SlideShare a Scribd company logo
1 of 56
Artificial Intelligence
in Radiation Oncology
Wookjin Choi, PhD
Assistant Professor of Radiation Oncology
Sidney Kimmel Medical College at Thomas Jefferson University
Wookjin.Choi@Jefferson.edu
Mar 11, 2022 @ Mayo Clinic
Acknowledgements
Memorial Sloan Kettering Cancer Center
ā€¢ Wei Lu PhD
ā€¢ Sadegh Riyahi, PhD
ā€¢ Jung Hun Oh, PhD
ā€¢ Saad Nadeem, PhD
ā€¢ Eric Aliotta, PhD
ā€¢ Joseph O. Deasy, PhD
ā€¢ Andreas Rimner, MD
ā€¢ Prasad Adusumilli, MD
Stony Brook University
ā€¢ Allen Tannenbaum, PhD
University of Virginia School of Medicine
ā€¢ Jeffrey Siebers, PhD
ā€¢ Victor Gabriel Leandro Alves, PhD
University of Maryland School of Medicine
ā€¢ Howard Zhang, PhD
ā€¢ Wengen Chen, MD, PhD
ā€¢ Charles White, MD
Thomas Jefferson University
ā€¢ Yevgeniy Vinogradskiy, PhD
ā€¢ Hamidreza Nourzadeh, PhD
ā€¢ Adam P. Dicker, MD
2
NIH/NCI Grant R01 CA222216, R01 CA172638 and
NIH/NCI Cancer Center Support Grant P30 CA008748 and 5P30 CA056036
The ESTRO Falcon project team, Scott Kaylor of EduCase, Benjamin Nelms of Proknow for the multi-
delineator contour data presented in this work
AI in Radiation Oncology
3
Huynh et al. Nat Rev Clin Oncol 2020
4
Netherton et al. Oncology 2021
Hype cycle for three major innovations
in radiation oncology
Automatable tasks in radiation oncology
for the modern clinic
Outline
ā€¢ Radiomics - Decision Support Tools
- Lung Cancer Screening
- Tumor Response Prediction and Evaluation
- Aggressive Lung ADC subtype prediction
- Multimodal data: Pathology, Multiomics, etc.
ā€¢ Auto Delineation and Variability Analysis
- Delineation Variability Quantification
- Dosimetric Consequences of Variabilities
- OARNet, Voxel2Mesh
5
Radiomics
6
ĀØ Controllable Feature Analysis
ĀØ More Interpretable
Lambin, et al. Eur J Cancer 2012
Aerts et al., Nature Communications, 2014
Radiomics Framework
7
Image
Registration
ā€¢ Multi-level rigid
ā€¢ Deformable
ā€¢ Pre/Post-CT
ā€¢ MSE, MI
Tumor
Segmentation
ā€¢ Adaptive region growing
ā€¢ Level set
ā€¢ Grow cut
ā€¢ Morphology filter
ā€¢ Multi-modality
image segmentation
Feature
Extraction
ā€¢ Intensity distribution
ā€¢ Spatial variations
(texture)
ā€¢ Geometric properties
ā€¢ Jacobian feature
from DVF
ā€¢ Feature selection
Predictive
Model
ā€¢ ROC analyses
ā€¢ Prediction models
ā€¢ Validation
ā€¢ Tumor response
ā€¢ Recurrence
ā€¢ Survival
Source codes: https://github.com/taznux/radiomics-tools
ā€¢ Automated Workflow
- Integrate all the radiomics components
- 3D Slicer, ITK (C++), Matlab, R, and Python
- Scalable: support multicore & GPU computing
Radiomics Framework
8
Image
Registration
ā€¢ Multi-level rigid
ā€¢ Deformable
ā€¢ Pre/Post-CT
ā€¢ MSE, MI
Tumor
Segmentation
ā€¢ Adaptive region growing
ā€¢ Level set
ā€¢ Grow cut
ā€¢ Morphology filter
ā€¢ Multi-modality
image segmentation
Feature
Extraction
ā€¢ Intensity distribution
ā€¢ Spatial variations
(texture)
ā€¢ Geometric properties
ā€¢ Jacobian feature
from DVF
ā€¢ Feature selection
Predictive
Model
ā€¢ ROC analyses
ā€¢ Prediction models
ā€¢ Validation
ā€¢ Tumor response
ā€¢ Recurrence
ā€¢ Survival
Source codes: https://github.com/taznux/radiomics-tools
Deep Learning Model
ā€¢ Automated Workflow
- Integrate all the radiomics components
- 3D Slicer, ITK (C++), Matlab, R, and Python
- Scalable: support multicore & GPU computing
Radiomics Framework
9
Image
Registration
ā€¢ Multi-level rigid
ā€¢ Deformable
ā€¢ Pre/Post-CT
ā€¢ MSE, MI
Tumor
Segmentation
ā€¢ Adaptive region growing
ā€¢ Level set
ā€¢ Grow cut
ā€¢ Morphology filter
ā€¢ Multi-modality
image segmentation
Feature
Extraction
ā€¢ Intensity distribution
ā€¢ Spatial variations
(texture)
ā€¢ Geometric properties
ā€¢ Jacobian feature
from DVF
ā€¢ Feature selection
Predictive
Model
ā€¢ ROC analyses
ā€¢ Prediction models
ā€¢ Validation
ā€¢ Tumor response
ā€¢ Recurrence
ā€¢ Survival
Source codes: https://github.com/taznux/radiomics-tools
Deep Learning Model
ā€¢ Automated Workflow
- Integrate all the radiomics components
- 3D Slicer, ITK (C++), Matlab, R, and Python
- Scalable: support multicore & GPU computing
Radiomics Framework
10
Image
Registration
ā€¢ Multi-level rigid
ā€¢ Deformable
ā€¢ Pre/Post-CT
ā€¢ MSE, MI
Tumor
Segmentation
ā€¢ Adaptive region growing
ā€¢ Level set
ā€¢ Grow cut
ā€¢ Morphology filter
ā€¢ Multi-modality
image segmentation
Feature
Extraction
ā€¢ Intensity distribution
ā€¢ Spatial variations
(texture)
ā€¢ Geometric properties
ā€¢ Jacobian feature
from DVF
ā€¢ Feature selection
Predictive
Model
ā€¢ ROC analyses
ā€¢ Prediction models
ā€¢ Validation
ā€¢ Tumor response
ā€¢ Recurrence
ā€¢ Survival
Source codes: https://github.com/taznux/radiomics-tools
Deep Learning Model
ā€¢ Automated Workflow
- Integrate all the radiomics components
- 3D Slicer, ITK (C++), Matlab, R, and Python
- Scalable: support multicore & GPU computing
Radiomics Framework
11
Image
Registration
ā€¢ Multi-level rigid
ā€¢ Deformable
ā€¢ Pre/Post-CT
ā€¢ MSE, MI
Tumor
Segmentation
ā€¢ Adaptive region growing
ā€¢ Level set
ā€¢ Grow cut
ā€¢ Morphology filter
ā€¢ Multi-modality
image segmentation
Feature
Extraction
ā€¢ Intensity distribution
ā€¢ Spatial variations
(texture)
ā€¢ Geometric properties
ā€¢ Jacobian feature
from DVF
ā€¢ Feature selection
Predictive
Model
ā€¢ ROC analyses
ā€¢ Prediction models
ā€¢ Validation
ā€¢ Tumor response
ā€¢ Recurrence
ā€¢ Survival
Source codes: https://github.com/taznux/radiomics-tools
Deep Learning Model
ā€¢ Automated Workflow
- Integrate all the radiomics components
- 3D Slicer, ITK (C++), Matlab, R, and Python
- Scalable: support multicore & GPU computing
Radiomics Framework
12
Image
Registration
ā€¢ Deep Learning Optical fl
ow
ā€¢ Action like flow
ā€¢ Differential warp
ā€¢ Dynamic filtering
ā€¢ ā€¦
Tumor
Segmentation
ā€¢ U-Net
ā€¢ Prob. U-Net
ā€¢ UANet
ā€¢ ā€¦
Feature
Extraction
ā€¢ AlexNet
ā€¢ ResNet
ā€¢ VGG
ā€¢ LeNet
ā€¢ ā€¦.
