Convolutional neural network for reconstruction of 7T-like images from 3T MRI using appearance and anatomical features

Khosro Bahrami; Feng Shi; Islem Rekik; Dinggang Shen

doi:10.1007/978-3-319-46976-8_5

Convolutional neural network for reconstruction of 7T-like images from 3T MRI using appearance and anatomical features

Khosro Bahrami, Feng Shi, Islem Rekik, Dinggang Shen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

56 Citations (Scopus)

Abstract

The advanced 7 Tesla (7T) Magnetic Resonance Imaging (MRI) scanners provide images with higher resolution anatomy than 3T MRI scanners, thus facilitating early diagnosis of brain diseases. However, 7T MRI scanners are less accessible, compared to the 3T MRI scanners. This motivates us to reconstruct 7T-like images from 3T MRI. We propose a deep architecture for Convolutional Neural Network (CNN), which uses the appearance (intensity) and anatomical (labels of brain tissues) features as input to non-linearly map 3T MRI to 7T MRI. In the training step, we train the CNN by feeding it with both appearance and anatomical features of the 3T patch. This outputs the intensity of center voxel in the corresponding 7T patch. In the testing step, we apply the trained CNN to map each input 3T patch to the 7T-like image patch. Our performance is evaluated on 15 subjects, each with both 3T and 7T MR images. Both visual and numerical results show that our method outperforms the comparison methods.

Original language	English
Title of host publication	Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings
Publisher	Springer Verlag
Pages	39-47
Number of pages	9
Volume	10008 LNCS
ISBN (Print)	9783319469751
DOIs	https://doi.org/10.1007/978-3-319-46976-8_5
Publication status	Published - 2016
Externally published	Yes
Event	1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016 - Athens, Greece Duration: 2016 Oct 21 → 2016 Oct 21

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10008 LNCS
ISSN (Print)	03029743
ISSN (Electronic)	16113349

Other

Other	1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016
Country	Greece
City	Athens
Period	16/10/21 → 16/10/21

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-319-46976-8_5

Fingerprint Dive into the research topics of 'Convolutional neural network for reconstruction of 7T-like images from 3T MRI using appearance and anatomical features'. Together they form a unique fingerprint.

Cite this

Bahrami, K., Shi, F., Rekik, I., & Shen, D. (2016). Convolutional neural network for reconstruction of 7T-like images from 3T MRI using appearance and anatomical features. In Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings (Vol. 10008 LNCS, pp. 39-47). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10008 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-46976-8_5

Convolutional neural network for reconstruction of 7T-like images from 3T MRI using appearance and anatomical features. / Bahrami, Khosro; Shi, Feng; Rekik, Islem; Shen, Dinggang.

Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings. Vol. 10008 LNCS Springer Verlag, 2016. p. 39-47 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10008 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Bahrami, K, Shi, F, Rekik, I & Shen, D 2016, Convolutional neural network for reconstruction of 7T-like images from 3T MRI using appearance and anatomical features. in Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings. vol. 10008 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10008 LNCS, Springer Verlag, pp. 39-47, 1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016, Athens, Greece, 16/10/21. https://doi.org/10.1007/978-3-319-46976-8_5

Bahrami K, Shi F, Rekik I, Shen D. Convolutional neural network for reconstruction of 7T-like images from 3T MRI using appearance and anatomical features. In Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings. Vol. 10008 LNCS. Springer Verlag. 2016. p. 39-47. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-46976-8_5

