SlideShare uma empresa Scribd logo
1 de 16
Baixar para ler offline
NNUNET
A self-configuring method for deep learning-based
biomedical image segmentation
Introduction
Medical Image Segmentation is difficult
because of the vast diversity of modalities,
each of which requires a specialized pipeline
for data pre-processing and training.
nnUNet (“no new U-Net”) seeks to establish
a standardized pipeline for the medical image
segmentation process.
2
Problem Statement
• Analysis of challenge leaderboard entries shows
that using superficially similar methods and model
architectures can lead to vastly different results
due to implementation details.
• Moreover, no single method stands out as being
necessary for high scores (except deep learning).
Method
• nnUNet divides hyper-parameters into 3
types:
(1) Fixed configurations
(2) Rule-Based configurations
(3) Empirical Configurations
• In all cases, validation set performance does
not affect training time, unlike in AutoML.
Fixed Configurations
• Model architecture (U-Net), hence the name “no new U-Net”.
• Learning rate value and scheduling (0.01 with poly decay).
• Optimizer (SGD with Nesterov momentum 0.9).
• Training procedure (250k iterations with 5-fold cross-validation and fore-ground over-sampling).
• Inference procedure (sliding window with Gaussian importance weighting).
5
Rule-Based Configurations
• Image Intensity Normalization (Use HU if CT, else use z normalization).
• Image Resampling Strategy (If anisotropic, use cubic spline if ratio is within 3, else use nearest
neighbor interpolation).
• Image Spacing (lowest 10th percentile if anisotropic, else median).
• Use 3D cascade (if image is too large).
• Model pooling depth (reduce anisotropic side until less than 3, pool until side length becomes 4).
• Mini-batch size (largest mini-batch that fits within 11 GB during training).
6
Rule-Based Model Configuration
• Network Topology, Patch Size, and Batch Size are configured at the start of training for effective
training within 11GB of memory.
• Networks are expected to give approximate GPU memory usage expectations, which makes
implementing new models somewhat cumbersome.
• Patch size is given high priority as large patch size is necessary for segmentation.
• Network topology is designed to pool until side length is 4 and anisotropic degree is within 3.
• Batch size is set to be lower than 5% of total data. Usually set to 1 or 2 as it has lowest priority.
7
Empirical Parameters
• Only 2 parameters are set empirically, both of which are post-processing steps and therefore do not
affect training time.
• Suppression of non-largest segmented organ (uses the prior that humans have one or two of each
organ).
• Ensemble selection of 2D, 3D, and 3D cascade trained networks. These are selected from cross-
validation results.
8
Comments
• Training time and memory requirements do not depend on performance on validation metrics.
• No pre-trained networks are necessary as inputs.
• No 2.5D, which may be effective in some anisotropic tasks. More robust to anisotropic data while
not abandoning the information from the anisotropic direction.
• No analysis on the effect of label quality. For example, what is the effect of nnUNet when
comparing clean and noisy labels. Is it robust to noisy labels? What attributes are important for
learning with sparsely labeled data.
9
Results
• First place in 33 of 53 challenges with
no modifications.
• Maintains a high rank in all challenges
submitted, though in some cases,
modifications to the original nnUNet
was necessary (e.g., 2020 COVID19
segmentation challenge).
• nnUNet is the baseline for most new
medical segmentation challenges.
Results
• Results for COVID19 segmentation
challenge in 2020.
• nnUNet features in the top-tier of
nearly all medical segmentation
challenges.
• First place used nnUNet to pseudo-
label more data for additional
training with nnUNet.
11
Code Analysis:
Dataset Conversion
First convert the data into nifti file
format for unified reading.
Files must be structured in the specified
format for proper training.
12
Code Analysis: Training & Inference
• Dataset experiment planning and pre-processing:
nnUNet_plan_and_preprocess -t XXX --verify_dataset_integrity
• Model training:
nnUNet_train CONFIGURATION TRAINER_CLASS_NAME TASK_NAME_OR_ID FOLD --npz (additional options)
• Find best configuration:
nnUNet_find_best_configuration -m 2d 3d_fullres 3d_lowres 3d_cascade_fullres -t XXX --strict
• Run inference:
nnUNet_predict -i INPUT_FOLDER -o OUTPUT_FOLDER -t TASK_NAME_OR_ID -m CONFIGURATION --save_npz
• Run ensemble:
nnUNet_ensemble -f FOLDER1 FOLDER2 ... -o OUTPUT_FOLDER -pp POSTPROCESSING_FILE
13
Resources
• Paper: https://www.nature.com/articles/s41592-020-01008-z (also read
the supplementary material)
• YouTube videos:
https://www.youtube.com/watch?v=3po8qVzz5Tc
https://www.youtube.com/watch?v=C6tpnJRpt90
• GitHub: https://github.com/MIC-DKFZ/nnUNet
14
Related Work: nnDetection
• Paper: https://arxiv.org/abs/2106.00817
• GitHub: https://github.com/MIC-DKFZ/nnDetection
15
16
THE END.

