Estimating CT Image from MRI Data Using Structured Random Forest and Auto-Context Model

Alzheimer's Disease Neuroimaging Initiative

Research output: Contribution to journalArticle

90 Citations (Scopus)

Abstract

Computed tomography (CT) imaging is an essential tool in various clinical diagnoses and radiotherapy treatment planning. Since CT image intensities are directly related to positron emission tomography (PET) attenuation coefficients, they are indispensable for attenuation correction (AC) of the PET images. However, due to the relatively high dose of radiation exposure in CT scan, it is advised to limit the acquisition of CT images. In addition, in the new PET and magnetic resonance (MR) imaging scanner, only MR images are available, which are unfortunately not directly applicable to AC. These issues greatly motivate the development of methods for reliable estimate of CT image from its corresponding MR image of the same subject. In this paper, we propose a learning-based method to tackle this challenging problem. Specifically, we first partition a given MR image into a set of patches. Then, for each patch, we use the structured random forest to directly predict a CT patch as a structured output, where a new ensemble model is also used to ensure the robust prediction. Image features are innovatively crafted to achieve multi-level sensitivity, with spatial information integrated through only rigid-body alignment to help avoiding the error-prone inter-subject deformable registration. Moreover, we use an auto-context model to iteratively refine the prediction. Finally, we combine all of the predicted CT patches to obtain the final prediction for the given MR image. We demonstrate the efficacy of our method on two datasets: human brain and prostate images. Experimental results show that our method can accurately predict CT images in various scenarios, even for the images undergoing large shape variation, and also outperforms two state-of-the-art methods.

Original languageEnglish
Article number7169564
Pages (from-to)174-183
Number of pages10
JournalIEEE Transactions on Medical Imaging
Volume35
Issue number1
DOIs
Publication statusPublished - 2016 Jan 1

Fingerprint

Magnetic resonance imaging
Tomography
Magnetic resonance
Positron emission tomography
Magnetic Resonance Spectroscopy
Positron-Emission Tomography
Imaging techniques
Forests
Radiotherapy
Dosimetry
Prostate
Brain
Magnetic Resonance Imaging
Learning
Radiation
Planning

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Radiological and Ultrasound Technology
  • Software

Cite this

Estimating CT Image from MRI Data Using Structured Random Forest and Auto-Context Model. / Alzheimer's Disease Neuroimaging Initiative.

In: IEEE Transactions on Medical Imaging, Vol. 35, No. 1, 7169564, 01.01.2016, p. 174-183.

Research output: Contribution to journalArticle

Alzheimer's Disease Neuroimaging Initiative. / Estimating CT Image from MRI Data Using Structured Random Forest and Auto-Context Model. In: IEEE Transactions on Medical Imaging. 2016 ; Vol. 35, No. 1. pp. 174-183.
@article{c9715931f21f48cc8fc2243950b81cbf,
title = "Estimating CT Image from MRI Data Using Structured Random Forest and Auto-Context Model",
abstract = "Computed tomography (CT) imaging is an essential tool in various clinical diagnoses and radiotherapy treatment planning. Since CT image intensities are directly related to positron emission tomography (PET) attenuation coefficients, they are indispensable for attenuation correction (AC) of the PET images. However, due to the relatively high dose of radiation exposure in CT scan, it is advised to limit the acquisition of CT images. In addition, in the new PET and magnetic resonance (MR) imaging scanner, only MR images are available, which are unfortunately not directly applicable to AC. These issues greatly motivate the development of methods for reliable estimate of CT image from its corresponding MR image of the same subject. In this paper, we propose a learning-based method to tackle this challenging problem. Specifically, we first partition a given MR image into a set of patches. Then, for each patch, we use the structured random forest to directly predict a CT patch as a structured output, where a new ensemble model is also used to ensure the robust prediction. Image features are innovatively crafted to achieve multi-level sensitivity, with spatial information integrated through only rigid-body alignment to help avoiding the error-prone inter-subject deformable registration. Moreover, we use an auto-context model to iteratively refine the prediction. Finally, we combine all of the predicted CT patches to obtain the final prediction for the given MR image. We demonstrate the efficacy of our method on two datasets: human brain and prostate images. Experimental results show that our method can accurately predict CT images in various scenarios, even for the images undergoing large shape variation, and also outperforms two state-of-the-art methods.",
author = "{Alzheimer's Disease Neuroimaging Initiative} and Tri Huynh and Yaozong Gao and Jiayin Kang and Li Wang and Pei Zhang and Jun Lian and Dinggang Shen",
year = "2016",
month = "1",
day = "1",
doi = "10.1109/TMI.2015.2461533",
language = "English",
volume = "35",
pages = "174--183",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