Bahrami, Khosro ; Shi, Feng ; Rekik, Islem ; Shen, Dinggang. / Convolutional neural network for reconstruction of 7T-like images from 3T MRI using appearance and anatomical features. Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings. Vol. 10008 LNCS Springer Verlag, 2016. pp. 39-47 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{0aa3e7f387154749ba6548a8dc06d8e5,

title = "Convolutional neural network for reconstruction of 7T-like images from 3T MRI using appearance and anatomical features",

abstract = "The advanced 7 Tesla (7T) Magnetic Resonance Imaging (MRI) scanners provide images with higher resolution anatomy than 3T MRI scanners, thus facilitating early diagnosis of brain diseases. However, 7T MRI scanners are less accessible, compared to the 3T MRI scanners. This motivates us to reconstruct 7T-like images from 3T MRI. We propose a deep architecture for Convolutional Neural Network (CNN), which uses the appearance (intensity) and anatomical (labels of brain tissues) features as input to non-linearly map 3T MRI to 7T MRI. In the training step, we train the CNN by feeding it with both appearance and anatomical features of the 3T patch. This outputs the intensity of center voxel in the corresponding 7T patch. In the testing step, we apply the trained CNN to map each input 3T patch to the 7T-like image patch. Our performance is evaluated on 15 subjects, each with both 3T and 7T MR images. Both visual and numerical results show that our method outperforms the comparison methods.",

author = "Khosro Bahrami and Feng Shi and Islem Rekik and Dinggang Shen",

year = "2016",

doi = "10.1007/978-3-319-46976-8_5",

language = "English",

isbn = "9783319469751",

volume = "10008 LNCS",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "39--47",

booktitle = "Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings",

note = "1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016 ; Conference date: 21-10-2016 Through 21-10-2016",

}

TY - GEN

T1 - Convolutional neural network for reconstruction of 7T-like images from 3T MRI using appearance and anatomical features

AU - Bahrami, Khosro

AU - Shi, Feng

AU - Rekik, Islem

AU - Shen, Dinggang

PY - 2016

Y1 - 2016

N2 - The advanced 7 Tesla (7T) Magnetic Resonance Imaging (MRI) scanners provide images with higher resolution anatomy than 3T MRI scanners, thus facilitating early diagnosis of brain diseases. However, 7T MRI scanners are less accessible, compared to the 3T MRI scanners. This motivates us to reconstruct 7T-like images from 3T MRI. We propose a deep architecture for Convolutional Neural Network (CNN), which uses the appearance (intensity) and anatomical (labels of brain tissues) features as input to non-linearly map 3T MRI to 7T MRI. In the training step, we train the CNN by feeding it with both appearance and anatomical features of the 3T patch. This outputs the intensity of center voxel in the corresponding 7T patch. In the testing step, we apply the trained CNN to map each input 3T patch to the 7T-like image patch. Our performance is evaluated on 15 subjects, each with both 3T and 7T MR images. Both visual and numerical results show that our method outperforms the comparison methods.

AB - The advanced 7 Tesla (7T) Magnetic Resonance Imaging (MRI) scanners provide images with higher resolution anatomy than 3T MRI scanners, thus facilitating early diagnosis of brain diseases. However, 7T MRI scanners are less accessible, compared to the 3T MRI scanners. This motivates us to reconstruct 7T-like images from 3T MRI. We propose a deep architecture for Convolutional Neural Network (CNN), which uses the appearance (intensity) and anatomical (labels of brain tissues) features as input to non-linearly map 3T MRI to 7T MRI. In the training step, we train the CNN by feeding it with both appearance and anatomical features of the 3T patch. This outputs the intensity of center voxel in the corresponding 7T patch. In the testing step, we apply the trained CNN to map each input 3T patch to the 7T-like image patch. Our performance is evaluated on 15 subjects, each with both 3T and 7T MR images. Both visual and numerical results show that our method outperforms the comparison methods.

UR - http://www.scopus.com/inward/record.url?scp=84992512809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84992512809&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-46976-8_5

DO - 10.1007/978-3-319-46976-8_5

M3 - Conference contribution

AN - SCOPUS:84992512809

SN - 9783319469751

VL - 10008 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 39

EP - 47

BT - Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings

PB - Springer Verlag

T2 - 1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016

Y2 - 21 October 2016 through 21 October 2016

ER -