Mais conteúdo relacionado

Mais procurados

Faster R-CNN - PR012
Faster R-CNN - PR012Faster R-CNN - PR012
Faster R-CNN - PR012Jinwon Lee
 
Batch normalization paper review
Batch normalization paper reviewBatch normalization paper review
Batch normalization paper reviewMinho Heo
 
Imagen: Photorealistic Text-to-Image Diffusion Models with Deep Language Unde...
Imagen: Photorealistic Text-to-Image Diffusion Models with Deep Language Unde...Imagen: Photorealistic Text-to-Image Diffusion Models with Deep Language Unde...
Imagen: Photorealistic Text-to-Image Diffusion Models with Deep Language Unde...Vitaly Bondar
 
"How does batch normalization help optimization" Paper Review
"How does batch normalization help optimization" Paper Review"How does batch normalization help optimization" Paper Review
"How does batch normalization help optimization" Paper ReviewLEE HOSEONG
 
Practical SPU Programming in God of War III
Practical SPU Programming in God of War IIIPractical SPU Programming in God of War III
Practical SPU Programming in God of War IIISlide_N
 
Introduction to Diffusion Models
Introduction to Diffusion ModelsIntroduction to Diffusion Models
Introduction to Diffusion ModelsSangwoo Mo
 
Batch normalization presentation
Batch normalization presentationBatch normalization presentation
Batch normalization presentationOwin Will
 
Diffusion models beat gans on image synthesis
Diffusion models beat gans on image synthesisDiffusion models beat gans on image synthesis
Diffusion models beat gans on image synthesisBeerenSahu
 
“Introduction to DNN Model Compression Techniques,” a Presentation from Xailient
“Introduction to DNN Model Compression Techniques,” a Presentation from Xailient“Introduction to DNN Model Compression Techniques,” a Presentation from Xailient
“Introduction to DNN Model Compression Techniques,” a Presentation from XailientEdge AI and Vision Alliance
 
Variational Auto Encoder and the Math Behind
Variational Auto Encoder and the Math BehindVariational Auto Encoder and the Math Behind
Variational Auto Encoder and the Math BehindVarun Reddy
 
Convolutional neural networks 이론과 응용
Convolutional neural networks 이론과 응용Convolutional neural networks 이론과 응용
Convolutional neural networks 이론과 응용홍배 김
 
Deep learning lecture - part 1 (basics, CNN)
Deep learning lecture - part 1 (basics, CNN)Deep learning lecture - part 1 (basics, CNN)
Deep learning lecture - part 1 (basics, CNN)SungminYou
 
Variational Autoencoder
Variational AutoencoderVariational Autoencoder
Variational AutoencoderMark Chang
 
[기초개념] Graph Convolutional Network (GCN)
[기초개념] Graph Convolutional Network (GCN)[기초개념] Graph Convolutional Network (GCN)
[기초개념] Graph Convolutional Network (GCN)Donghyeon Kim
 
Dimensionality Reduction
Dimensionality ReductionDimensionality Reduction
Dimensionality ReductionSaad Elbeleidy
 
HRNET : Deep High-Resolution Representation Learning for Human Pose Estimation
HRNET : Deep High-Resolution Representation Learning for Human Pose EstimationHRNET : Deep High-Resolution Representation Learning for Human Pose Estimation
HRNET : Deep High-Resolution Representation Learning for Human Pose Estimationtaeseon ryu
 
Backpropagation in Convolutional Neural Network
Backpropagation in Convolutional Neural NetworkBackpropagation in Convolutional Neural Network
Backpropagation in Convolutional Neural NetworkHiroshi Kuwajima
 
Revisiting the Calibration of Modern Neural Networks
Revisiting the Calibration of Modern Neural NetworksRevisiting the Calibration of Modern Neural Networks
Revisiting the Calibration of Modern Neural NetworksSungchul Kim
 
A 2.5D Culling for Forward+ (SIGGRAPH ASIA 2012)
A 2.5D Culling for Forward+ (SIGGRAPH ASIA 2012)A 2.5D Culling for Forward+ (SIGGRAPH ASIA 2012)
A 2.5D Culling for Forward+ (SIGGRAPH ASIA 2012)Takahiro Harada
 
Convolutional Neural Network Models - Deep Learning
Convolutional Neural Network Models - Deep LearningConvolutional Neural Network Models - Deep Learning
Convolutional Neural Network Models - Deep LearningMohamed Loey
 

Mais procurados (20)

Faster R-CNN - PR012
Faster R-CNN - PR012Faster R-CNN - PR012
Faster R-CNN - PR012
 
Batch normalization paper review
Batch normalization paper reviewBatch normalization paper review
Batch normalization paper review
 
Imagen: Photorealistic Text-to-Image Diffusion Models with Deep Language Unde...
Imagen: Photorealistic Text-to-Image Diffusion Models with Deep Language Unde...Imagen: Photorealistic Text-to-Image Diffusion Models with Deep Language Unde...
Imagen: Photorealistic Text-to-Image Diffusion Models with Deep Language Unde...
 