Predictive
Model
ā€¢ ROC analyses
ā€¢ Prediction models
ā€¢ Validation
ā€¢ Tumor response
ā€¢ Recurrence
ā€¢ Survival
ā€¢ Automated Workflow
- Integrate all the radiomics components
- 3D Slicer, ITK (C++), Matlab, R, and Python
- Scalable: support multicore & GPU computing
Lung Cancer Screening
13
ĀØ Early detection of lung cancer by LDCT can reduce mortality
ĀØ Known features correlated with PN malignancy
Ā¤ Size, growth rate (Lung-RADS)
Ā¤ Calcification, enhancement, solidity ā†’ texture features
Ā¤ Boundary margins (spiculation, lobulation), attachment ā†’ shape and
appearance features
Malignant nodules Benign nodules
Size Total Malignancy
< 4mm 2038 0%
4-7 mm 1034 1%
8-20 mm 268 15%
> 20 mm 16 75%
ACR Lung-RADS 1.0
Category Baseline Screening Malignancy
1 No PNs; PNs with calcification
Negative
<1% chance of malignancy
2
Solid/part-solid: <6 mm
GGN: <20 mm
Benign appearance
<1% chance of malignancy
3
Solid: ā‰„6 to <8 mm
Part-solid: ā‰„6 mm with solid component <6 mm
GGN: ā‰„20 mm
Probably benign
1-2% chance of malignancy
4A
Solid: ā‰„8 to <15 mm
Part-solid: ā‰„8 mm with solid component ā‰„6 and <8 mm
Suspicious
5-15% chance of malignancy
4B
Solid: ā‰„15 mm
Part-solid: Solid component ā‰„8 mm
>15% chance of malignancy
4X
Category 3 or 4 PNs with suspicious features (e.g., enlarged lymph
nodes) or suspicious imaging findings (e.g., spiculation)
>15% chance of malignancy
14
ACR: American College of Radiology
Lung-RADS: Lung CT Screening Reporting and Data System
Lung Cancer Screening (Methodology)
ā€¢ TCIA LIDC-IDRI public data set (n=1,010)
- Multi-institutional data
- 72 cases evaluated (31 benign and 41 malignant cases)
ā€¢ Consensus contour
15
GLCM GLRM
Texture features Intensity features
2D
Shape features
3D
Lung Cancer Screening (SVM-LASSO Model )
16
SVM classification
Distinctive feature identification
Malignant?
Predicted malignancy
Feature extraction
Yes
10x10-fold
CV
10-fold
CV
LASSO feature selection
ā€¢ Size (BB_AP) : Highly correlated with the axial longest diameter
and its perpendicular diameter (r = 0.96, larger ā€“ more
malignant)
ā€¢ Texture (SD_IDM) : Tumor heterogeneity (smaller ā€“ more
malignant)
Lung Screening (Results: Comparison)
Sensitivity Specificity Accuracy AUC
Lung-RADS
Clinical guideline
73.3% 70.4% 72.2% 0.74
Hawkins et al. (2016)
Radiomics ā€“ 23 features
51.7 % 92.9% 80.0% 0.83
Ma et al. (2016)
Radiomics ā€“ 583 features
80.0% 85.5% 82.7%
Buty et al. (2016)
DL ā€“ 400 SH and 4096 AlexNet features
82.4%
Kumar et al. (2015)
DL: 5000 features
79.1% 76.1% 77.5%
Proposed
Radiomics: two features (Size and Texture)
87.2% 81.2% 84.6% 0.89
17
DL: Deep Learning, SH: Spherical Harmonics
Choi et al., Medical Physics, 2018.
ACR Lung-RADS 1.0
Category Baseline Screening Malignancy
1 No PNs; PNs with calcification
Negative
<1% chance of malignancy
2
Solid/part-solid: <6 mm
GGN: <20 mm
Benign appearance
<1% chance of malignancy
3
Solid: ā‰„6 to <8 mm
Part-solid: ā‰„6 mm with solid component <6 mm
GGN: ā‰„20 mm
Probably benign
1-2% chance of malignancy
4A
Solid: ā‰„8 to <15 mm
Part-solid: ā‰„8 mm with solid component ā‰„6 and <8 mm
Suspicious
5-15% chance of malignancy
4B
Solid: ā‰„15 mm
Part-solid: Solid component ā‰„8 mm
>15% chance of malignancy
4X
Category 3 or 4 PNs with suspicious features (e.g., enlarged lymph
nodes) or suspicious imaging findings (e.g., spiculation)
>15% chance of malignancy
18
ACR: American College of Radiology
Lung-RADS: Lung CT Screening Reporting and Data System
Spiculation Quantification (Motivation)
ā€¢ Semantic Features
ā€¢ Semi-automatic Segmentation
- GrowCut and LevelSet
19
Radiologists spiculation score (RS) for different pulmonary nodules
1 2 3 4 5
Choi et al. in CMPB 2021
Spiculation Quantification (Methodology)
20
!!: = log
āˆ‘",$ ( ([+(,!), +(,"), +(,$)])
āˆ‘",$ ( ([,!, ,", ,$])
Area Distortion Map
Spherical Mapping
Eigenfunction
Spiculation Quantification (Results)
21
Number of Spiculations and Radiologists spiculation score (RS) for different pulmonary nodules
1 2 3 4 5
Choi et al. in CMPB 2021
0 1 4 8 14
Spiculation Quantification (Model Validation)
22
Spiculation Quantification (Results: Comparison)
23
Choi et al. in CMPB 2021
Progression-free survival Prediction
after SBRT for early-stage NSCLC
24
Thor, Choi et al. ASTRO 2020
ā€¢ 412 patients treated between 2006 and 2017
ā€¢ PETs and CTs within three months prior to SBRT start.
