SlideShare a Scribd company logo
Jaeyoung Huh1,
Shujaat Khan1, and Jong Chul Ye2
BISPL - BioImaging, Signal Processing, and Learning lab.
1 Dept. Bio & Brain Engineering
2 Kim Jaechul Graduate School of AI
KAIST, Korea
Multi-Domain Unpaired Ultrasound Image
Artifact Removal Using
a Single Convolutional Neural Network
Introduction
 Ultrasound image artifact
- Blurring artifact from limitation by the
bandwidth of the transducer.
"Unsupervised deconvolution neural network for high quality ultrasound imaging." International Ultrasonics Symposium (IUS). IEEE, 2020.
No interference Constructive
interference
Destructive
interference
Signal
1
Signal
2
Signal
summation
 Deconvolution  Despeckle
- Speckle pattern generated by
instructive and destructive signal
interference.
Motivation
 Conventional Method
 US artifact hinder accurate diagnosis.
 Many researchers have proposed
various model-based iterative
algorithms
 computationally expensive
"Increasing axial resolution of ultrasonic imaging with a joint sparse representation model." TUFFC, 2016.
"A non-local low-rank framework for ultrasound speckle reduction." CVPR, 2017.
Motivation
 Deep learning based single domain image to
image translation such as CycleGAN.
 Still have some technical hurdles to wide
acceptance.
 Different types of artifact needs distinct choice of
artifact suppression algorithms.
 Single domain translation model
Motivation
 Recently, multi-domain image translation
models are proposed such as StarGAN.
 StarGAN require bi-directional translation
which is unnecessary in US image.
 Multi-domain translation model
Background
 Representative single-domain unpaired image-to-image translation method.
 It minimizes the distance between 𝜇, 𝜇𝜃 and 𝜈, 𝜈𝜙.
 Geometry of CycleGAN  CycleGAN structure
 CycleGAN
"Unpaired image-to-image translation using cycle-consistent adversarial networks.“ CVPR, 2017.
Background
 StarGAN
"Stargan: Unified generative adversarial networks for multi-domain image-to-image translation.“ CVPR, 2018.
 StarGAN structure
 Representative multi-domain unpaired image-to-image translation method.
 It can translate into each domain by using single generator.
 Each domain is separated by mask vector composed of one-hot vector.
Strategy
 Multi-domain Ultrasound image artifact removal using a single CNN
 Delay-And-Sum (DAS) image is generated from
raw data obtained from the scanner.
=> not necessary to re-generate from other
domain image.
 By combining the concept of CycleGAN and
StarGAN, we proposed uni-directional multi-
domain translation method for US image artifact
removal.
Our contribution
 Proposed method
Our contribution
 Generator, Discriminator Structure
 The generator is U-Net with residual block.
 The discriminator is multi-head structure.
3x3 Convolution with stride=1 + InstNorm + Relu
3x3 Convolution with stride=2 + InstNorm + Relu
3x3 Convolution with stride=1 + InstNorm
3x3 Transposed convolution with stride=2 + InstNorm + Relu
3x3 Convolution with stride=1 + Tanh
4x4 Convolution with stride=2 + LeakyRelu
3x3 Convolution with stride=1
4x4 Convolution with stride=1
Domain
classification
Real/Fake
classification
Training Details
Implementation Details
Total Epoch 1000
Learning Rate
Linear decreasing from 1e-4 after the
half of total epoch
Optimizer Adam Optimizer
Batch size 4
Parameter (𝜆𝑐𝑦𝑐, 𝜆𝐺𝑃, 𝜆𝑐𝑙𝑠)= (20,30,1)
Optimization
Formulation
WGAN with Gradient-Penalty
Dataset Details
Training set
Input - 304 images (8 subjects)
Input -125 images (Phantom)
Target - 429 images
(generated using input image)
Test set
80 images (2 subjects)
16 images (tissue mimicking phantom)
Data
Augmentation
Flipping, Rotating, Random Scaling
Normalization Normalized all images to -1~1
Target
generation
- Deconvolution
"Increasing axial resolution of ultrasonic imaging
with a joint sparse representation model." TUFFC,
2016.
- Despeckle
"A non-local low-rank framework for ultrasound
speckle reduction." CVPR, 2017.
Results
 Qualitative
Results
 Quantitative
Conclusion
 Ultrasound (US) image suffer from many artifacts from various sources.
 We proposed multi-domain US image artifact removal method using single convolution
neural network.
 The method is based on representative image-to-image (I2I) translation algorithm,
CycleGAN and StarGAN.
 A single network can provide blurring removed or speckle suppressed image.
Thank You