"How does batch normalization help optimization" Paper Review
"How does batch normalization help optimization" Paper Review"How does batch normalization help optimization" Paper Review
"How does batch normalization help optimization" Paper Review
 
Practical SPU Programming in God of War III
Practical SPU Programming in God of War IIIPractical SPU Programming in God of War III
Practical SPU Programming in God of War III
 
Introduction to Diffusion Models
Introduction to Diffusion ModelsIntroduction to Diffusion Models
Introduction to Diffusion Models
 
Batch normalization presentation
Batch normalization presentationBatch normalization presentation
Batch normalization presentation
 
Diffusion models beat gans on image synthesis
Diffusion models beat gans on image synthesisDiffusion models beat gans on image synthesis
Diffusion models beat gans on image synthesis
 
“Introduction to DNN Model Compression Techniques,” a Presentation from Xailient
“Introduction to DNN Model Compression Techniques,” a Presentation from Xailient“Introduction to DNN Model Compression Techniques,” a Presentation from Xailient
“Introduction to DNN Model Compression Techniques,” a Presentation from Xailient
 
Variational Auto Encoder and the Math Behind
Variational Auto Encoder and the Math BehindVariational Auto Encoder and the Math Behind
Variational Auto Encoder and the Math Behind
 
Convolutional neural networks 이론과 응용
Convolutional neural networks 이론과 응용Convolutional neural networks 이론과 응용
Convolutional neural networks 이론과 응용
 
Deep learning lecture - part 1 (basics, CNN)
Deep learning lecture - part 1 (basics, CNN)Deep learning lecture - part 1 (basics, CNN)
Deep learning lecture - part 1 (basics, CNN)
 
Variational Autoencoder
Variational AutoencoderVariational Autoencoder
Variational Autoencoder
 
[기초개념] Graph Convolutional Network (GCN)
[기초개념] Graph Convolutional Network (GCN)[기초개념] Graph Convolutional Network (GCN)
[기초개념] Graph Convolutional Network (GCN)
 
Dimensionality Reduction
Dimensionality ReductionDimensionality Reduction
Dimensionality Reduction
 
HRNET : Deep High-Resolution Representation Learning for Human Pose Estimation
HRNET : Deep High-Resolution Representation Learning for Human Pose EstimationHRNET : Deep High-Resolution Representation Learning for Human Pose Estimation
HRNET : Deep High-Resolution Representation Learning for Human Pose Estimation
 
Backpropagation in Convolutional Neural Network
Backpropagation in Convolutional Neural NetworkBackpropagation in Convolutional Neural Network
Backpropagation in Convolutional Neural Network
 
Revisiting the Calibration of Modern Neural Networks
Revisiting the Calibration of Modern Neural NetworksRevisiting the Calibration of Modern Neural Networks
Revisiting the Calibration of Modern Neural Networks
 
A 2.5D Culling for Forward+ (SIGGRAPH ASIA 2012)
A 2.5D Culling for Forward+ (SIGGRAPH ASIA 2012)A 2.5D Culling for Forward+ (SIGGRAPH ASIA 2012)
A 2.5D Culling for Forward+ (SIGGRAPH ASIA 2012)
 
Convolutional Neural Network Models - Deep Learning
Convolutional Neural Network Models - Deep LearningConvolutional Neural Network Models - Deep Learning
Convolutional Neural Network Models - Deep Learning
 

Semelhante a nnUNet

PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...
PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...
PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...Jinwon Lee
 
Table of Contents
Table of ContentsTable of Contents
Table of Contentsbutest
 
Comparing Incremental Learning Strategies for Convolutional Neural Networks
Comparing Incremental Learning Strategies for Convolutional Neural NetworksComparing Incremental Learning Strategies for Convolutional Neural Networks
Comparing Incremental Learning Strategies for Convolutional Neural NetworksVincenzo Lomonaco
 
Kcc201728apr2017 170828235330
Kcc201728apr2017 170828235330Kcc201728apr2017 170828235330
Kcc201728apr2017 170828235330JEE HYUN PARK
 
PR-231: A Simple Framework for Contrastive Learning of Visual Representations
PR-231: A Simple Framework for Contrastive Learning of Visual RepresentationsPR-231: A Simple Framework for Contrastive Learning of Visual Representations
PR-231: A Simple Framework for Contrastive Learning of Visual RepresentationsJinwon Lee
 
ResNeSt: Split-Attention Networks
ResNeSt: Split-Attention NetworksResNeSt: Split-Attention Networks
ResNeSt: Split-Attention NetworksSeunghyun Hwang
 