ā€¢ The median prescription dose was 50Gy in 5 fractions.
Progression-free survival Prediction (Results)
ā€¢ PET entropy, CT number of peaks,
CT major axis, and gender.
ā€¢ The most frequently selected model
included PET entropy and CT
number of peaks
- The c-index in the validation
subset was 0.77
- The prediction-stratified survival
indicated a clear separation
between the observed HR and
LR
- e.g. a PFS of 60% was observed
at 12 months in HR vs. 22
months in LR.
25
Thor, Choi et al. ASTRO 2020
Local tumor morphological changes
26
Jacobian Map
- Jacobian matrix: calculates rate of displacement change in each direction.
- Determinant indicates volumetric ratio of shrinkage/expansion.
012 3 = 4
012 3 > 1 volume expansion
012 3 = 1 no volume change
012 3 < 1 volume shrinkage
012 3 = 1.2 = 20% expansion
012 3 = 0.8 = 20% shrinkage (-20%)
Riyahi, Choi et al., PMB 2018
Local tumor morphological changes (Results)
27
Local tumor morphological changes (Results)
Features P-value AUC Correlation to responders
Minimum Jacobian 0.009 0.98 -0.79
Median Jacobian 0.046 0.95 -0.72
The P-value, AUC and correlation to responders for all significant features in univariate analysis
28
Riyahi, Choi et al., PMB 2018
SVM-LASSO: AUC 0.91
Local Metabolic Tumor Volume Changes
29
Riyahi, Choi et al., DATRA@MICCAI 2018 AUC=0.81
Aggressive Lung ADC Subtype Prediction (Motivation)
30
CT
MIP
PET/CT
Soild
CT PET/CT
Five classifications of lung ADC Travis et
al. JTO 2011
ĀØ Solid and MIP components: poor surgery/SBRT prognosis factor
Ā¤ Benefit from lobectomy rather than limited resection
ĀØ Core biopsy (Leeman et al. IJROBP 2017)
Ā¤ Minimally invasive, not routinely performed, sampling error (about 60%
agreement with pathology)
ĀØ Preoperative diagnostic CT and FDG PET/CT radiomics
Ā¤ Non-invasive and routinely performed
Aggressive Lung ADC Subtype Prediction (Method & Results)
ā€¢ Retrospectively enrolled 120 patients
- Stage I lung ADC, ā‰¤2cm
- Preoperative diagnostic CT and FDG PET/CT
ā€¢ Histopathologic endpoint
- Aggressiveness (Solid : 18 cases, MIP : 5 cases)
ā€¢ 206 radiomic features & 14 clinical parameters
ā€¢ SVM-LASSO model
31
Performance of the SVM-LASSO model to predict aggressive lung ADC
Choi et al. Manuscript under review
Box plots of SUVmax (FDR q=0.004) and PET Mean of
Cluster Shade (q=0.002)
Feature Sensitivity Specificity PPV NPV Accuracy AUC
Conventional SUVmax 57.8Ā±4.6% 78.5Ā±1.4% 39.2Ā±2.3% 88.6Ā±1.1% 74.5Ā±1.4% 0.64Ā±0.01
SVM-LASSO PET Mean of Cluster Shade 67.4Ā±3.1% 86.0Ā±1.1% 53.7Ā±2.1% 91.7Ā±1.0% 82.4Ā±1.0% 0.78Ā±0.01
p-value SUVmax vs. SVM-LASSO 0.002 1e-5 7e-8 3e-5 5e-8 0.03
Unsupervised Learning of Deep Learned Features
from Breast Cancer Images
32
Lee, Choi et al. IEEE BIBE 2020
Unsupervised Learning of Deep Learned Features
33
Slide name
Silhouette
optimal
number
Cluster set Accuracy F1-score
TCGA-A7-A0DA 29
[25, 22, 6,
2, 24, 14,
0, 20, 10]
0.8829 0.8929
TCGA-A2-A0YM 20 [7, 6, 1, 5] 0.8360 0.8863
TCGA-A2-A3XT 19
[13, 10, 5,
1, 2]
0.9316 0.9514
TCGA-BH-A0BG 8 [1, 5] 0.7828 0.6857
TCGA-E2-A1LS 26 [18, 5] 0.8495 0.8680
TCGA-OL-A66I 25
[7, 18, 12,
20, 1, 0]
0.7761 0.7091
TCGA-C8-A26Y 21
[9, 18, 12,
16]
0.8594 0.8122
Lee, Choi et al. IEEE BIBE 2020
PathCNN: interpretable convolutional neural networks
for survival prediction and pathway analysis applied to glioblastoma
34
ā€¢ CNNs have achieved great success
ā€¢ A lack of interpretability remains a
key barrier
ā€¢ Moreover, because biological array
data are generally represented in a
non-grid structured format
ā€¢ PathCNN
An interpretable CNN model on
integrated multi-omics data using a
newly defined pathway image.
Oh, Choi et al. Bioinformatics 2021, joint first author Source code: https://github.com/mskspi/PathCNN.
PathCNN: interpretable CNNs (results)
35
Cancer PathCNN Logistic
regression SVM with RBF Neural network MiNet
GBM 0.755ā€‰Ā±ā€‰0.009 0.668ā€‰Ā±ā€‰0.039 0.685ā€‰Ā±ā€‰0.037 0.692ā€‰Ā±ā€‰0.030 0.690ā€‰Ā±ā€‰0.032
LGG 0.877ā€‰Ā±ā€‰0.007 0.816ā€‰Ā±ā€‰0.036 0.884ā€‰Ā±ā€‰0.017 0.791ā€‰Ā±ā€‰0.031 0.854ā€‰Ā±ā€‰0.027
LUAD 0.637ā€‰Ā±ā€‰0.014 0.581ā€‰Ā±ā€‰0.028 0.624ā€‰Ā±ā€‰0.034 0.573ā€‰Ā±ā€‰0.031 0.597ā€‰Ā±ā€‰0.042
KIRC 0.709ā€‰Ā±ā€‰0.009 0.654ā€‰Ā±ā€‰0.034 0.684ā€‰Ā±ā€‰0.027 0.702ā€‰Ā±ā€‰0.028 0.659ā€‰Ā±ā€‰0.030
Comparison of predictive performance with benchmark methods in terms of the area
under the curve (AUC: mean Ā± standard deviation) over 30 iterations of the 5-fold
cross validation
Note: AUCs for PathCNN were obtained with three principal components. Bold = Highest AUC for each dataset.
SVM, support vector machine; RBF, radial basis function; MiNet, Multi-omics Integrative Net; GBM, glioblastoma
multiforme; LGG, low-grade glioma; LUAD, lung adenocarcinoma; KIRC, kidney cancer.