More Related Content

Similar to MULTI-DOMAIN UNPAIRED ULTRASOUND IMAGE ARTIFACT REMOVAL USING A SINGLE CONVOLUTIONAL NEURAL NETWORK.pptx

Towards fine-precision automated immobilization in maskless radiosurgery
Towards fine-precision automated immobilization in maskless radiosurgeryTowards fine-precision automated immobilization in maskless radiosurgery
Towards fine-precision automated immobilization in maskless radiosurgery
Olalekan Ogunmolu
 
CT Image reconstruction
CT Image reconstructionCT Image reconstruction
CT Image reconstruction
Santosh Ojha
 
PhD Qualifying Exam Slides
PhD Qualifying Exam SlidesPhD Qualifying Exam Slides
PhD Qualifying Exam Slides
Olalekan Ogunmolu
 
ct_meeting_final_jcy (1).pdf
ct_meeting_final_jcy (1).pdfct_meeting_final_jcy (1).pdf
ct_meeting_final_jcy (1).pdf
ssuser2c7393
 
15 20 jul17 6jun 16151 gunawanedgenn jul17final(edit)n
15 20 jul17 6jun 16151 gunawanedgenn jul17final(edit)n15 20 jul17 6jun 16151 gunawanedgenn jul17final(edit)n
15 20 jul17 6jun 16151 gunawanedgenn jul17final(edit)n
IAESIJEECS
 
Hairong Qi V Swaminathan
Hairong Qi V SwaminathanHairong Qi V Swaminathan
Hairong Qi V Swaminathan
FNian
 
Machine learning for Tomographic Imaging.pdf
Machine learning for Tomographic Imaging.pdfMachine learning for Tomographic Imaging.pdf
Machine learning for Tomographic Imaging.pdf
Munir Ahmad
 
Machine learning for Tomographic Imaging.pptx
Machine learning for Tomographic Imaging.pptxMachine learning for Tomographic Imaging.pptx
Machine learning for Tomographic Imaging.pptx
Munir Ahmad
 
Super Resolution of Image
Super Resolution of ImageSuper Resolution of Image
Super Resolution of Image
Satheesh K
 
Ieee 2016 nss mic poster N30-21
Ieee 2016 nss mic poster N30-21Ieee 2016 nss mic poster N30-21
Ieee 2016 nss mic poster N30-21
Dae Woon Kim
 
Final Poster
Final PosterFinal Poster
Final Poster
Elizabeth Koshelev
 
Universal plane wave compounding for high quality us imaging using deep learning
Universal plane wave compounding for high quality us imaging using deep learningUniversal plane wave compounding for high quality us imaging using deep learning
Universal plane wave compounding for high quality us imaging using deep learning
Shujaat Khan
 
Deep learning for image super resolution
Deep learning for image super resolutionDeep learning for image super resolution
Deep learning for image super resolution
Prudhvi Raj
 
Deep learning for image super resolution
Deep learning for image super resolutionDeep learning for image super resolution
Deep learning for image super resolution
Prudhvi Raj
 
An Efficient Thresholding Neural Network Technique for High Noise Densities E...
An Efficient Thresholding Neural Network Technique for High Noise Densities E...An Efficient Thresholding Neural Network Technique for High Noise Densities E...
An Efficient Thresholding Neural Network Technique for High Noise Densities E...
CSCJournals
 
W6P3622650776P65
W6P3622650776P65W6P3622650776P65
W6P3622650776P65
Ricardo Ferrari
 
