Hierarchical pictorial structures for simultaneously localizing multiple organs in volumetric pre-scan CT

Albert Montillo, Qi Song, Bipul Das, Zhye Yin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Parsing volumetric computed tomography (CT) into 10 or more salient organs simultaneously is a challenging task with many applications such as personalized scan planning and dose reporting. In the clinic, pre-scan data can come in the form of very low dose volumes acquired just prior to the primary scan or from an existing primary scan. To localize organs in such diverse data, we propose a new learning based framework that we call hierarchical pictorial structures (HPS) which builds multiple levels of models in a tree-like hierarchy that mirrors the natural decomposition of human anatomy from gross structures to finer structures. Each node of our hierarchical model learns (1) the local appearance and shape of structures, and (2) a generative global model that learns probabilistic, structural arrangement. Our main contribution is twofold. First we embed the pictorial structures approach in a hierarchical framework which reduces test time image interpretation and allows for the incorporation of additional geometric constraints that robustly guide model fitting in the presence of noise. Second we guide our HPS framework with the probabilistic cost maps extracted using random decision forests using volumetric 3D HOG features which makes our model fast to train and fast to apply to novel test data and posses a high degree of invariance to shape distortion and imaging artifacts. All steps require approximate 3 mins to compute and all organs are located with suitably high accuracy for our clinical applications such as personalized scan planning for radiation dose reduction. We assess our method using a database of volumetric CT scans from 81 subjects with widely varying age and pathology and with simulated ultra-low dose cadaver pre-scan data.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2015
Subtitle of host publicationImage Processing
PublisherSPIE
Volume9413
ISBN (Electronic)9781628415032
DOIs
StatePublished - Jan 1 2015
EventMedical Imaging 2015: Image Processing - Orlando, United States
Duration: Feb 24 2015Feb 26 2015

Other

OtherMedical Imaging 2015: Image Processing
CountryUnited States
CityOrlando
Period2/24/152/26/15

Fingerprint

Cone-Beam Computed Tomography
organs
Tomography
tomography
Statistical Models
dosage
Cadaver
Artifacts
Noise
Anatomy
Learning
Databases
planning
Radiation
Pathology
Costs and Cost Analysis
Planning
anatomy
pathology
Invariance

Keywords

  • 3D HOG image descriptor
  • hierarchical pictorial structures
  • probabilistic decision forest

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Montillo, A., Song, Q., Das, B., & Yin, Z. (2015). Hierarchical pictorial structures for simultaneously localizing multiple organs in volumetric pre-scan CT. In Medical Imaging 2015: Image Processing (Vol. 9413). [94130T] SPIE. https://doi.org/10.1117/12.2082183

Hierarchical pictorial structures for simultaneously localizing multiple organs in volumetric pre-scan CT. / Montillo, Albert; Song, Qi; Das, Bipul; Yin, Zhye.

Medical Imaging 2015: Image Processing. Vol. 9413 SPIE, 2015. 94130T.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Montillo, A, Song, Q, Das, B & Yin, Z 2015, Hierarchical pictorial structures for simultaneously localizing multiple organs in volumetric pre-scan CT. in Medical Imaging 2015: Image Processing. vol. 9413, 94130T, SPIE, Medical Imaging 2015: Image Processing, Orlando, United States, 2/24/15. https://doi.org/10.1117/12.2082183
Montillo, Albert ; Song, Qi ; Das, Bipul ; Yin, Zhye. / Hierarchical pictorial structures for simultaneously localizing multiple organs in volumetric pre-scan CT. Medical Imaging 2015: Image Processing. Vol. 9413 SPIE, 2015.
@inproceedings{b856ac50258c466392d305fe98bd57a1,
title = "Hierarchical pictorial structures for simultaneously localizing multiple organs in volumetric pre-scan CT",
abstract = "Parsing volumetric computed tomography (CT) into 10 or more salient organs simultaneously is a challenging task with many applications such as personalized scan planning and dose reporting. In the clinic, pre-scan data can come in the form of very low dose volumes acquired just prior to the primary scan or from an existing primary scan. To localize organs in such diverse data, we propose a new learning based framework that we call hierarchical pictorial structures (HPS) which builds multiple levels of models in a tree-like hierarchy that mirrors the natural decomposition of human anatomy from gross structures to finer structures. Each node of our hierarchical model learns (1) the local appearance and shape of structures, and (2) a generative global model that learns probabilistic, structural arrangement. Our main contribution is twofold. First we embed the pictorial structures approach in a hierarchical framework which reduces test time image interpretation and allows for the incorporation of additional geometric constraints that robustly guide model fitting in the presence of noise. Second we guide our HPS framework with the probabilistic cost maps extracted using random decision forests using volumetric 3D HOG features which makes our model fast to train and fast to apply to novel test data and posses a high degree of invariance to shape distortion and imaging artifacts. All steps require approximate 3 mins to compute and all organs are located with suitably high accuracy for our clinical applications such as personalized scan planning for radiation dose reduction. We assess our method using a database of volumetric CT scans from 81 subjects with widely varying age and pathology and with simulated ultra-low dose cadaver pre-scan data.",
keywords = "3D HOG image descriptor, hierarchical pictorial structures, probabilistic decision forest",
author = "Albert Montillo and Qi Song and Bipul Das and Zhye Yin",
year = "2015",
month = "1",
day = "1",
doi = "10.1117/12.2082183",
language = "English (US)",
volume = "9413",
booktitle = "Medical Imaging 2015",
publisher = "SPIE",