Oh, Choi et al. Bioinformatics 2021, joint first author Source code: https://github.com/mskspi/PathCNN.
36
A matrix of adjusted P-values. The row represents the 146 KEGG pathways ordered on pathway images.
The columns represent the first two principal components of each omics type. The red color indicates
key pathways with adjusted P-values < 0.001
Outline
ā€¢ Radiomics - Decision Support Tools
- Lung Cancer Screening
- Tumor Response Prediction and Evaluation
- Aggressive Lung ADC subtype prediction
- Multimodal data: Pathology, Multiomics, etc.
ā€¢ Auto Delineation and Variability Analysis
- Delineation Variability Quantification
- Dosimetric Consequences of Variabilities
- OARNet, Voxel2Mesh
37
Delineation Variability Quantification and Simulation
A framework for radiation therapy variability analysis
38
RT plan
Structure Set
CT Image
Dose Distribution
Structure Sets
DV simulation
ASSD
GrowCut
RW
Other delineators
SV analysis
DV analysis
Geometric
Dosimetric
Variability analysis
Human DV
Simulated
DV
Consensus SS
OARNet
Choi et al., AAPM, 2019.
Delineation Variability Quantification and
Simulation
ā€¢ ESTRO Falcon contour workshop (EduCase)
- A HNC case, Larynx, 70 Gy and 35 fractions
- 14 independent manually delineated (MD) OAR structure sets (SS)
- BrainStem, Esophagus, OralCavity, Parotid_L, Parotid_R, SpinalCord,
and Thyroid
ā€¢ Consensus MD SS
- The simultaneous truth and performance level estimation (STAPLE)
39
Choi, Nourzadeh et al., AAPM, 2019.
Delineation Variability Quantification and
Simulation (Methods)
ā€¢ Geometric analysis
- Similarity: Dice coefficient (Volumetric, Surface)
- Distance: Hausdorff distance (HD), Actual Average Surface
Distance (AASD)
- Reference: STAPLE SS
ā€¢ Dosimetric analysis
- Single dose distribution planned from a human SS
- DVH confidence bands (90%tile)
- !!"#$, !!#%, !!&$, !'(
40
Choi, Nourzadeh et al., AAPM, 2019.
Delineation Variability Quantification and Simulation (Results)
ā€¢ DVH variability not predicted by geometric measures
ā€¢ Large human variability
41
100%
50%
0%
100%
50%
0%
Human ASSD GrowCut RW
Right
Parotid
Left
Parotid
Choi, Nourzadeh et al., AAPM, 2019.
Human ASSD GrowCut RW
100%
50%
0%
Education
100%
50%
0%
Clinic
42
ā€¢ Plan Competition Data Set: A HNC case, Nasopharynx , 70 Gy
ā€¢ PV: 409 plans for IMRT, VMAT, and Tomotherapy on various TPS
Eclipse, Monaco , Pinnacle, RayStation, Tomotherapy
ā€¢ SV: Setup error simulation using Radiation Therapy Robustness
Analyzer (RTRA)
- 1000 simulations: 3mm translation and 5-degree rotation
ā€¢ DV: 5 manually delineated (MD) SSs
- BrainStem, Esophagus, OralCavity, Parotid_L, Parotid_R, and
SpinalCord
- 3x5 Simulated DVs using Radiation Therapy Variability Analyzer (RTVA)
Variability Analysis
Plan Variability, Setup Variability, Delineation Variability
43
Choi, Nourzadeh et al., AAPM, 2020.
Dose Volume Coverage Map (DVCM)
44
Plan Variability Setup Variability Delineation Variability
Choi, Nourzadeh et al., AAPM, 2020.
DVCM Analysis
45
Fractional
Volume
Dose
OAR Dose Constraint
Probability
Threshold
5%
0.3
DVCM Summary
46
TV SVƗPV SVƗDV PVƗDV SV PV DV Average
BrainStem 0.15 0.11 0.10 0.07 0.03 0.01 0.02 0.07 >0.95
Chiasm 0.68 0.00 0.65 0.56 0.00 0.00 0.49 0.34
Eye_L 0.07 0.05 0.04 0.01 0.02 0.01 0.01 0.03
Eye_R 0.02 0.01 0.01 0.01 0.00 0.00 0.01 0.01 >0.5
Lens_L 1.00 1.00 1.00 1.00 0.99 1.00 0.01 0.86
Lens_R 1.00 1.00 1.00 1.00 1.00 1.00 0.01 0.86 >0.3
Mandible 0.04 0.04 0.00 0.07 0.00 0.08 0.01 0.03 >0.2
OpticNerve_L 0.51 0.42 0.43 0.36 0.17 0.00 0.34 0.32 >0.05
OpticNerve_R 0.56 0.38 0.61 0.38 0.20 0.00 0.39 0.36 ā‰¤0.05
SpinalCord 0.13 0.12 0.05 0.05 0.04 0.01 0.02 0.06 =0.00
Average 0.42 0.31 0.39 0.35 0.25 0.21 0.13 0.29
Fractional volume affected by different variations when constraint failing probability > 5% (worst-case scenario)
Choi, Nourzadeh et al., AAPM, 2020.
OARNet: auto-delineate organs-at-risk (OARs) in
head and neck (H&N) CT image
47
Soomro, Nourzadeh, Choi et al., arXiv:2108.13987.
OARNet results
48
Soomro, Nourzadeh, Choi et al., arXiv:2108.13987.
Interpretable Radiomics Toolkit
End-to-End Deep Learning Model for Malignancy Prediction
49
Input Ground Truth Voxel2Mesh
Choi et al., Manuscript in Preparation
Network AUC Accuracy Sensitivity Specificity F1
LIDC-PM Mesh Only 0.937 83.33 77.78 88.89 82.35
Mesh+Encoder 0.903 88.89 91.67 86.11 89.19
LUNGx Mesh Only 0.711 63.33 73.33 53.33 66.67
Mesh+Encoder 0.687 53.3 83.3 23.33 64.11
Summary
ā€¢ Radiomics - Decision Support Tools
- Lung Cancer Screening
- Tumor Response Prediction and Evaluation
- Aggressive Lung ADC subtype prediction
- Multimodal data: Pathology, Multiomics, etc.
ā€¢ Auto Delineation and Variability Analysis
- Delineation Variability Quantification
- Dosimetric Consequences of Variabilities
50
Short-term Future Works
ā€¢ Develop interpretable radiomic features
- Improve spiculation quantification and multi-institution validation
- Multimodal data integration
ā€¢ Human-Variability aware auto-delineation
- Variability quantification and simulation using generative models
- AI-guided interactive delineation editing
ā€¢ Integrate the radiomics framework into TPS
- Eclipse (C#) and MIM (Python)
51
Long-term Future Works
ā€¢ Comprehensive Framework for Cancer Imaging
- Multi-modal imaging
- Response prediction and evaluation (Pre, Mid, and Post)
- Longitudinal analysis of tumor change during treatment (MRgRT)
- Shape analysis (e.g., Spiculation)
- Deep learning models
ā€¢ Automation of Clinical Workflow
- Big Data Analytics: EMR, PACS, ROIS, Genomics, etc.