Ijri ece-01-02 image enhancement aided denoising using dual tree complex wave...
Ijri ece-01-02 image enhancement aided denoising using dual tree complex wave...Ijri ece-01-02 image enhancement aided denoising using dual tree complex wave...
Ijri ece-01-02 image enhancement aided denoising using dual tree complex wave...
Ijripublishers Ijri
 
Defying Nyquist in Analog to Digital Conversion
Defying Nyquist in Analog to Digital ConversionDefying Nyquist in Analog to Digital Conversion
Defying Nyquist in Analog to Digital Conversion
Distinguished Lecturer Series - Leon The Mathematician
 
November 9, Planning and Control of Unmanned Aircraft Systems in Realistic C...
November 9, Planning and Control of Unmanned Aircraft Systems  in Realistic C...November 9, Planning and Control of Unmanned Aircraft Systems  in Realistic C...
November 9, Planning and Control of Unmanned Aircraft Systems in Realistic C...
University of Colorado at Boulder
 
Automatic System for Detection and Classification of Brain Tumors
Automatic System for Detection and Classification of Brain TumorsAutomatic System for Detection and Classification of Brain Tumors
Automatic System for Detection and Classification of Brain Tumors
Fatma Sayed Ibrahim
 

Similar to MULTI-DOMAIN UNPAIRED ULTRASOUND IMAGE ARTIFACT REMOVAL USING A SINGLE CONVOLUTIONAL NEURAL NETWORK.pptx (20)

Towards fine-precision automated immobilization in maskless radiosurgery
Towards fine-precision automated immobilization in maskless radiosurgeryTowards fine-precision automated immobilization in maskless radiosurgery
Towards fine-precision automated immobilization in maskless radiosurgery
 
CT Image reconstruction
CT Image reconstructionCT Image reconstruction
CT Image reconstruction
 
PhD Qualifying Exam Slides
PhD Qualifying Exam SlidesPhD Qualifying Exam Slides
PhD Qualifying Exam Slides
 
ct_meeting_final_jcy (1).pdf
ct_meeting_final_jcy (1).pdfct_meeting_final_jcy (1).pdf
ct_meeting_final_jcy (1).pdf
 
15 20 jul17 6jun 16151 gunawanedgenn jul17final(edit)n
15 20 jul17 6jun 16151 gunawanedgenn jul17final(edit)n15 20 jul17 6jun 16151 gunawanedgenn jul17final(edit)n
15 20 jul17 6jun 16151 gunawanedgenn jul17final(edit)n
 
Hairong Qi V Swaminathan
Hairong Qi V SwaminathanHairong Qi V Swaminathan
Hairong Qi V Swaminathan
 
Machine learning for Tomographic Imaging.pdf
Machine learning for Tomographic Imaging.pdfMachine learning for Tomographic Imaging.pdf
Machine learning for Tomographic Imaging.pdf
 
Machine learning for Tomographic Imaging.pptx
Machine learning for Tomographic Imaging.pptxMachine learning for Tomographic Imaging.pptx
Machine learning for Tomographic Imaging.pptx
 
Super Resolution of Image
Super Resolution of ImageSuper Resolution of Image
Super Resolution of Image
 
Ieee 2016 nss mic poster N30-21
Ieee 2016 nss mic poster N30-21Ieee 2016 nss mic poster N30-21
Ieee 2016 nss mic poster N30-21
 
Final Poster
Final PosterFinal Poster
Final Poster
 
Universal plane wave compounding for high quality us imaging using deep learning
Universal plane wave compounding for high quality us imaging using deep learningUniversal plane wave compounding for high quality us imaging using deep learning
Universal plane wave compounding for high quality us imaging using deep learning
 
Deep learning for image super resolution
Deep learning for image super resolutionDeep learning for image super resolution
Deep learning for image super resolution
 
Deep learning for image super resolution
Deep learning for image super resolutionDeep learning for image super resolution
Deep learning for image super resolution
 
An Efficient Thresholding Neural Network Technique for High Noise Densities E...
An Efficient Thresholding Neural Network Technique for High Noise Densities E...An Efficient Thresholding Neural Network Technique for High Noise Densities E...
An Efficient Thresholding Neural Network Technique for High Noise Densities E...
 
