, for the development of deep learning (DL) algorithms, a software tool supporting training data annotation and GUI construction is highly desired. In this work, we constructed AnatomySketch, an extensible open-source software platform with a friendly GUI and a flexible plugin interface for integrating user-developed algorithm modules. Through the plugin interface, algorithm developers can quickly create a GUI-based software prototype for clinical validation. AnatomySketch supports image annotation using the stylus and multi-touch screen. It also provides efficient tools to facilitate the collaboration between human experts and artificial intelligent (AI) algorithms. We demonstrate four exemplar applications including customized MRI image diagnosis, interactive lung lobe segmentation, human-AI collaborated spine disc segmentation and Annotation-by-iterative-Deep-Learning (AID) for DL model training. Using AnatomySketch, the gap between laboratory prototyping and clinical testing is bridged and the development of MIA algorithms is accelerated. The software is opened at https://github.com/DlutMedimgGroup/AnatomySketch-Software .

Keywords: Algorithm development; Deep learning; Image annotation; Medical image analysis; User interaction.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
a AnatomySketch interface. b The function module panel. c An example of the drop list of user-defined function modules, mostly segmentation methods in this case, including deep network models. d An example of calling the software GUI (the highlighted line of code) to visualize intermediate variables.
Fig. 2
Fig. 2
The architecture diagram of the software platform
Fig. 3
Fig. 3
Interactive data annotation. a The tablet mode layout supporting stylus sketching and multi-touch gestures. b Multiple annotation tools are provided by the software
Fig. 4
Fig. 4
Proofreading of the automatic segmentation result. The inaccurate boundary can be dragged towards the correct position (the dashed curve) using the stylus or the mouse
Fig. 5
Fig. 5
The extension module. a The workflow of the extension module. b An example of the configuration file. c The widget panel generated by AnatomySketch according to the configuration file of (b)
Fig. 6
Fig. 6
An example of user-defined extension module for ITSS analysis in MR images
Fig.7
Fig.7
An example of user-developed plugin module for lung lobe annotation in CT images
Fig. 8
Fig. 8
An example of 3D FFD proofreading. a The pink area in white contour is the automatic segmentation result of V-Net model. The red area and the white contour depict the under-segmented part. b Human expert proofreading result (the adjusted white contour) using the FFD tool
Fig. 9
Fig. 9
AID annotation results of two exemplar CT slices, showing that the retrained network yield more accurate results than the preliminary network. The ground truth comes from human expert manual labelling
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Yoo TS, Ackerman MJ, Lorensen WE, Schroeder W, Chalana V, Aylward S, Metaxas D, Whitaker R. Engineering and algorithm design for an image processing Api: a technical report on ITK–the Insight Toolkit. Stud Health Technol Inform. 2002;85:586–592. - PubMed
    1. McCormick M, Liu X, Jomier J, Marion C, Ibanez L. ITK: enabling reproducible research and open science. Frontiers in neuroinformatics. 2014;8:13. doi: 10.3389/fninf.2014.00013. - DOI - PMC - PubMed
    1. Schroeder, W.; Martin, K.; Lorensen, B. The Visualization Toolkit (4th ed.); Kitware: 2006.
    1. Klein S, Staring M, Murphy K, Viergever MA, Pluim JPW. elastix: A Toolbox for Intensity-Based Medical Image Registration. IEEE Transactions on Medical Imaging. 2010;29:196–205. doi: 10.1109/TMI.2009.2035616. - DOI - PubMed
    1. Shamonin D, Bron E, Lelieveldt B, Smits M, Klein S, Staring M. Fast Parallel Image Registration on CPU and GPU for Diagnostic Classification of Alzheimer's Disease. Frontiers in Neuroinformatics. 2014 doi: 10.3389/fninf.2013.00050. - DOI - PMC - PubMed

Publication types

LinkOut - more resources