Semi-Supervised Deep Learning
Semi-Supervised Deep LearningSemi-Supervised Deep Learning
Semi-Supervised Deep LearningKamer Ali Yuksel
 
PR-433: Test-time Training with Masked Autoencoders
PR-433: Test-time Training with Masked AutoencodersPR-433: Test-time Training with Masked Autoencoders
PR-433: Test-time Training with Masked AutoencodersSunghoon Joo
 
Presentation 7.pptx
Presentation 7.pptxPresentation 7.pptx
Presentation 7.pptxShivam327815
 
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...Edge AI and Vision Alliance
 
Distributed approach for Peptide Identification
Distributed approach for Peptide IdentificationDistributed approach for Peptide Identification
Distributed approach for Peptide Identificationabhinav vedanbhatla
 
IRJET- Analysis of PV Fed Vector Controlled Induction Motor Drive
IRJET- Analysis of PV Fed Vector Controlled Induction Motor DriveIRJET- Analysis of PV Fed Vector Controlled Induction Motor Drive
IRJET- Analysis of PV Fed Vector Controlled Induction Motor DriveIRJET Journal
 
IRJET- Deep Learning Model to Predict Hardware Performance
IRJET- Deep Learning Model to Predict Hardware PerformanceIRJET- Deep Learning Model to Predict Hardware Performance
IRJET- Deep Learning Model to Predict Hardware PerformanceIRJET Journal
 
How can pre-training help to solve the cold start problem?
How can pre-training help to solve the cold start problem?How can pre-training help to solve the cold start problem?
How can pre-training help to solve the cold start problem?Lokesh Vadlamudi
 
Deep learning summary
Deep learning summaryDeep learning summary
Deep learning summaryankit_ppt
 
IEEE 2014 JAVA DATA MINING PROJECTS A similarity measure for text classificat...
IEEE 2014 JAVA DATA MINING PROJECTS A similarity measure for text classificat...IEEE 2014 JAVA DATA MINING PROJECTS A similarity measure for text classificat...
IEEE 2014 JAVA DATA MINING PROJECTS A similarity measure for text classificat...IEEEFINALYEARSTUDENTPROJECTS
 
2014 IEEE JAVA DATA MINING PROJECT A similarity measure for text classificati...
2014 IEEE JAVA DATA MINING PROJECT A similarity measure for text classificati...2014 IEEE JAVA DATA MINING PROJECT A similarity measure for text classificati...
2014 IEEE JAVA DATA MINING PROJECT A similarity measure for text classificati...IEEEMEMTECHSTUDENTSPROJECTS
 
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...Edge AI and Vision Alliance
 

Semelhante a nnUNet (20)

PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...
PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...
PR-330: How To Train Your ViT? Data, Augmentation, and Regularization in Visi...
 
Table of Contents
Table of ContentsTable of Contents
Table of Contents
 
Comparing Incremental Learning Strategies for Convolutional Neural Networks
Comparing Incremental Learning Strategies for Convolutional Neural NetworksComparing Incremental Learning Strategies for Convolutional Neural Networks
Comparing Incremental Learning Strategies for Convolutional Neural Networks
 
Kcc201728apr2017 170828235330
Kcc201728apr2017 170828235330Kcc201728apr2017 170828235330
Kcc201728apr2017 170828235330
 
PR-231: A Simple Framework for Contrastive Learning of Visual Representations
PR-231: A Simple Framework for Contrastive Learning of Visual RepresentationsPR-231: A Simple Framework for Contrastive Learning of Visual Representations
PR-231: A Simple Framework for Contrastive Learning of Visual Representations
 
ResNeSt: Split-Attention Networks
ResNeSt: Split-Attention NetworksResNeSt: Split-Attention Networks
ResNeSt: Split-Attention Networks
 
Semi-Supervised Deep Learning
Semi-Supervised Deep LearningSemi-Supervised Deep Learning
Semi-Supervised Deep Learning
 
PR-433: Test-time Training with Masked Autoencoders
PR-433: Test-time Training with Masked AutoencodersPR-433: Test-time Training with Masked Autoencoders
PR-433: Test-time Training with Masked Autoencoders
 
Presentation 7.pptx
Presentation 7.pptxPresentation 7.pptx
Presentation 7.pptx
 
KCC2017 28APR2017
KCC2017 28APR2017KCC2017 28APR2017
KCC2017 28APR2017
 
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
 
Distributed approach for Peptide Identification
Distributed approach for Peptide IdentificationDistributed approach for Peptide Identification
Distributed approach for Peptide Identification
 
IRJET- Analysis of PV Fed Vector Controlled Induction Motor Drive
IRJET- Analysis of PV Fed Vector Controlled Induction Motor DriveIRJET- Analysis of PV Fed Vector Controlled Induction Motor Drive
IRJET- Analysis of PV Fed Vector Controlled Induction Motor Drive
 
IRJET- Deep Learning Model to Predict Hardware Performance
IRJET- Deep Learning Model to Predict Hardware PerformanceIRJET- Deep Learning Model to Predict Hardware Performance
IRJET- Deep Learning Model to Predict Hardware Performance
 
How can pre-training help to solve the cold start problem?
How can pre-training help to solve the cold start problem?How can pre-training help to solve the cold start problem?
How can pre-training help to solve the cold start problem?
 
