Anatomical correlations for a hierarchical multi-atlas segmentation of CT images

55 %
45 %
Information about Anatomical correlations for a hierarchical multi-atlas segmentation of...
Technology

Published on May 16, 2014

Author: IIG_HES

Source: slideshare.net

Description

Many medical image analysis techniques require an initial localization and segmentation of anatomical structures. As part of the VISCERAL benchmarks on Anatomy segmentation, a hierarchical multi-atlas multi-structure segmentation approach guided by anatomical correlations is proposed. The method begins with a global alignment of the volumes and refines the alignment of the structures locally. The alignment of the bigger structures is used as reference for the smaller and harder to segment structures. The method is evaluated in the ISBI VISCERAL testset on ten anatomical structures in both contrast-enhanced and non-enhanced computed tomography scans. The proposed method obtained the highest DICE overlap score for some structures like kidneys and gallbladder. Similar segmentation accuracies compared to the highest results of the other methods proposed in the challenge are obtained for most of the other structures segmented with the method.

Anatomical correlations for a hierarchical multi-atlas segmentation of CT images Oscar A. Jiménez del Toro University of Applied Sciences Western Switzerland (HES-SO)

Overview •  Motivation •  VISCERAL •  Method •  Multi-atlas segmentation •  Image registration •  Hierarchical registration approach •  Experimental setup •  Results •  Conclusion 2

Overview •  Motivation •  VISCERAL •  Method •  Multi-atlas segmentation •  Image registration •  Hierarchical registration approach •  Experimental setup •  Results •  Conclusion 3

Motivation •  Anatomical segmentation is fundamental for further image analysis and Computer-Aided Diagnosis1 •  Manual annotation and visual inspection is time consuming for radiologists •  Accurate large scale data analysis techniques are needed 1 K.Doi. Current status and future potential of computer-aided diagnosis in medical imaging. British Journal of Radiology, 78:3-19, 2005. 4

VISCERAL Benchmarks •  Automatic segmentation of anatomical structures (20) – Visceral Benchmark 1: 12 ceCT test volumes*10 structures – ISBI challenge: 5 ceCT, 5 wbCT test volumes*10 structures •  CT and MR images (contrast-enhanced and non-enhanced)

Overview •  Motivation •  VISCERAL •  Method •  Multi-atlas segmentation •  Image registration •  Hierarchical registration approach •  Experimental setup •  Results •  Conclusion 6

Overview •  Motivation •  VISCERAL •  Method •  Multi-atlas segmentation •  Image registration •  Hierarchical registration approach •  Experimental setup •  Results •  Conclusion 7

Hierarchical multi-atlas segmentation •  Use multiple atlases for the estimation on a target image •  Global and local alignment •  Hierarchical selection of the registrations improves results2 •  Label fusion 2Jiménez del Toro et.al., Multi-structure Atlas-Based Segmentation using Anatomical Regions of Interest. In proceeding of: Medical Image Computing and Computer Assisted Intervention (MICCAI2013) MCV workshop, Nagoya, Japan, 2013

Image Registration •  Atlas = Patient volume + labels •  Coordinate transformation that increases spatial correlation between images •  Multi-scale gaussian pyramid

Affine alignment •  Global

Affine alignment •  Global

Affine alignment •  Global

Affine alignment •  Global •  Local refinement for independent structures

Affine alignment •  Global •  Local refinement for independent structures •  Regions of interest based on the morphologically dilated initial estimations

Affine alignment •  Global •  Local refinement for independent structures •  Regions of interest based on the morphologically dilated initial estimations

Affine alignment •  Global •  Local refinement for independent structures •  Regions of interest based on the morphologically dilated initial estimations

Right Kidney Liver Global alignment Urinary Bladder Right Lung Left Lung 1st Lumbar Vertebra Gall- bladder Left KidneyTrachea Spleen 2nd Local Affine Hierarchical Registration approach Affine Local Affine B-spline non- rigid

Non-rigid alignment •  Non-rigid •  B-spline •  Multi-scale approach •  Faster optimization due to better initial alignment

Label fusion •  Majority voting threshold •  Classification on a per-voxel basis •  Local registration errors are reduced •  Threshold optimization