W6P3622650776P65
W6P3622650776P65W6P3622650776P65
W6P3622650776P65
 
Ijri ece-01-02 image enhancement aided denoising using dual tree complex wave...
Ijri ece-01-02 image enhancement aided denoising using dual tree complex wave...Ijri ece-01-02 image enhancement aided denoising using dual tree complex wave...
Ijri ece-01-02 image enhancement aided denoising using dual tree complex wave...
 
Defying Nyquist in Analog to Digital Conversion
Defying Nyquist in Analog to Digital ConversionDefying Nyquist in Analog to Digital Conversion
Defying Nyquist in Analog to Digital Conversion
 
November 9, Planning and Control of Unmanned Aircraft Systems in Realistic C...
November 9, Planning and Control of Unmanned Aircraft Systems  in Realistic C...November 9, Planning and Control of Unmanned Aircraft Systems  in Realistic C...
November 9, Planning and Control of Unmanned Aircraft Systems in Realistic C...
 
Automatic System for Detection and Classification of Brain Tumors
Automatic System for Detection and Classification of Brain TumorsAutomatic System for Detection and Classification of Brain Tumors
Automatic System for Detection and Classification of Brain Tumors
 

Recently uploaded

Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
University of Maribor
 
bank management system in java and mysql report1.pdf
bank management system in java and mysql report1.pdfbank management system in java and mysql report1.pdf
bank management system in java and mysql report1.pdf
Divyam548318
 
Embedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoringEmbedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoring
IJECEIAES
 
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
insn4465
 
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMSA SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
IJNSA Journal
 
Literature Review Basics and Understanding Reference Management.pptx
Literature Review Basics and Understanding Reference Management.pptxLiterature Review Basics and Understanding Reference Management.pptx
Literature Review Basics and Understanding Reference Management.pptx
Dr Ramhari Poudyal
 
PPT on GRP pipes manufacturing and testing
PPT on GRP pipes manufacturing and testingPPT on GRP pipes manufacturing and testing
PPT on GRP pipes manufacturing and testing
anoopmanoharan2
 
Properties Railway Sleepers and Test.pptx
Properties Railway Sleepers and Test.pptxProperties Railway Sleepers and Test.pptx
Properties Railway Sleepers and Test.pptx
MDSABBIROJJAMANPAYEL
 
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdfIron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
RadiNasr
 
CHINA’S GEO-ECONOMIC OUTREACH IN CENTRAL ASIAN COUNTRIES AND FUTURE PROSPECT
CHINA’S GEO-ECONOMIC OUTREACH IN CENTRAL ASIAN COUNTRIES AND FUTURE PROSPECTCHINA’S GEO-ECONOMIC OUTREACH IN CENTRAL ASIAN COUNTRIES AND FUTURE PROSPECT
CHINA’S GEO-ECONOMIC OUTREACH IN CENTRAL ASIAN COUNTRIES AND FUTURE PROSPECT
jpsjournal1
 
Exception Handling notes in java exception
Exception Handling notes in java exceptionException Handling notes in java exception
Exception Handling notes in java exception
Ratnakar Mikkili
 
Heat Resistant Concrete Presentation ppt
Heat Resistant Concrete Presentation pptHeat Resistant Concrete Presentation ppt
Heat Resistant Concrete Presentation ppt
mamunhossenbd75
 
22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt
KrishnaveniKrishnara1
 
14 Template Contractual Notice - EOT Application
14 Template Contractual Notice - EOT Application14 Template Contractual Notice - EOT Application
14 Template Contractual Notice - EOT Application
SyedAbiiAzazi1
 
Understanding Inductive Bias in Machine Learning
Understanding Inductive Bias in Machine LearningUnderstanding Inductive Bias in Machine Learning
Understanding Inductive Bias in Machine Learning
SUTEJAS
 
basic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdfbasic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdf
NidhalKahouli2
 
2. Operations Strategy in a Global Environment.ppt
2. Operations Strategy in a Global Environment.ppt2. Operations Strategy in a Global Environment.ppt
2. Operations Strategy in a Global Environment.ppt
PuktoonEngr
 
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODELDEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
gerogepatton
 