HW03 (1).pdf
HW03 (1).pdfHW03 (1).pdf
HW03 (1).pdf
 
Deep learning summary
Deep learning summaryDeep learning summary
Deep learning summary
 
IEEE 2014 JAVA DATA MINING PROJECTS A similarity measure for text classificat...
IEEE 2014 JAVA DATA MINING PROJECTS A similarity measure for text classificat...IEEE 2014 JAVA DATA MINING PROJECTS A similarity measure for text classificat...
IEEE 2014 JAVA DATA MINING PROJECTS A similarity measure for text classificat...
 
2014 IEEE JAVA DATA MINING PROJECT A similarity measure for text classificati...
2014 IEEE JAVA DATA MINING PROJECT A similarity measure for text classificati...2014 IEEE JAVA DATA MINING PROJECT A similarity measure for text classificati...
2014 IEEE JAVA DATA MINING PROJECT A similarity measure for text classificati...
 
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
 

Mais de Joonhyung Lee

Rethinking Attention with Performers
Rethinking Attention with PerformersRethinking Attention with Performers
Rethinking Attention with PerformersJoonhyung Lee
 
Denoising Unpaired Low Dose CT Images with Self-Ensembled CycleGAN
Denoising Unpaired Low Dose CT Images with Self-Ensembled CycleGANDenoising Unpaired Low Dose CT Images with Self-Ensembled CycleGAN
Denoising Unpaired Low Dose CT Images with Self-Ensembled CycleGANJoonhyung Lee
 
Deep Learning Fast MRI Using Channel Attention in Magnitude Domain
Deep Learning Fast MRI Using Channel Attention in Magnitude DomainDeep Learning Fast MRI Using Channel Attention in Magnitude Domain
Deep Learning Fast MRI Using Channel Attention in Magnitude DomainJoonhyung Lee
 
InfoGAN: Interpretable Representation Learning by Information Maximizing Gene...
InfoGAN: Interpretable Representation Learning by Information Maximizing Gene...InfoGAN: Interpretable Representation Learning by Information Maximizing Gene...
InfoGAN: Interpretable Representation Learning by Information Maximizing Gene...Joonhyung Lee
 
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...Joonhyung Lee
 
AlphaZero: A General Reinforcement Learning Algorithm that Masters Chess, Sho...
AlphaZero: A General Reinforcement Learning Algorithm that Masters Chess, Sho...AlphaZero: A General Reinforcement Learning Algorithm that Masters Chess, Sho...
AlphaZero: A General Reinforcement Learning Algorithm that Masters Chess, Sho...Joonhyung Lee
 
Squeeze Excitation Networks, The simple idea that won the final ImageNet Chal...
Squeeze Excitation Networks, The simple idea that won the final ImageNet Chal...Squeeze Excitation Networks, The simple idea that won the final ImageNet Chal...
Squeeze Excitation Networks, The simple idea that won the final ImageNet Chal...Joonhyung Lee
 
DeepLab V3+: Encoder-Decoder with Atrous Separable Convolution for Semantic I...
DeepLab V3+: Encoder-Decoder with Atrous Separable Convolution for Semantic I...DeepLab V3+: Encoder-Decoder with Atrous Separable Convolution for Semantic I...
DeepLab V3+: Encoder-Decoder with Atrous Separable Convolution for Semantic I...Joonhyung Lee
 
AlphaGo Zero: Mastering the Game of Go Without Human Knowledge
AlphaGo Zero: Mastering the Game of Go Without Human KnowledgeAlphaGo Zero: Mastering the Game of Go Without Human Knowledge
AlphaGo Zero: Mastering the Game of Go Without Human KnowledgeJoonhyung Lee
 
Deep Learning in Bio-Medical Imaging
Deep Learning in Bio-Medical ImagingDeep Learning in Bio-Medical Imaging
Deep Learning in Bio-Medical ImagingJoonhyung Lee
 

Mais de Joonhyung Lee (11)

Rethinking Attention with Performers
Rethinking Attention with PerformersRethinking Attention with Performers
Rethinking Attention with Performers
 
Denoising Unpaired Low Dose CT Images with Self-Ensembled CycleGAN
Denoising Unpaired Low Dose CT Images with Self-Ensembled CycleGANDenoising Unpaired Low Dose CT Images with Self-Ensembled CycleGAN
Denoising Unpaired Low Dose CT Images with Self-Ensembled CycleGAN
 
Deep Learning Fast MRI Using Channel Attention in Magnitude Domain
Deep Learning Fast MRI Using Channel Attention in Magnitude DomainDeep Learning Fast MRI Using Channel Attention in Magnitude Domain
Deep Learning Fast MRI Using Channel Attention in Magnitude Domain
 
InfoGAN: Interpretable Representation Learning by Information Maximizing Gene...
InfoGAN: Interpretable Representation Learning by Information Maximizing Gene...InfoGAN: Interpretable Representation Learning by Information Maximizing Gene...
InfoGAN: Interpretable Representation Learning by Information Maximizing Gene...
 