- Provide an informatics platform for comprehensive cancer therapy
53
Selected Publications
1. Jung Hun Oh*, Wookjin Choi* et al., ā€œPathCNN: interpretable convolutional neural networks for survival
prediction and pathway analysis applied to glioblastomaā€, Bioinformatics, 2021, *joint first author
2. Wookjin Choi et al., ā€œ Reproducible and Interpretable Spiculation Quantification for Lung Cancer Screeningā€,
Computer Methods and Programs in Biomedicineā€, 2021
3. Noemi Garau, Wookjin Choi, et al., ā€œ External validation of radiomics-based predictive models in low-dose CT
screening for early lung cancer diagnosisā€, Medical Physics, 2020
4. Jiahui Wang, Wookjin Choi et al., ā€œPrediction of anal cancer recurrence after chemoradiotherapy using
quantitative image features extracted from serial 18F-FDG PET/CTā€, Frontiers in oncology, 2019
5. Wookjin Choi et al., ā€œRadiomics analysis of pulmonary nodules in low-dose CT for early detection of lung cancerā€,
Medical Physics, 2018
6. Sadegh Riyahi, Wookjin Choi, et al., ā€œQuantifying local tumor morphological changes with Jacobian map for
prediction of pathologic tumor response to chemo-radiotherapy in locally advanced esophageal cancerā€, Physics
in Medicine and Biology, 2018
7. Shan Tan, Laquan Li, Wookjin Choi, et al., ā€œAdaptive region-growing with maximum curvature strategy for tumor
segmentation in 18F-FDG PETā€, Physics in Medicine and Biology, 2017
8. Wookjin Choi et al., ā€œIndividually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in Pancreatic
Ductal Adenocarcinomaā€, Medical Physics, 2016
9. Wookjin Choi, Tae-Sun Choi, ā€œAutomated Pulmonary Nodule Detection based on Three-dimensional Shape-based
Feature Descriptorā€, Computer Methods and Programs in Biomedicine, 2014
54
Complete list of publications: https://scholar.google.com/citations?user=iHgsGLUAAAAJ
Post-Doctoral Research Fellow
Developing Interpretable Predictive Models for Radiation
Therapy
ā€¢ PI: Wookjin Choi, PhD - Wookjin.Choi@Jefferson.edu
ā€¢ 2 Years
ā€¢ Machine Learning/Deep Learning: Radiomics (PET/CT & MR) and Bioinformatics
ā€¢ Computational Medical Physics: Development of Predictive Models and
Automated Workflows, and Improve Clinical Workflow
ā€¢ Internal or Extramural Research Funding Opportunities
Qualifications
ā€¢ Ph.D. in Computer Science, Electrical Engineering, Medical Physics, or related
field required
Thank You!
Q & A
https://quradiomics.com
Wookjin.Choi@Jefferson.edu

More Related Content

What's hot

Rrecent advances in linear accelerators [MR linac]
Rrecent advances in linear accelerators [MR linac]Rrecent advances in linear accelerators [MR linac]
Rrecent advances in linear accelerators [MR linac]Upasna Saxena
Ā 
How Radiation Therapy is Used to Treat Soft Tissue Sarcoma
How Radiation Therapy is Used to Treat Soft Tissue SarcomaHow Radiation Therapy is Used to Treat Soft Tissue Sarcoma
How Radiation Therapy is Used to Treat Soft Tissue SarcomaDana-Farber Cancer Institute
Ā 
Intra Operative Radiotherapy
Intra Operative RadiotherapyIntra Operative Radiotherapy
Intra Operative RadiotherapySasikumar Sambasivam
Ā 
Radiotherapy sarcomas
Radiotherapy sarcomas Radiotherapy sarcomas
Radiotherapy sarcomas Ashutosh Mukherji
Ā 
RE-IRRADIATION IN HEAD AND NECK CANCER
RE-IRRADIATION IN HEAD AND NECK CANCERRE-IRRADIATION IN HEAD AND NECK CANCER
RE-IRRADIATION IN HEAD AND NECK CANCERMUNEER khalam
Ā 
University of Toronto - Radiomics for Oncology - 2017
University of Toronto  - Radiomics for Oncology - 2017University of Toronto  - Radiomics for Oncology - 2017
University of Toronto - Radiomics for Oncology - 2017Andre Dekker
Ā 
Radiation for Lung Cancer
Radiation for Lung CancerRadiation for Lung Cancer
Radiation for Lung CancerRobert J Miller MD
Ā 
Tumor board locally advanced rectal cancer
Tumor board locally advanced rectal cancerTumor board locally advanced rectal cancer
Tumor board locally advanced rectal cancerRanjita Pallavi
Ā 
FAST Forward Trial breast cancer
FAST Forward Trial breast cancerFAST Forward Trial breast cancer
FAST Forward Trial breast cancerKanhu Charan
Ā 
Head and neck reirradiation
Head and neck reirradiationHead and neck reirradiation
Head and neck reirradiationKanhu Charan
Ā 
Artificial Intelligence in Radiation Oncology
Artificial Intelligence in Radiation OncologyArtificial Intelligence in Radiation Oncology
Artificial Intelligence in Radiation OncologyWookjin Choi
Ā 
Summary of embrace protocol
Summary of embrace protocolSummary of embrace protocol
Summary of embrace protocolDr. Ankita Pandey
Ā 
Radiotherapy planning in carcinoma cervix dr rekha
Radiotherapy planning in carcinoma cervix dr rekhaRadiotherapy planning in carcinoma cervix dr rekha
Radiotherapy planning in carcinoma cervix dr rekhaDr Rekha Arya
Ā 
Induction chemotherapy followed by concurrent ct rt versus ct-rt in advanced ...
Induction chemotherapy followed by concurrent ct rt versus ct-rt in advanced ...Induction chemotherapy followed by concurrent ct rt versus ct-rt in advanced ...