Series of visio cisco devices Cisco_Icons.ppt
Series of visio cisco devices Cisco_Icons.pptSeries of visio cisco devices Cisco_Icons.ppt
Series of visio cisco devices Cisco_Icons.ppt
PauloRodrigues104553
 
Modelagem de um CSTR com reação endotermica.pdf
Modelagem de um CSTR com reação endotermica.pdfModelagem de um CSTR com reação endotermica.pdf
Modelagem de um CSTR com reação endotermica.pdf
camseq
 

Recently uploaded (20)

Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
Presentation of IEEE Slovenia CIS (Computational Intelligence Society) Chapte...
 
bank management system in java and mysql report1.pdf
bank management system in java and mysql report1.pdfbank management system in java and mysql report1.pdf
bank management system in java and mysql report1.pdf
 
Embedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoringEmbedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoring
 
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
哪里办理(csu毕业证书)查尔斯特大学毕业证硕士学历原版一模一样
 
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMSA SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMS
 
Literature Review Basics and Understanding Reference Management.pptx
Literature Review Basics and Understanding Reference Management.pptxLiterature Review Basics and Understanding Reference Management.pptx
Literature Review Basics and Understanding Reference Management.pptx
 
PPT on GRP pipes manufacturing and testing
PPT on GRP pipes manufacturing and testingPPT on GRP pipes manufacturing and testing
PPT on GRP pipes manufacturing and testing
 
Properties Railway Sleepers and Test.pptx
Properties Railway Sleepers and Test.pptxProperties Railway Sleepers and Test.pptx
Properties Railway Sleepers and Test.pptx
 
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdfIron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
Iron and Steel Technology Roadmap - Towards more sustainable steelmaking.pdf
 
CHINA’S GEO-ECONOMIC OUTREACH IN CENTRAL ASIAN COUNTRIES AND FUTURE PROSPECT
CHINA’S GEO-ECONOMIC OUTREACH IN CENTRAL ASIAN COUNTRIES AND FUTURE PROSPECTCHINA’S GEO-ECONOMIC OUTREACH IN CENTRAL ASIAN COUNTRIES AND FUTURE PROSPECT
CHINA’S GEO-ECONOMIC OUTREACH IN CENTRAL ASIAN COUNTRIES AND FUTURE PROSPECT
 
Exception Handling notes in java exception
Exception Handling notes in java exceptionException Handling notes in java exception
Exception Handling notes in java exception
 
Heat Resistant Concrete Presentation ppt
Heat Resistant Concrete Presentation pptHeat Resistant Concrete Presentation ppt
Heat Resistant Concrete Presentation ppt
 
22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt
 
14 Template Contractual Notice - EOT Application
14 Template Contractual Notice - EOT Application14 Template Contractual Notice - EOT Application
14 Template Contractual Notice - EOT Application
 
Understanding Inductive Bias in Machine Learning
Understanding Inductive Bias in Machine LearningUnderstanding Inductive Bias in Machine Learning
Understanding Inductive Bias in Machine Learning
 
basic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdfbasic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdf
 
2. Operations Strategy in a Global Environment.ppt
2. Operations Strategy in a Global Environment.ppt2. Operations Strategy in a Global Environment.ppt
2. Operations Strategy in a Global Environment.ppt
 
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODELDEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
 
Series of visio cisco devices Cisco_Icons.ppt
Series of visio cisco devices Cisco_Icons.pptSeries of visio cisco devices Cisco_Icons.ppt
Series of visio cisco devices Cisco_Icons.ppt
 
Modelagem de um CSTR com reação endotermica.pdf
Modelagem de um CSTR com reação endotermica.pdfModelagem de um CSTR com reação endotermica.pdf
Modelagem de um CSTR com reação endotermica.pdf
 

