Chapter title |
Estimating CT Image from MRI Data Using 3D Fully Convolutional Networks
|
---|---|
Chapter number | 18 |
Book title |
Deep Learning and Data Labeling for Medical Applications
|
Published in |
Lecture notes in computer science, September 2016
|
DOI | 10.1007/978-3-319-46976-8_18 |
Pubmed ID | |
Book ISBNs |
978-3-31-946975-1, 978-3-31-946976-8
|
Authors |
Dong Nie, Xiaohuan Cao, Yaozong Gao, Li Wang, Dinggang Shen, Nie, Dong, Cao, Xiaohuan, Gao, Yaozong, Wang, Li, Shen, Dinggang |
Abstract |
Computed tomography (CT) is critical for various clinical applications, e.g., radiotherapy treatment planning and also PET attenuation correction. However, CT exposes radiation during CT imaging, which may cause side effects to patients. Compared to CT, magnetic resonance imaging (MRI) is much safer and does not involve any radiation. Therefore, recently researchers are greatly motivated to estimate CT image from its corresponding MR image of the same subject for the case of radiotherapy planning. In this paper, we propose a 3D deep learning based method to address this challenging problem. Specifically, a 3D fully convolutional neural network (FCN) is adopted to learn an end-to-end nonlinear mapping from MR image to CT image. Compared to the conventional convolutional neural network (CNN), FCN generates structured output and can better preserve the neighborhood information in the predicted CT image. We have validated our method in a real pelvic CT/MRI dataset. Experimental results show that our method is accurate and robust for predicting CT image from MRI image, and also outperforms three state-of-the-art methods under comparison. In addition, the parameters, such as network depth and activation function, are extensively studied to give an insight for deep learning based regression tasks in our application. |
X Demographics
Geographical breakdown
Country | Count | As % |
---|---|---|
United States | 15 | 36% |
Australia | 5 | 12% |
Canada | 4 | 10% |
Germany | 2 | 5% |
Switzerland | 1 | 2% |
Brazil | 1 | 2% |
Russia | 1 | 2% |
Saudi Arabia | 1 | 2% |
Ireland | 1 | 2% |
Other | 1 | 2% |
Unknown | 10 | 24% |
Demographic breakdown
Type | Count | As % |
---|---|---|
Members of the public | 28 | 67% |
Scientists | 12 | 29% |
Practitioners (doctors, other healthcare professionals) | 1 | 2% |
Science communicators (journalists, bloggers, editors) | 1 | 2% |
Mendeley readers
Geographical breakdown
Country | Count | As % |
---|---|---|
Netherlands | 1 | <1% |
Unknown | 179 | 99% |
Demographic breakdown
Readers by professional status | Count | As % |
---|---|---|
Student > Ph. D. Student | 37 | 21% |
Student > Master | 32 | 18% |
Researcher | 26 | 14% |
Student > Bachelor | 17 | 9% |
Professor > Associate Professor | 5 | 3% |
Other | 14 | 8% |
Unknown | 49 | 27% |
Readers by discipline | Count | As % |
---|---|---|
Engineering | 31 | 17% |
Computer Science | 29 | 16% |
Medicine and Dentistry | 22 | 12% |
Physics and Astronomy | 10 | 6% |
Biochemistry, Genetics and Molecular Biology | 5 | 3% |
Other | 18 | 10% |
Unknown | 65 | 36% |