Induction chemotherapy followed by concurrent ct rt versus ct-rt in advanced ...Santam Chakraborty
Ā 
Radiotherapy techniques for Breast Cancer
Radiotherapy techniques for Breast CancerRadiotherapy techniques for Breast Cancer
Radiotherapy techniques for Breast CancerAnimesh Agrawal
Ā 
Adaptive radiotherapy in head and neck cancer
Adaptive radiotherapy in head and neck cancerAdaptive radiotherapy in head and neck cancer
Adaptive radiotherapy in head and neck cancerDr. Rituparna Biswas
Ā 
RADIOTHERAPY IN CARCINOMA OVARY
RADIOTHERAPY IN CARCINOMA OVARYRADIOTHERAPY IN CARCINOMA OVARY
RADIOTHERAPY IN CARCINOMA OVARYDR DEBASHIS PANDA
Ā 
Plan evaluation in Radiotherapy- Dr Kiran
Plan evaluation in Radiotherapy- Dr KiranPlan evaluation in Radiotherapy- Dr Kiran
Plan evaluation in Radiotherapy- Dr KiranKiran Ramakrishna
Ā 
Contouring in breast cancer current practice and future directions
Contouring in breast cancer current practice and future directions Contouring in breast cancer current practice and future directions
Contouring in breast cancer current practice and future directions Anil Gupta
Ā 

What's hot (20)

Rrecent advances in linear accelerators [MR linac]
Rrecent advances in linear accelerators [MR linac]Rrecent advances in linear accelerators [MR linac]
Rrecent advances in linear accelerators [MR linac]
Ā 
How Radiation Therapy is Used to Treat Soft Tissue Sarcoma
How Radiation Therapy is Used to Treat Soft Tissue SarcomaHow Radiation Therapy is Used to Treat Soft Tissue Sarcoma
How Radiation Therapy is Used to Treat Soft Tissue Sarcoma
Ā 
Intra Operative Radiotherapy
Intra Operative RadiotherapyIntra Operative Radiotherapy
Intra Operative Radiotherapy
Ā 
Radiotherapy sarcomas
Radiotherapy sarcomas Radiotherapy sarcomas
Radiotherapy sarcomas
Ā 
RE-IRRADIATION IN HEAD AND NECK CANCER
RE-IRRADIATION IN HEAD AND NECK CANCERRE-IRRADIATION IN HEAD AND NECK CANCER
RE-IRRADIATION IN HEAD AND NECK CANCER
Ā 
University of Toronto - Radiomics for Oncology - 2017
University of Toronto  - Radiomics for Oncology - 2017University of Toronto  - Radiomics for Oncology - 2017
University of Toronto - Radiomics for Oncology - 2017
Ā 
Radiation for Lung Cancer
Radiation for Lung CancerRadiation for Lung Cancer
Radiation for Lung Cancer
Ā 
Tumor board locally advanced rectal cancer
Tumor board locally advanced rectal cancerTumor board locally advanced rectal cancer
Tumor board locally advanced rectal cancer
Ā 
FAST Forward Trial breast cancer
FAST Forward Trial breast cancerFAST Forward Trial breast cancer
FAST Forward Trial breast cancer
Ā 
Head and neck reirradiation
Head and neck reirradiationHead and neck reirradiation
Head and neck reirradiation
Ā 
Artificial Intelligence in Radiation Oncology
Artificial Intelligence in Radiation OncologyArtificial Intelligence in Radiation Oncology
Artificial Intelligence in Radiation Oncology
Ā 
Summary of embrace protocol
Summary of embrace protocolSummary of embrace protocol
Summary of embrace protocol
Ā 
Icru 38
Icru   38Icru   38
Icru 38
Ā 
Radiotherapy planning in carcinoma cervix dr rekha
Radiotherapy planning in carcinoma cervix dr rekhaRadiotherapy planning in carcinoma cervix dr rekha
Radiotherapy planning in carcinoma cervix dr rekha
Ā 
Induction chemotherapy followed by concurrent ct rt versus ct-rt in advanced ...
Induction chemotherapy followed by concurrent ct rt versus ct-rt in advanced ...Induction chemotherapy followed by concurrent ct rt versus ct-rt in advanced ...
Induction chemotherapy followed by concurrent ct rt versus ct-rt in advanced ...
Ā 
Radiotherapy techniques for Breast Cancer
Radiotherapy techniques for Breast CancerRadiotherapy techniques for Breast Cancer
Radiotherapy techniques for Breast Cancer
Ā 
Adaptive radiotherapy in head and neck cancer
Adaptive radiotherapy in head and neck cancerAdaptive radiotherapy in head and neck cancer
Adaptive radiotherapy in head and neck cancer
Ā 
RADIOTHERAPY IN CARCINOMA OVARY
RADIOTHERAPY IN CARCINOMA OVARYRADIOTHERAPY IN CARCINOMA OVARY
RADIOTHERAPY IN CARCINOMA OVARY
Ā 
Plan evaluation in Radiotherapy- Dr Kiran
Plan evaluation in Radiotherapy- Dr KiranPlan evaluation in Radiotherapy- Dr Kiran
Plan evaluation in Radiotherapy- Dr Kiran
Ā 
Contouring in breast cancer current practice and future directions
Contouring in breast cancer current practice and future directions Contouring in breast cancer current practice and future directions
Contouring in breast cancer current practice and future directions
Ā 

Similar to Artificial Intelligence in Radiation Oncology

Quantitative Cancer Image Analysis
Quantitative Cancer Image AnalysisQuantitative Cancer Image Analysis
Quantitative Cancer Image AnalysisWookjin Choi
Ā 
Radiomics Analysis of Pulmonary Nodules in Low Dose CT for Early Detection of...
Radiomics Analysis of Pulmonary Nodules in Low Dose CT for Early Detection of...Radiomics Analysis of Pulmonary Nodules in Low Dose CT for Early Detection of...
Radiomics Analysis of Pulmonary Nodules in Low Dose CT for Early Detection of...Wookjin Choi
Ā 
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRIProstate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRISaifeng (Aaron) Liu
Ā 
Radiomics and Deep Learning for Lung Cancer Screening
Radiomics and Deep Learning for Lung Cancer ScreeningRadiomics and Deep Learning for Lung Cancer Screening
Radiomics and Deep Learning for Lung Cancer ScreeningWookjin Choi
Ā 
An introduction to The Cancer Imaging Archive (Hands on)
An introduction to The Cancer Imaging Archive (Hands on)An introduction to The Cancer Imaging Archive (Hands on)
An introduction to The Cancer Imaging Archive (Hands on)CancerImagingInforma
Ā 
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI: Pr...
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI: Pr...Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI: Pr...
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI: Pr...Saifeng (Aaron) Liu
Ā 
AI and whole slide imaging biomarkers
AI and whole slide imaging biomarkersAI and whole slide imaging biomarkers
AI and whole slide imaging biomarkersJoel Saltz
Ā 
Longitudinal Plasma Samples: Paving the Way for Precision Oncology
Longitudinal Plasma Samples: Paving the Way for Precision OncologyLongitudinal Plasma Samples: Paving the Way for Precision Oncology
Longitudinal Plasma Samples: Paving the Way for Precision OncologyInsideScientific
Ā 
Longitudinal Plasma Samples: Paving the Way for Precision Oncology
Longitudinal Plasma Samples: Paving the Way for Precision OncologyLongitudinal Plasma Samples: Paving the Way for Precision Oncology
Longitudinal Plasma Samples: Paving the Way for Precision OncologyInsideScientific
Ā 
A practical guide to using The Cancer Imaging Archive for QIN Challenges and ...
