Implementing a Grand Challenge for Segmenting Volumetric CT Data
ADSA02

The final report for this workshop is available at:
https://alert.northeastern.edu/transitioning-technology/adsa/final-reports-and-presentations/

This workshop was conducted to discuss the implementation of a grand challenge for segmenting objects of interest (OOI) from volumetric CT scans of baggage. OOIs are known items that are inserted into baggage along with objects that are normally packed into baggage. Segmentation means finding the voxels corresponding to the OOIs in the images that result from the volumetric CT scans. The OOIs along with contents in the baggage are designed to create scenarios that a segmentation algorithm would encounter from scans on state-of-the-art CT scanners used in security applications.

Segmentation and classification are the two steps that are usually found in algorithms that perform automated threat recognition (ATR). Only the segmentation step of ATR is of interest for this grand challenge.

The objectives of the workshop were to discuss the following aspects of executing the grand challenge:

  • CT segmentation grand challenge definition
  • Dataset creation
  • Participant identification
  • Entry criteria and funds allocation
  • Segmentation algorithm development and testing
  • Independent validation and testing of the segmentation algorithms
  • Demonstration of algorithms
  • Creation of final report

Workshop Outcomes

  • The grand challenge for segmenting OOIs from volumetric CT images should be performed.
  • A number of refinements to the grand challenge were suggested.
  • There are relevant precedents in the medical imaging and other communities that should be researched in order to follow their best practices.
  • A  precise specification for the grand challenge will lead to better results.
  • Collaboration among researchers increases the speed of technology development.
  • It is speculated that advanced reconstruction algorithms will have a bigger impact on the performance of CT-based explosives detection equipment compared to advances in segmentation. However, a prerequisite for developing reconstruction algorithms is having segmentation algorithms available in order to assess the impact of improved image quality on segmentation. Therefore, it may be necessary to complete the grand challenge for CT segmentation before implementing a grand challenge for reconstruction.
  • Working groups should be held, instead of workshops, in order to get the grand challenges initiated. The first working groups should discuss the problem cases for segmentation and the specifications for the grand challenges.
  • Third parties have begun to work on the problems described in the two workshops held to date. This is due, in part, to disclosing problem statements and putting datasets into the public domain.
  • Providing mentorship to the participants will enhance the value of the segmentation algorithms developed by the participants.