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...
 
AlphaZero: A General Reinforcement Learning Algorithm that Masters Chess, Sho...
AlphaZero: A General Reinforcement Learning Algorithm that Masters Chess, Sho...AlphaZero: A General Reinforcement Learning Algorithm that Masters Chess, Sho...
AlphaZero: A General Reinforcement Learning Algorithm that Masters Chess, Sho...
 
Squeeze Excitation Networks, The simple idea that won the final ImageNet Chal...
Squeeze Excitation Networks, The simple idea that won the final ImageNet Chal...Squeeze Excitation Networks, The simple idea that won the final ImageNet Chal...
Squeeze Excitation Networks, The simple idea that won the final ImageNet Chal...
 
DeepLab V3+: Encoder-Decoder with Atrous Separable Convolution for Semantic I...
DeepLab V3+: Encoder-Decoder with Atrous Separable Convolution for Semantic I...DeepLab V3+: Encoder-Decoder with Atrous Separable Convolution for Semantic I...
DeepLab V3+: Encoder-Decoder with Atrous Separable Convolution for Semantic I...
 
AlphaGo Zero: Mastering the Game of Go Without Human Knowledge
AlphaGo Zero: Mastering the Game of Go Without Human KnowledgeAlphaGo Zero: Mastering the Game of Go Without Human Knowledge
AlphaGo Zero: Mastering the Game of Go Without Human Knowledge
 
StarGAN
StarGANStarGAN
StarGAN
 
Deep Learning in Bio-Medical Imaging
Deep Learning in Bio-Medical ImagingDeep Learning in Bio-Medical Imaging
Deep Learning in Bio-Medical Imaging
 

Último

Generative AI - Gitex v1Generative AI - Gitex v1.pptx
Generative AI - Gitex v1Generative AI - Gitex v1.pptxGenerative AI - Gitex v1Generative AI - Gitex v1.pptx
Generative AI - Gitex v1Generative AI - Gitex v1.pptxfnnc6jmgwh
 
Time Series Foundation Models - current state and future directions
Time Series Foundation Models - current state and future directionsTime Series Foundation Models - current state and future directions
Time Series Foundation Models - current state and future directionsNathaniel Shimoni
 
Generative Artificial Intelligence: How generative AI works.pdf
Generative Artificial Intelligence: How generative AI works.pdfGenerative Artificial Intelligence: How generative AI works.pdf
Generative Artificial Intelligence: How generative AI works.pdfIngrid Airi González
 
A Framework for Development in the AI Age
A Framework for Development in the AI AgeA Framework for Development in the AI Age
A Framework for Development in the AI AgeCprime
 
Genislab builds better products and faster go-to-market with Lean project man...
Genislab builds better products and faster go-to-market with Lean project man...Genislab builds better products and faster go-to-market with Lean project man...
Genislab builds better products and faster go-to-market with Lean project man...Farhan Tariq
 
Connecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdfConnecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdfNeo4j
 
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...Wes McKinney
 
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...Alkin Tezuysal
 
How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.Curtis Poe
 
The Ultimate Guide to Choosing WordPress Pros and Cons
The Ultimate Guide to Choosing WordPress Pros and ConsThe Ultimate Guide to Choosing WordPress Pros and Cons
The Ultimate Guide to Choosing WordPress Pros and ConsPixlogix Infotech
 
Moving Beyond Passwords: FIDO Paris Seminar.pdf
Moving Beyond Passwords: FIDO Paris Seminar.pdfMoving Beyond Passwords: FIDO Paris Seminar.pdf
Moving Beyond Passwords: FIDO Paris Seminar.pdfLoriGlavin3
 
Potential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and InsightsPotential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and InsightsRavi Sanghani
 
Microsoft 365 Copilot: How to boost your productivity with AI – Part one: Ado...
Microsoft 365 Copilot: How to boost your productivity with AI – Part one: Ado...Microsoft 365 Copilot: How to boost your productivity with AI – Part one: Ado...
Microsoft 365 Copilot: How to boost your productivity with AI – Part one: Ado...Nikki Chapple
 
A Deep Dive on Passkeys: FIDO Paris Seminar.pptx
A Deep Dive on Passkeys: FIDO Paris Seminar.pptxA Deep Dive on Passkeys: FIDO Paris Seminar.pptx
A Deep Dive on Passkeys: FIDO Paris Seminar.pptxLoriGlavin3
 
React Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App FrameworkReact Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App FrameworkPixlogix Infotech
 
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptxThe Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptxLoriGlavin3
 
Abdul Kader Baba- Managing Cybersecurity Risks and Compliance Requirements i...
Abdul Kader Baba- Managing Cybersecurity Risks  and Compliance Requirements i...Abdul Kader Baba- Managing Cybersecurity Risks  and Compliance Requirements i...
Abdul Kader Baba- Managing Cybersecurity Risks and Compliance Requirements i...itnewsafrica
 
Data governance with Unity Catalog Presentation
Data governance with Unity Catalog PresentationData governance with Unity Catalog Presentation
Data governance with Unity Catalog PresentationKnoldus Inc.
 