A practical guide to using The Cancer Imaging Archive for QIN Challenges and ...A practical guide to using The Cancer Imaging Archive for QIN Challenges and ...
A practical guide to using The Cancer Imaging Archive for QIN Challenges and ...CancerImagingInforma
Ā 
Interpretable Spiculation Quantification for Lung Cancer Screening
Interpretable Spiculation Quantification for Lung Cancer ScreeningInterpretable Spiculation Quantification for Lung Cancer Screening
Interpretable Spiculation Quantification for Lung Cancer ScreeningWookjin Choi
Ā 
Dekker trog - learning outcome prediction models from cancer data - 2017
Dekker   trog  - learning outcome prediction models from cancer data - 2017Dekker   trog  - learning outcome prediction models from cancer data - 2017
Dekker trog - learning outcome prediction models from cancer data - 2017Andre Dekker
Ā 
Radiomics in Lung Cancer
Radiomics in Lung CancerRadiomics in Lung Cancer
Radiomics in Lung CancerWookjin Choi
Ā 
Utilization of NGS to Identify Clinically-Relevant Mutations in cfDNA: Meet t...
Utilization of NGS to Identify Clinically-Relevant Mutations in cfDNA: Meet t...Utilization of NGS to Identify Clinically-Relevant Mutations in cfDNA: Meet t...
Utilization of NGS to Identify Clinically-Relevant Mutations in cfDNA: Meet t...QIAGEN
Ā 
Pathomics Based Biomarkers, Tools, and Methods
Pathomics Based Biomarkers, Tools, and MethodsPathomics Based Biomarkers, Tools, and Methods
Pathomics Based Biomarkers, Tools, and Methodsimgcommcall
Ā 
Dekker trog - big data for radiation oncology - 2017
Dekker   trog  - big data for radiation oncology - 2017Dekker   trog  - big data for radiation oncology - 2017
Dekker trog - big data for radiation oncology - 2017Andre Dekker
Ā 
IMAGING BIOMARKER PANELS AND MULTI-OMICS AI MODELS FOR OUTCOMES PREDICTION
IMAGING BIOMARKER PANELS AND MULTI-OMICS AI MODELS FOR OUTCOMES PREDICTIONIMAGING BIOMARKER PANELS AND MULTI-OMICS AI MODELS FOR OUTCOMES PREDICTION
IMAGING BIOMARKER PANELS AND MULTI-OMICS AI MODELS FOR OUTCOMES PREDICTIONiQHub
Ā 
NCI HTAN, cancer trajectories, precision oncology
NCI HTAN, cancer trajectories, precision oncologyNCI HTAN, cancer trajectories, precision oncology
NCI HTAN, cancer trajectories, precision oncologyWarren Kibbe
Ā 
Imaging prostate cancer astellas
Imaging prostate cancer astellasImaging prostate cancer astellas
Imaging prostate cancer astellasMohamed Abdulla
Ā 

Similar to Artificial Intelligence in Radiation Oncology (20)

Quantitative Cancer Image Analysis
Quantitative Cancer Image AnalysisQuantitative Cancer Image Analysis
Quantitative Cancer Image Analysis
Ā 
Radiomics Analysis of Pulmonary Nodules in Low Dose CT for Early Detection of...
Radiomics Analysis of Pulmonary Nodules in Low Dose CT for Early Detection of...Radiomics Analysis of Pulmonary Nodules in Low Dose CT for Early Detection of...
Radiomics Analysis of Pulmonary Nodules in Low Dose CT for Early Detection of...
Ā 
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRIProstate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI
Ā 
Radiomics and Deep Learning for Lung Cancer Screening
Radiomics and Deep Learning for Lung Cancer ScreeningRadiomics and Deep Learning for Lung Cancer Screening
Radiomics and Deep Learning for Lung Cancer Screening
Ā 
An introduction to The Cancer Imaging Archive (Hands on)
An introduction to The Cancer Imaging Archive (Hands on)An introduction to The Cancer Imaging Archive (Hands on)
An introduction to The Cancer Imaging Archive (Hands on)
Ā 
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI: Pr...
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI: Pr...Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI: Pr...
Prostate Cancer Diagnosis using Deep Learning with 3D Multiparametric MRI: Pr...
Ā 
AI and whole slide imaging biomarkers
AI and whole slide imaging biomarkersAI and whole slide imaging biomarkers
AI and whole slide imaging biomarkers
Ā 
Longitudinal Plasma Samples: Paving the Way for Precision Oncology
Longitudinal Plasma Samples: Paving the Way for Precision OncologyLongitudinal Plasma Samples: Paving the Way for Precision Oncology
Longitudinal Plasma Samples: Paving the Way for Precision Oncology
Ā 
Longitudinal Plasma Samples: Paving the Way for Precision Oncology
Longitudinal Plasma Samples: Paving the Way for Precision OncologyLongitudinal Plasma Samples: Paving the Way for Precision Oncology
Longitudinal Plasma Samples: Paving the Way for Precision Oncology
Ā 
A practical guide to using The Cancer Imaging Archive for QIN Challenges and ...
A practical guide to using The Cancer Imaging Archive for QIN Challenges and ...A practical guide to using The Cancer Imaging Archive for QIN Challenges and ...
A practical guide to using The Cancer Imaging Archive for QIN Challenges and ...
Ā 
Interpretable Spiculation Quantification for Lung Cancer Screening
Interpretable Spiculation Quantification for Lung Cancer ScreeningInterpretable Spiculation Quantification for Lung Cancer Screening
Interpretable Spiculation Quantification for Lung Cancer Screening
Ā 
Dekker trog - learning outcome prediction models from cancer data - 2017
Dekker   trog  - learning outcome prediction models from cancer data - 2017Dekker   trog  - learning outcome prediction models from cancer data - 2017
Dekker trog - learning outcome prediction models from cancer data - 2017
Ā 
Radiomics in Lung Cancer
Radiomics in Lung CancerRadiomics in Lung Cancer
Radiomics in Lung Cancer
Ā 
Utilization of NGS to Identify Clinically-Relevant Mutations in cfDNA: Meet t...
Utilization of NGS to Identify Clinically-Relevant Mutations in cfDNA: Meet t...Utilization of NGS to Identify Clinically-Relevant Mutations in cfDNA: Meet t...
Utilization of NGS to Identify Clinically-Relevant Mutations in cfDNA: Meet t...