MULTI-DOMAIN UNPAIRED ULTRASOUND IMAGE ARTIFACT REMOVAL USING A SINGLE CONVOLUTIONAL NEURAL NETWORK.pptx

  • 1. Jaeyoung Huh1, Shujaat Khan1, and Jong Chul Ye2 BISPL - BioImaging, Signal Processing, and Learning lab. 1 Dept. Bio & Brain Engineering 2 Kim Jaechul Graduate School of AI KAIST, Korea Multi-Domain Unpaired Ultrasound Image Artifact Removal Using a Single Convolutional Neural Network
  • 2. Introduction  Ultrasound image artifact - Blurring artifact from limitation by the bandwidth of the transducer. "Unsupervised deconvolution neural network for high quality ultrasound imaging." International Ultrasonics Symposium (IUS). IEEE, 2020. No interference Constructive interference Destructive interference Signal 1 Signal 2 Signal summation  Deconvolution  Despeckle - Speckle pattern generated by instructive and destructive signal interference.
  • 3. Motivation  Conventional Method  US artifact hinder accurate diagnosis.  Many researchers have proposed various model-based iterative algorithms  computationally expensive "Increasing axial resolution of ultrasonic imaging with a joint sparse representation model." TUFFC, 2016. "A non-local low-rank framework for ultrasound speckle reduction." CVPR, 2017.
  • 4. Motivation  Deep learning based single domain image to image translation such as CycleGAN.  Still have some technical hurdles to wide acceptance.  Different types of artifact needs distinct choice of artifact suppression algorithms.  Single domain translation model
  • 5. Motivation  Recently, multi-domain image translation models are proposed such as StarGAN.  StarGAN require bi-directional translation which is unnecessary in US image.  Multi-domain translation model
  • 6. Background  Representative single-domain unpaired image-to-image translation method.  It minimizes the distance between 𝜇, 𝜇𝜃 and 𝜈, 𝜈𝜙.  Geometry of CycleGAN  CycleGAN structure  CycleGAN "Unpaired image-to-image translation using cycle-consistent adversarial networks.“ CVPR, 2017.
  • 7. Background  StarGAN "Stargan: Unified generative adversarial networks for multi-domain image-to-image translation.“ CVPR, 2018.  StarGAN structure  Representative multi-domain unpaired image-to-image translation method.  It can translate into each domain by using single generator.  Each domain is separated by mask vector composed of one-hot vector.
  • 8. Strategy  Multi-domain Ultrasound image artifact removal using a single CNN  Delay-And-Sum (DAS) image is generated from raw data obtained from the scanner. => not necessary to re-generate from other domain image.  By combining the concept of CycleGAN and StarGAN, we proposed uni-directional multi- domain translation method for US image artifact removal.
  • 10. Our contribution  Generator, Discriminator Structure  The generator is U-Net with residual block.  The discriminator is multi-head structure. 3x3 Convolution with stride=1 + InstNorm + Relu 3x3 Convolution with stride=2 + InstNorm + Relu 3x3 Convolution with stride=1 + InstNorm 3x3 Transposed convolution with stride=2 + InstNorm + Relu 3x3 Convolution with stride=1 + Tanh 4x4 Convolution with stride=2 + LeakyRelu 3x3 Convolution with stride=1 4x4 Convolution with stride=1 Domain classification Real/Fake classification
  • 11. Training Details Implementation Details Total Epoch 1000 Learning Rate Linear decreasing from 1e-4 after the half of total epoch Optimizer Adam Optimizer Batch size 4 Parameter (𝜆𝑐𝑦𝑐, 𝜆𝐺𝑃, 𝜆𝑐𝑙𝑠)= (20,30,1) Optimization Formulation WGAN with Gradient-Penalty Dataset Details Training set Input - 304 images (8 subjects) Input -125 images (Phantom) Target - 429 images (generated using input image) Test set 80 images (2 subjects) 16 images (tissue mimicking phantom) Data Augmentation Flipping, Rotating, Random Scaling Normalization Normalized all images to -1~1 Target generation - Deconvolution "Increasing axial resolution of ultrasonic imaging with a joint sparse representation model." TUFFC, 2016. - Despeckle "A non-local low-rank framework for ultrasound speckle reduction." CVPR, 2017.
  • 14. Conclusion  Ultrasound (US) image suffer from many artifacts from various sources.  We proposed multi-domain US image artifact removal method using single convolution neural network.  The method is based on representative image-to-image (I2I) translation algorithm, CycleGAN and StarGAN.  A single network can provide blurring removed or speckle suppressed image.