Bridging Between CAD & GIS: 6 Ways to Automate Your Data Integration
Bridging Between CAD & GIS:  6 Ways to Automate Your Data IntegrationBridging Between CAD & GIS:  6 Ways to Automate Your Data Integration
Bridging Between CAD & GIS: 6 Ways to Automate Your Data Integrationmarketing932765
 
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptxUse of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptxLoriGlavin3
 

Último (20)

Generative AI - Gitex v1Generative AI - Gitex v1.pptx
Generative AI - Gitex v1Generative AI - Gitex v1.pptxGenerative AI - Gitex v1Generative AI - Gitex v1.pptx
Generative AI - Gitex v1Generative AI - Gitex v1.pptx
 
Time Series Foundation Models - current state and future directions
Time Series Foundation Models - current state and future directionsTime Series Foundation Models - current state and future directions
Time Series Foundation Models - current state and future directions
 
Generative Artificial Intelligence: How generative AI works.pdf
Generative Artificial Intelligence: How generative AI works.pdfGenerative Artificial Intelligence: How generative AI works.pdf
Generative Artificial Intelligence: How generative AI works.pdf
 
A Framework for Development in the AI Age
A Framework for Development in the AI AgeA Framework for Development in the AI Age
A Framework for Development in the AI Age
 
Genislab builds better products and faster go-to-market with Lean project man...
Genislab builds better products and faster go-to-market with Lean project man...Genislab builds better products and faster go-to-market with Lean project man...
Genislab builds better products and faster go-to-market with Lean project man...
 
Connecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdfConnecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdf
 
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
The Future Roadmap for the Composable Data Stack - Wes McKinney - Data Counci...
 
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...
Unleashing Real-time Insights with ClickHouse_ Navigating the Landscape in 20...
 
How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.
 
The Ultimate Guide to Choosing WordPress Pros and Cons
The Ultimate Guide to Choosing WordPress Pros and ConsThe Ultimate Guide to Choosing WordPress Pros and Cons
The Ultimate Guide to Choosing WordPress Pros and Cons
 
Moving Beyond Passwords: FIDO Paris Seminar.pdf
Moving Beyond Passwords: FIDO Paris Seminar.pdfMoving Beyond Passwords: FIDO Paris Seminar.pdf
Moving Beyond Passwords: FIDO Paris Seminar.pdf
 
Potential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and InsightsPotential of AI (Generative AI) in Business: Learnings and Insights
Potential of AI (Generative AI) in Business: Learnings and Insights
 
Microsoft 365 Copilot: How to boost your productivity with AI – Part one: Ado...
Microsoft 365 Copilot: How to boost your productivity with AI – Part one: Ado...Microsoft 365 Copilot: How to boost your productivity with AI – Part one: Ado...
Microsoft 365 Copilot: How to boost your productivity with AI – Part one: Ado...
 
A Deep Dive on Passkeys: FIDO Paris Seminar.pptx
A Deep Dive on Passkeys: FIDO Paris Seminar.pptxA Deep Dive on Passkeys: FIDO Paris Seminar.pptx
A Deep Dive on Passkeys: FIDO Paris Seminar.pptx
 
React Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App FrameworkReact Native vs Ionic - The Best Mobile App Framework
React Native vs Ionic - The Best Mobile App Framework
 
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptxThe Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
 
Abdul Kader Baba- Managing Cybersecurity Risks and Compliance Requirements i...
Abdul Kader Baba- Managing Cybersecurity Risks  and Compliance Requirements i...Abdul Kader Baba- Managing Cybersecurity Risks  and Compliance Requirements i...
Abdul Kader Baba- Managing Cybersecurity Risks and Compliance Requirements i...
 
