Back to Search
Start Over
A deep learning method for image-based subject-specific local SAR assessment
- Source :
- Magnetic Resonance in Medicine, 83(2), 695. John Wiley and Sons Inc., Magnetic Resonance in Medicine, 83(2), 695-711. Wiley, Magnetic Resonance in Medicine
- Publication Year :
- 2020
-
Abstract
- Purpose: Local specific absorption rate (SAR) cannot be measured and is usually evaluated by offline numerical simulations using generic body models that of course will differ from the patient's anatomy. An additional safety margin is needed to include this intersubject variability. In this work, we present a deep learning–based method for image-based subject-specific local SAR assessment. We propose to train a convolutional neural network to learn a “surrogate SAR model” to map the relation between subject-specific (Formula presented.) maps and the corresponding local SAR. Method: Our database of 23 subject-specific models with an 8–transmit channel body array for prostate imaging at 7 T was used to build 5750 training samples. These synthetic complex (Formula presented.) maps and local SAR distributions were used to train a conditional generative adversarial network. Extra penalization for local SAR underestimation errors was included in the loss function. In silico and in vivo validation were performed. Results: In silico cross-validation shows a good qualitative and quantitative match between predicted and ground-truth local SAR distributions. The peak local SAR estimation error distribution shows a mean overestimation error of 15% with 13% probability of underestimation. The higher accuracy of the proposed method allows the use of less conservative safety factors compared with standard procedures. In vivo validation shows that the method is applicable with realistic measurement data with impressively good qualitative and quantitative agreement to simulations. Conclusion: The proposed deep learning method allows online image-based subject-specific local SAR assessment. It greatly reduces the uncertainty in current state-of-the-art SAR assessment methods, reducing the time in the examination protocol by almost 25%.
- Subjects :
- Male
Databases, Factual
Computer science
convolutional neural network
Signal-To-Noise Ratio
Convolutional neural network
030218 nuclear medicine & medical imaging
Full Papers—Computer Processing and Modeling
0302 clinical medicine
ultrahigh‐field MRI
Image Processing, Computer-Assisted
subject‐specific SAR assessment
skin and connective tissue diseases
Full Paper
Phantoms, Imaging
Prostate
Specific absorption rate
SAR assessment
Magnetic Resonance Imaging
Healthy Volunteers
specific absorption rate
Radiology Nuclear Medicine and imaging
ultrahigh-field MRI
subject-specific SAR assessment
Algorithms
MRI
SAR
Relation (database)
Channel (digital image)
Image (mathematics)
03 medical and health sciences
Journal Article
Humans
Computer Simulation
Radiology, Nuclear Medicine and imaging
parallel transmit
Protocol (science)
Models, Statistical
business.industry
Deep learning
fungi
Prostatic Neoplasms
Reproducibility of Results
deep learning
Pattern recognition
Function (mathematics)
body regions
Neural Networks, Computer
Artificial intelligence
business
subject-specific
030217 neurology & neurosurgery
Subjects
Details
- Language :
- English
- ISSN :
- 07403194
- Database :
- OpenAIRE
- Journal :
- Magnetic Resonance in Medicine, 83(2), 695. John Wiley and Sons Inc., Magnetic Resonance in Medicine, 83(2), 695-711. Wiley, Magnetic Resonance in Medicine
- Accession number :
- edsair.doi.dedup.....d2d433778ab5221ba904f8aa93c91b4f