}

TY - GEN

T1 - Hierarchical pictorial structures for simultaneously localizing multiple organs in volumetric pre-scan CT

AU - Montillo, Albert

AU - Song, Qi

AU - Das, Bipul

AU - Yin, Zhye

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Parsing volumetric computed tomography (CT) into 10 or more salient organs simultaneously is a challenging task with many applications such as personalized scan planning and dose reporting. In the clinic, pre-scan data can come in the form of very low dose volumes acquired just prior to the primary scan or from an existing primary scan. To localize organs in such diverse data, we propose a new learning based framework that we call hierarchical pictorial structures (HPS) which builds multiple levels of models in a tree-like hierarchy that mirrors the natural decomposition of human anatomy from gross structures to finer structures. Each node of our hierarchical model learns (1) the local appearance and shape of structures, and (2) a generative global model that learns probabilistic, structural arrangement. Our main contribution is twofold. First we embed the pictorial structures approach in a hierarchical framework which reduces test time image interpretation and allows for the incorporation of additional geometric constraints that robustly guide model fitting in the presence of noise. Second we guide our HPS framework with the probabilistic cost maps extracted using random decision forests using volumetric 3D HOG features which makes our model fast to train and fast to apply to novel test data and posses a high degree of invariance to shape distortion and imaging artifacts. All steps require approximate 3 mins to compute and all organs are located with suitably high accuracy for our clinical applications such as personalized scan planning for radiation dose reduction. We assess our method using a database of volumetric CT scans from 81 subjects with widely varying age and pathology and with simulated ultra-low dose cadaver pre-scan data.

AB - Parsing volumetric computed tomography (CT) into 10 or more salient organs simultaneously is a challenging task with many applications such as personalized scan planning and dose reporting. In the clinic, pre-scan data can come in the form of very low dose volumes acquired just prior to the primary scan or from an existing primary scan. To localize organs in such diverse data, we propose a new learning based framework that we call hierarchical pictorial structures (HPS) which builds multiple levels of models in a tree-like hierarchy that mirrors the natural decomposition of human anatomy from gross structures to finer structures. Each node of our hierarchical model learns (1) the local appearance and shape of structures, and (2) a generative global model that learns probabilistic, structural arrangement. Our main contribution is twofold. First we embed the pictorial structures approach in a hierarchical framework which reduces test time image interpretation and allows for the incorporation of additional geometric constraints that robustly guide model fitting in the presence of noise. Second we guide our HPS framework with the probabilistic cost maps extracted using random decision forests using volumetric 3D HOG features which makes our model fast to train and fast to apply to novel test data and posses a high degree of invariance to shape distortion and imaging artifacts. All steps require approximate 3 mins to compute and all organs are located with suitably high accuracy for our clinical applications such as personalized scan planning for radiation dose reduction. We assess our method using a database of volumetric CT scans from 81 subjects with widely varying age and pathology and with simulated ultra-low dose cadaver pre-scan data.

KW - 3D HOG image descriptor

KW - hierarchical pictorial structures

KW - probabilistic decision forest

UR - http://www.scopus.com/inward/record.url?scp=84929586494&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84929586494&partnerID=8YFLogxK

U2 - 10.1117/12.2082183

DO - 10.1117/12.2082183

M3 - Conference contribution

VL - 9413

BT - Medical Imaging 2015

PB - SPIE

ER -