TY - JOUR

T1 - Estimating CT Image from MRI Data Using Structured Random Forest and Auto-Context Model

AU - Alzheimer's Disease Neuroimaging Initiative

AU - Huynh, Tri

AU - Gao, Yaozong

AU - Kang, Jiayin

AU - Wang, Li

AU - Zhang, Pei

AU - Lian, Jun

AU - Shen, Dinggang

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Computed tomography (CT) imaging is an essential tool in various clinical diagnoses and radiotherapy treatment planning. Since CT image intensities are directly related to positron emission tomography (PET) attenuation coefficients, they are indispensable for attenuation correction (AC) of the PET images. However, due to the relatively high dose of radiation exposure in CT scan, it is advised to limit the acquisition of CT images. In addition, in the new PET and magnetic resonance (MR) imaging scanner, only MR images are available, which are unfortunately not directly applicable to AC. These issues greatly motivate the development of methods for reliable estimate of CT image from its corresponding MR image of the same subject. In this paper, we propose a learning-based method to tackle this challenging problem. Specifically, we first partition a given MR image into a set of patches. Then, for each patch, we use the structured random forest to directly predict a CT patch as a structured output, where a new ensemble model is also used to ensure the robust prediction. Image features are innovatively crafted to achieve multi-level sensitivity, with spatial information integrated through only rigid-body alignment to help avoiding the error-prone inter-subject deformable registration. Moreover, we use an auto-context model to iteratively refine the prediction. Finally, we combine all of the predicted CT patches to obtain the final prediction for the given MR image. We demonstrate the efficacy of our method on two datasets: human brain and prostate images. Experimental results show that our method can accurately predict CT images in various scenarios, even for the images undergoing large shape variation, and also outperforms two state-of-the-art methods.

AB - Computed tomography (CT) imaging is an essential tool in various clinical diagnoses and radiotherapy treatment planning. Since CT image intensities are directly related to positron emission tomography (PET) attenuation coefficients, they are indispensable for attenuation correction (AC) of the PET images. However, due to the relatively high dose of radiation exposure in CT scan, it is advised to limit the acquisition of CT images. In addition, in the new PET and magnetic resonance (MR) imaging scanner, only MR images are available, which are unfortunately not directly applicable to AC. These issues greatly motivate the development of methods for reliable estimate of CT image from its corresponding MR image of the same subject. In this paper, we propose a learning-based method to tackle this challenging problem. Specifically, we first partition a given MR image into a set of patches. Then, for each patch, we use the structured random forest to directly predict a CT patch as a structured output, where a new ensemble model is also used to ensure the robust prediction. Image features are innovatively crafted to achieve multi-level sensitivity, with spatial information integrated through only rigid-body alignment to help avoiding the error-prone inter-subject deformable registration. Moreover, we use an auto-context model to iteratively refine the prediction. Finally, we combine all of the predicted CT patches to obtain the final prediction for the given MR image. We demonstrate the efficacy of our method on two datasets: human brain and prostate images. Experimental results show that our method can accurately predict CT images in various scenarios, even for the images undergoing large shape variation, and also outperforms two state-of-the-art methods.

UR - http://www.scopus.com/inward/record.url?scp=84959335586&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959335586&partnerID=8YFLogxK

U2 - 10.1109/TMI.2015.2461533

DO - 10.1109/TMI.2015.2461533

M3 - Article

C2 - 26241970

AN - SCOPUS:84959335586

VL - 35

SP - 174

EP - 183

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 1

M1 - 7169564

ER -