Elsevier

Applied Soft Computing

Volume 46, September 2016, Pages 537-542
Applied Soft Computing

Prediction of relative position of CT slices using a computational intelligence system

https://doi.org/10.1016/j.asoc.2015.09.021Get rights and content

Abstract

One of the most common techniques in radiology is the computerized tomography (CT) scan. Automatically determining the relative position of a single CT slice within the human body can be very useful. It can allow for an efficient retrieval of slices from the same body region taken in other volume scans and provide useful information to the non-expert user. This work addresses the problem of determining which portion of the body is shown by a stack of axial CT image slices. To tackle this problem, this work proposes a computational intelligence system that combines semantics-based operators for Genetic Programming with a local search algorithm, coupling the exploration ability of the former with the exploitation ability of the latter. This allows the search process to quickly converge towards (near-)optimal solutions. Experimental results, using a large database of CT images, have confirmed the suitability of the proposed system for the prediction of the relative position of a CT slice. In particular, the new method achieves a median localization error of 3.4 cm on unseen data, outperforming standard Genetic Programming and other techniques that have been applied to the same dataset. In summary, this paper makes two contributions: (i) in the radiology domain, the proposed system outperforms current state-of-the-art techniques; (ii) from the computational intelligence perspective, the results show that including a local searcher in Geometric Semantic Genetic Programming can speed up convergence without degrading test performance.

Section snippets

Background

Scanning large parts of a patient's body with computerized tomography (CT) is common practice in radiology. As reported in [1], the amount of image data resulting from a full body scan varies between 40 MB to more than 1 GB, which has to be stored in a medical picture archiving and communication system (PACS). The increasing amount of data poses various problems for physicians and the PACS. A clinician often needs to compare different scans of the same body region for differential diagnoses or

Method

The field of evolutionary computation is devoted to the development of search and optimization algorithms based on the core principles of Neo-Darwinian evolutionary theory [11]. Evolutionary algorithms are population-based meta-heuristics, where candidate solutions are stochastically selected and modified to produce new, and possibly better, solutions for a particular problem. In particular, in standard GP each individual is encoded using a tree structure, also known as a program tree, which

Data set information

Each CT slice is described by a compound radial image descriptor, which is generated using the following steps: unifying the image resolutions, extracting the patient's body and combining the two image descriptors to a single radial descriptor. The complete process used to build the dataset is explained with rigorous detail in [21], but can be summarized as follows. The data was retrieved from 53,500 CT images taken from 74 different patients (43 male, 31 female). Each CT slice is described by

Conclusions

This paper proposes a computational intelligence system to automatically determine the relative position of a single CT slice within a full body scan. Knowing the relative position in a scan allows the efficient retrieval of similar slices from the same body region in other volume scans. Moreover, the relative position is often important information for a non-expert user that only has access to a single CT slice of a scan.

The proposed system is based on a variant of GP. In particular, the GP

Acknowledgments

The authors acknowledge project MassGP (PTDC/EEI-CTP/2975/2012), FCT, Portugal.

References (26)

  • J. Feulner et al.

    Comparing axial ct slices in quantized n-dimensional surf descriptor space to estimate the visible body region

    Comput. Med. Imaging Graph.

    (2011)
  • T. Emrich et al.

    CT slice localization via instance-based regression

  • DICOM standard, 2015,...
  • J.R. Koza

    Human-competitive results produced by genetic programming

    Genet. Programm. Evol. Mach.

    (2010)
  • M.O. Güld et al.

    Quality of DICOM header information for image categorization

  • B. Haas et al.

    Automatic segmentation of thoracic and pelvic CT images for radiotherapy planning using implicit anatomic knowledge and organ-specific segmentation strategies

    Phys. Med. Biol.

    (2008)
  • S. Seifert et al.

    Hierarchical parsing and semantic navigation of full body CT data

  • C. Bürger

    Automatic Localisation of Body Regions in CT Topograms

    (2008)
  • L. Yang et al.

    A boosting framework for visuality-preserving distance metric learning and its application to medical image retrieval

    IEEE Trans. Pattern Anal. Mach. Intell.

    (2010)
  • J. Duncan et al.

    Medical image analysis: progress over two decades and the challenges ahead

    IEEE Trans. Pattern Anal. Mach. Intell.

    (2000)
  • J.H. Holland

    Adaptation in Natural and Artificial Systems

    (1992)
  • J.R. Koza

    Genetic Programming: On the Programming of Computers by Means of Natural Selection

    (1992)
  • L. Vanneschi et al.

    A survey of semantic methods in genetic programming

    Genet. Programm. Evol. Mach.

    (2014)
  • Cited by (0)

    View full text