Overview •  Motivation •  VISCERAL •  Method •  Multi-atlas segmentation •  Image registration •  Hierarchical registration approach •  Experimental setup •  Results •  Conclusion 20

Overview •  Motivation •  VISCERAL •  Method •  Multi-atlas segmentation •  Image registration •  Hierarchical registration approach •  Experimental setup •  Results •  Conclusion 21

Experimental setup •  VISCERAL ISBI testset •  5 contrast-enhanced CT volumes of the trunk •  5 non-enhanced whole body CT •  Applied to 10 anatomical structures: – Liver, lungs, kidneys, gallbladder, urinary bladder, 1st lumbar vertebra, trachea and spleen •  7 independent atlases as trainingset

Results ISBI Challenge Structure DICE ceCT DICE wbCT Liver 0.908 0.823 Right Kidney 0.905 0.649 Left Kidney 0.923 0.678 Right Lung 0.963 0.967 Left Lung 0.952 0.969 Spleen 0.859 0.677 Trachea 0.83 0.855 Gallbladder 0.4 0.271 Urinary bladder 0.68 0.616 1st Lumbar vertebra 0.472 0.44

Results ISBI Challenge Structure DICE ceCT DICE wbCT Liver 0.908 0.823 Right Kidney 0.905 0.649 Left Kidney 0.923 0.678 Right Lung 0.963 0.967 Left Lung 0.952 0.969 Spleen 0.859 0.677 Trachea 0.83 0.855 Gallbladder 0.4 0.271 Urinary bladder 0.68 0.616 1st Lumbar vertebra 0.472 0.44

Conclusion •  Straightforward and fully automatic method •  Showed robustness in the segmentation of multiple structures with high overlap for the bigger structures (e.g. kidneys, liver, lungs) •  Smaller structures fared well compared to the other approaches •  Future work: –  Extend to method to other modalities (CTwb ISBI challenge, MR) –  Improve speed of the algorithm

Questions???

Add a comment

Related presentations

Presentación que realice en el Evento Nacional de Gobierno Abierto, realizado los ...

In this presentation we will describe our experience developing with a highly dyna...

Presentation to the LITA Forum 7th November 2014 Albuquerque, NM

Un recorrido por los cambios que nos generará el wearabletech en el futuro

Um paralelo entre as novidades & mercado em Wearable Computing e Tecnologias Assis...

Microsoft finally joins the smartwatch and fitness tracker game by introducing the...

Related pages

Hierarchical Multi{atlas Based Segmentation for Anatomical ...

Hierarchical Multi{atlas Based Segmentation for ... enforce the spatial correlation of each anatomical ... of liver and spleen from CT images using ...
Read more

Hierarchic Anatomical Structure Segmentation Guided by ...

Hierarchic Anatomical Structure Segmentation Guided ... a hierarchical multi{atlas multi ... from CT images using normalized probabilistic atlases and ...
Read more

CiteSeerX — Hierarchical Multi–structure Segmentation ...

... a hierarchical multi–atlas multi–structure segmentation approach guided by anatomical correlations ... hierarchical multi atlas ...
Read more

Evaluation of a Hierarchical Anatomical Segmentation ...

Evaluation of a Hierarchical Anatomical Segmentation Approach in ... • CT and MR images ... increases spatial correlation between images3!!
Read more

Multi-atlas Segmentation in Head and Neck CT Scans

Multi-atlas Segmentation in Head and Neck CT Scans ... 3D anatomical struc-tures. Image segmentation is a well ... Our multi-atlas segmentation method ...
Read more

Anatomical correlations for a hierarchical multi-atlas ...

L’Institut Informatique de gestion, spécialisé dans le développement de systèmes d’information et étroitement lié à l’offre de formation ...
Read more

Multi-atlas based segmentation of brain images: Atlas ...

... accuracy of a multi-atlas segmentation of a query image based on ... using a hierarchical ... anatomical brain MRI segmentation combining ...
Read more

Hierarchical Multi{structure Segmentation Guided by ...

Hierarchical Multi{structure Segmentation Guided by ... a hierarchical multi{atlas multi ... Hierarchical Segmentation via Anatomical Correlations 33.
Read more