Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ITKIT: Feasible CT Image Analysis based on SimpleITK and MMEngine

Yiqin Zhang, Meiling Chen

Abstract

CT images are widely used in clinical diagnosis and treatment, and their data have formed a de facto standard - DICOM. It is clear and easy to use, and can be efficiently utilized by data-driven analysis methods such as deep learning. In the past decade, many program frameworks for medical image analysis have emerged in the open-source community. ITKIT analyzed the characteristics of these frameworks and hopes to provide a better choice in terms of ease of use and configurability. ITKIT offers a complete pipeline from DICOM to 3D segmentation inference. Its basic practice only includes some essential steps, enabling users with relatively weak computing capabilities to quickly get started using the CLI according to the documentation. For advanced users, the OneDL-MMEngine framework provides a flexible model configuration and deployment entry. This paper conducted 12 typical experiments to verify that ITKIT can meet the needs of most basic scenarios.

ITKIT: Feasible CT Image Analysis based on SimpleITK and MMEngine

Abstract

CT images are widely used in clinical diagnosis and treatment, and their data have formed a de facto standard - DICOM. It is clear and easy to use, and can be efficiently utilized by data-driven analysis methods such as deep learning. In the past decade, many program frameworks for medical image analysis have emerged in the open-source community. ITKIT analyzed the characteristics of these frameworks and hopes to provide a better choice in terms of ease of use and configurability. ITKIT offers a complete pipeline from DICOM to 3D segmentation inference. Its basic practice only includes some essential steps, enabling users with relatively weak computing capabilities to quickly get started using the CLI according to the documentation. For advanced users, the OneDL-MMEngine framework provides a flexible model configuration and deployment entry. This paper conducted 12 typical experiments to verify that ITKIT can meet the needs of most basic scenarios.
Paper Structure (16 sections, 4 figures, 4 tables)

This paper contains 16 sections, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Data-Driven CT Image Segmentation
  3. Roles of CT Image Processing Softwares
  4. Comparison of Other Existing Projects with This Work
  5. Methods and Implementations
  6. Data Model
  7. Processing Toolchain
  8. OneDL-MMEngine-Based Deep Learning
  9. Task Definition
  10. Integrated Datasets
  11. Integrated Neural Networks
  12. Integrated Pipelines
  13. Inferencer and 3D Slicer Extension
  14. Results
  15. Discussion
  16. ...and 1 more sections

Figures (4)

  • Figure 1: Recommended Dataset Structure for ITKIT. The image and label folders contain the volumetric data and corresponding semantic masks, respectively. The meta.json file stores metadata for the entire dataset, while crop_meta.json contains metadata for any cropped or patched sub-volumes derived from the original data.
  • Figure 2: Scope of ITKIT's Functional Abstractions
  • Figure 3: ITKIT 3D Slicer Extension Screenshot. The left side displays the ITKIT extension, providing an interface for configuring the parameters required to perform segmentation inference. The right side shows the results sematic mask of spleen, generated using an ONNX model trained and converted by ITKIT. In this case, the segment engine runs in a Docker container under WSL on Windows.
  • Figure 5: ITKIT Visualization. The image is saved to both local files and TensorBoard during training.