Data governance with Unity Catalog Presentation
Data governance with Unity Catalog PresentationData governance with Unity Catalog Presentation
Data governance with Unity Catalog Presentation
 
Bridging Between CAD & GIS: 6 Ways to Automate Your Data Integration
Bridging Between CAD & GIS:  6 Ways to Automate Your Data IntegrationBridging Between CAD & GIS:  6 Ways to Automate Your Data Integration
Bridging Between CAD & GIS: 6 Ways to Automate Your Data Integration
 
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptxUse of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
 

nnUNet

  • 1. NNUNET A self-configuring method for deep learning-based biomedical image segmentation
  • 2. Introduction Medical Image Segmentation is difficult because of the vast diversity of modalities, each of which requires a specialized pipeline for data pre-processing and training. nnUNet (“no new U-Net”) seeks to establish a standardized pipeline for the medical image segmentation process. 2
  • 3. Problem Statement • Analysis of challenge leaderboard entries shows that using superficially similar methods and model architectures can lead to vastly different results due to implementation details. • Moreover, no single method stands out as being necessary for high scores (except deep learning).
  • 4. Method • nnUNet divides hyper-parameters into 3 types: (1) Fixed configurations (2) Rule-Based configurations (3) Empirical Configurations • In all cases, validation set performance does not affect training time, unlike in AutoML.
  • 5. Fixed Configurations • Model architecture (U-Net), hence the name “no new U-Net”. • Learning rate value and scheduling (0.01 with poly decay). • Optimizer (SGD with Nesterov momentum 0.9). • Training procedure (250k iterations with 5-fold cross-validation and fore-ground over-sampling). • Inference procedure (sliding window with Gaussian importance weighting). 5
  • 6. Rule-Based Configurations • Image Intensity Normalization (Use HU if CT, else use z normalization). • Image Resampling Strategy (If anisotropic, use cubic spline if ratio is within 3, else use nearest neighbor interpolation). • Image Spacing (lowest 10th percentile if anisotropic, else median). • Use 3D cascade (if image is too large). • Model pooling depth (reduce anisotropic side until less than 3, pool until side length becomes 4). • Mini-batch size (largest mini-batch that fits within 11 GB during training). 6
  • 7. Rule-Based Model Configuration • Network Topology, Patch Size, and Batch Size are configured at the start of training for effective training within 11GB of memory. • Networks are expected to give approximate GPU memory usage expectations, which makes implementing new models somewhat cumbersome. • Patch size is given high priority as large patch size is necessary for segmentation. • Network topology is designed to pool until side length is 4 and anisotropic degree is within 3. • Batch size is set to be lower than 5% of total data. Usually set to 1 or 2 as it has lowest priority. 7
  • 8. Empirical Parameters • Only 2 parameters are set empirically, both of which are post-processing steps and therefore do not affect training time. • Suppression of non-largest segmented organ (uses the prior that humans have one or two of each organ). • Ensemble selection of 2D, 3D, and 3D cascade trained networks. These are selected from cross- validation results. 8
  • 9. Comments • Training time and memory requirements do not depend on performance on validation metrics. • No pre-trained networks are necessary as inputs. • No 2.5D, which may be effective in some anisotropic tasks. More robust to anisotropic data while not abandoning the information from the anisotropic direction. • No analysis on the effect of label quality. For example, what is the effect of nnUNet when comparing clean and noisy labels. Is it robust to noisy labels? What attributes are important for learning with sparsely labeled data. 9
  • 10. Results • First place in 33 of 53 challenges with no modifications. • Maintains a high rank in all challenges submitted, though in some cases, modifications to the original nnUNet was necessary (e.g., 2020 COVID19 segmentation challenge). • nnUNet is the baseline for most new medical segmentation challenges.
  • 11. Results • Results for COVID19 segmentation challenge in 2020. • nnUNet features in the top-tier of nearly all medical segmentation challenges. • First place used nnUNet to pseudo- label more data for additional training with nnUNet. 11
  • 12. Code Analysis: Dataset Conversion First convert the data into nifti file format for unified reading. Files must be structured in the specified format for proper training. 12
  • 13. Code Analysis: Training & Inference • Dataset experiment planning and pre-processing: nnUNet_plan_and_preprocess -t XXX --verify_dataset_integrity • Model training: nnUNet_train CONFIGURATION TRAINER_CLASS_NAME TASK_NAME_OR_ID FOLD --npz (additional options) • Find best configuration: nnUNet_find_best_configuration -m 2d 3d_fullres 3d_lowres 3d_cascade_fullres -t XXX --strict • Run inference: nnUNet_predict -i INPUT_FOLDER -o OUTPUT_FOLDER -t TASK_NAME_OR_ID -m CONFIGURATION --save_npz • Run ensemble: nnUNet_ensemble -f FOLDER1 FOLDER2 ... -o OUTPUT_FOLDER -pp POSTPROCESSING_FILE 13
  • 14. Resources • Paper: https://www.nature.com/articles/s41592-020-01008-z (also read the supplementary material) • YouTube videos: https://www.youtube.com/watch?v=3po8qVzz5Tc https://www.youtube.com/watch?v=C6tpnJRpt90 • GitHub: https://github.com/MIC-DKFZ/nnUNet 14
  • 15. Related Work: nnDetection • Paper: https://arxiv.org/abs/2106.00817 • GitHub: https://github.com/MIC-DKFZ/nnDetection 15