Slicer 5.0: Summary, Highlights and Changelog

Table of Contents

Summary

The community of 3D Slicer developers is proud to announce that version 5.0 is now available for download. This version introduces hundreds of feature enhancements and bug fixes for better performance and stability. It includes many new-and-improved core modules and more than 40 new-and-improved extensions. Many of these features have been available for months to users of the Slicer Preview version, but now there is a release that developers and users can rely on for stability and maintenance.

3D Slicer 5.0 builds on the success of the version 4 with over one million downloads of the core program and 5.3 million downloads of extensions during the last decade.

The development of 3D Slicer—including its numerous modules, extensions, datasets, pull requests, patches, issues reports, suggestions—is made possible by users, developers, contributors and commercial partners around the world. 3D Slicer is based on a stack of open-source softwares and we are working constantly on updating the underlying packages. This development is funded by various grants and agencies. For more details, please see the 3D Slicer Acknowledgments page.

One of the main distinguishing features of 3D Slicer is its easy extensibility enabled by good quality documentation, a thriving community, cross-platform support & open source code. 3D Slicer supports a number of extension types which offer a different mix of simplicity and range of control of the software.
A rigorous quality assurance system leveraging automated testing and user feedback ensures the timely detection of issues caused by ongoing work in different components of the platform. See the contributing guidelines

slicer.org is the portal to the application, training materials, and the development community.

The Slicer Tutorials page provides a series of tutorials and data sets for training in the use of Slicer.

Please note that Slicer continues to be a research package and is not intended for clinical use (clinical users must obtain the necessary ethics or regulatory approvals).

Highlights

Markups improvements

  • New built-in markups types
    unnamed (1)

  • Support for computing markups measurements (length for line and curve, angle for angle markups, curvature mean and curvature max for curve markups and area for plane and curve markups)
    unnamed

  • New interaction handles for translating, rotating and scaling markups from 2D and 3D views.
    unnamed (2)

  • Pluggable markups infrastructure: The infrastructure has been generalized to enable developers add new markup types in extensions. For example, the SurfaceMarkups extension adds editable NURBS surfaces for modelling thin objects.

Artificial Intelligence support

MONAILabel extension

The MONAILabel extension builds on the very popular MONAI toolkit for building AI models for automatic segmentation of medical images. The extension allows users to segment images (using classic and AI-assisted Segment Editor effects) and uses these data sets to train an AI model. MONAILabel can be set up without a dedicated server, on any computer with a strong GPU.

This is the most popular framework for users who want to train their own AI models for image segmentation.

3D Slicer and MONAI Label were highlighted by the CEO of NVIDIA during his 2022 GCT Keynote Address as the suggested tools for working with DICOM data.

NVIDIA AI-assisted annotation extension

The new Slicer extension Nvidia AI-assisted annotation (AIAA) adds a new effect in the Segment Editor module.

Example result of automatic segmentation:

A community-supported AI segmentation server is provided for Slicer users to allow easy evaluation of the extension, without the need for users to have access to a computer with a strong GPU. The server hosts a number of trained models for segmentation of liver, lungs, brain tumor, etc. This extension is recommended quick evaluation and for users who find the already available pre-trained models suitable for their work. For creating new AI models, MONAILabel extension is recommended.

More AI tools deployed via 3D Slicer

Several other AI-assisted extensions are added to the Extensions Manager, including:

HD Brain Extraction Tool HD-BET skull stripping tool from DKFZ/Heidelberg
DensityLungSegmentation AI-based lung segmentation
RVXLiverSegmentation AI-based liver segmentation
SlicerAIGT Real-time ultrasound and optical image stream segmentation and volume reconstruction
SlicerPyTorch Helper extension for installing Pytorch, used by extensions that require PyTorch
TorchIO Graphical user interface for the TorchIO AI helper toolkit

DICOM improvements

Cloud/Web support

Extensions Manager improvements

We have implemented some long-awaited improvements in the Extensions Manager in Slicer-5.0.3. Most importantly extensions can be updated automatically and bookmarks can be used for convenient reinstallation of a list of extensions across different Slicer versions, computers, or users.

For the list of new features as well as usability, stability and performance improvements. See New feature: Automatic update of extensions

image

Translation (internationalization) support

Slicer was awarded a CZI Essential Open Source Software grant for translation of the application and extensions to French and develop infrastructure for allowing translation to any other languages. This is still a work in progress (expected to be complete in September 2023), but language selection and translation of a large part of the displayed text in the Slicer core application is already translatable.

Infrastructure improvements

Other improvements

  • Integration with the Python ecosystem. See details in the Scripting and Dependencies sections below

  • Library updates. See details in the Dependencies section below

  • More than 40 new extensions have been added. See complete list in the Extensions section below.

  • Although we have a common understanding about how to conduct operations in the Slicer community, consensus was reached and we now formally adopted a code of conduct.

7 Likes

Changelog

Core

Application

Extension

Scripting

  • Slicer now uses Python 3!
  • Slicer’s python is now binary-compatible with pypi packages
  • Users can install most packages using slicer.util.pip_install

Installation

Internationalization/Localization

IO

Rendering & Display

Registration

New Modules

Text

Sequences

Lights

Improved Modules

Markups

DICOM

Models

Segmentations

Infrastructure

Build System

  • Embrace C++11 for coding style
  • Remove last remnants of TCL support. See details in dependencies

Platform Support

  • Support for building on newer Mac OSX (10.15+) and newer Linux builds (Ubuntu 18.04+)

Dependencies

Extensions

Listed below are extensions added, removed or updated since the 4.10.2 release.

The Slicer extensions manager enables Slicer users to install more than 150 extensions written and contributed by their peers from around the world.

New

Updated

Referenced below are new features added to existing extensions. Note all the existing extensions have been maintained and updated to account for API and build environment changes.

Removed

List of archived extensions is documented at Slicer/ExtensionsIndex/ARCHIVE/README.md

10 Likes