Ā 
Pathomics Based Biomarkers, Tools, and Methods
Pathomics Based Biomarkers, Tools, and MethodsPathomics Based Biomarkers, Tools, and Methods
Pathomics Based Biomarkers, Tools, and Methods
Ā 
Dekker trog - big data for radiation oncology - 2017
Dekker   trog  - big data for radiation oncology - 2017Dekker   trog  - big data for radiation oncology - 2017
Dekker trog - big data for radiation oncology - 2017
Ā 
IMAGING BIOMARKER PANELS AND MULTI-OMICS AI MODELS FOR OUTCOMES PREDICTION
IMAGING BIOMARKER PANELS AND MULTI-OMICS AI MODELS FOR OUTCOMES PREDICTIONIMAGING BIOMARKER PANELS AND MULTI-OMICS AI MODELS FOR OUTCOMES PREDICTION
IMAGING BIOMARKER PANELS AND MULTI-OMICS AI MODELS FOR OUTCOMES PREDICTION
Ā 
Oncology
OncologyOncology
Oncology
Ā 
NCI HTAN, cancer trajectories, precision oncology
NCI HTAN, cancer trajectories, precision oncologyNCI HTAN, cancer trajectories, precision oncology
NCI HTAN, cancer trajectories, precision oncology
Ā 
Imaging prostate cancer astellas
Imaging prostate cancer astellasImaging prostate cancer astellas
Imaging prostate cancer astellas
Ā 

More from Wookjin Choi

Deep Learning-based Histological Segmentation Differentiates Cavitation Patte...
Deep Learning-based Histological SegmentationDifferentiates Cavitation Patte...Deep Learning-based Histological SegmentationDifferentiates Cavitation Patte...
Deep Learning-based Histological Segmentation Differentiates Cavitation Patte...Wookjin Choi
Ā 
Artificial Intelligence To Reduce Radiation-induced Cardiotoxicity In Lung Ca...
Artificial Intelligence To Reduce Radiation-induced Cardiotoxicity In Lung Ca...Artificial Intelligence To Reduce Radiation-induced Cardiotoxicity In Lung Ca...
Artificial Intelligence To Reduce Radiation-induced Cardiotoxicity In Lung Ca...Wookjin Choi
Ā 
Novel Functional Radiomics for Prediction of Cardiac PET Avidity in Lung Canc...
Novel Functional Radiomics for Prediction of Cardiac PET Avidity in Lung Canc...Novel Functional Radiomics for Prediction of Cardiac PET Avidity in Lung Canc...
Novel Functional Radiomics for Prediction of Cardiac PET Avidity in Lung Canc...Wookjin Choi
Ā 
Novel Deep Learning Segmentation Models for Accurate GTV and OAR Segmentation...
Novel Deep Learning Segmentation Models for Accurate GTV and OAR Segmentation...Novel Deep Learning Segmentation Models for Accurate GTV and OAR Segmentation...
Novel Deep Learning Segmentation Models for Accurate GTV and OAR Segmentation...Wookjin Choi
Ā 
Novel Functional Delta-Radiomics for Predicting Overall Survival in Lung Canc...
Novel Functional Delta-Radiomics for Predicting Overall Survival in Lung Canc...Novel Functional Delta-Radiomics for Predicting Overall Survival in Lung Canc...
Novel Functional Delta-Radiomics for Predicting Overall Survival in Lung Canc...Wookjin Choi
Ā 
CIRDataset: A large-scale Dataset for Clinically-Interpretable lung nodule Ra...
CIRDataset: A large-scale Dataset for Clinically-Interpretable lung nodule Ra...CIRDataset: A large-scale Dataset for Clinically-Interpretable lung nodule Ra...
CIRDataset: A large-scale Dataset for Clinically-Interpretable lung nodule Ra...Wookjin Choi
Ā 
Automatic motion tracking system for analysis of insect behavior
Automatic motion tracking system for analysis of insect behaviorAutomatic motion tracking system for analysis of insect behavior
Automatic motion tracking system for analysis of insect behaviorWookjin Choi
Ā 
Assessing the Dosimetric Links between Organ-At-Risk Delineation Variability ...
Assessing the Dosimetric Links between Organ-At-Risk Delineation Variability ...Assessing the Dosimetric Links between Organ-At-Risk Delineation Variability ...
Assessing the Dosimetric Links between Organ-At-Risk Delineation Variability ...Wookjin Choi
Ā 
Simulation of Realistic Organ-At-Risk Delineation Variability in Head and Nec...
Simulation of Realistic Organ-At-Risk Delineation Variability in Head and Nec...Simulation of Realistic Organ-At-Risk Delineation Variability in Head and Nec...
Simulation of Realistic Organ-At-Risk Delineation Variability in Head and Nec...Wookjin Choi
Ā 
Quantitative image analysis for cancer diagnosis and radiation therapy
Quantitative image analysis for cancer diagnosis and radiation therapyQuantitative image analysis for cancer diagnosis and radiation therapy
Quantitative image analysis for cancer diagnosis and radiation therapyWookjin Choi
Ā 
Interpretable Spiculation Quantification for Lung Cancer Screening
Interpretable Spiculation Quantification for Lung Cancer ScreeningInterpretable Spiculation Quantification for Lung Cancer Screening
Interpretable Spiculation Quantification for Lung Cancer ScreeningWookjin Choi
Ā 
Quantitative Image Analysis for Cancer Diagnosis and Radiation Therapy
Quantitative Image Analysis for Cancer Diagnosis and Radiation TherapyQuantitative Image Analysis for Cancer Diagnosis and Radiation Therapy
Quantitative Image Analysis for Cancer Diagnosis and Radiation TherapyWookjin Choi
Ā 
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...Wookjin Choi
Ā 
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...Wookjin Choi
Ā 
Aggressive Lung Adenocarcinoma Subtype Prediction Using FDG-PET/CT Radiomics
Aggressive Lung Adenocarcinoma Subtype Prediction Using FDG-PET/CT RadiomicsAggressive Lung Adenocarcinoma Subtype Prediction Using FDG-PET/CT Radiomics
Aggressive Lung Adenocarcinoma Subtype Prediction Using FDG-PET/CT RadiomicsWookjin Choi
Ā 
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...Wookjin Choi
Ā 
Current Projects - Wookjin Choi
Current Projects - Wookjin ChoiCurrent Projects - Wookjin Choi
Current Projects - Wookjin ChoiWookjin Choi
Ā 
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...Wookjin Choi
Ā 
Identification of Robust Normal Lung CT Texture Features
Identification of Robust Normal Lung CT Texture FeaturesIdentification of Robust Normal Lung CT Texture Features
Identification of Robust Normal Lung CT Texture FeaturesWookjin Choi
Ā 
Robust breathing signal extraction from cone beam CT projections based on ada...
Robust breathing signal extraction from cone beam CT projections based on ada...Robust breathing signal extraction from cone beam CT projections based on ada...
Robust breathing signal extraction from cone beam CT projections based on ada...Wookjin Choi
Ā 

More from Wookjin Choi (20)

Deep Learning-based Histological Segmentation Differentiates Cavitation Patte...
Deep Learning-based Histological SegmentationDifferentiates Cavitation Patte...Deep Learning-based Histological SegmentationDifferentiates Cavitation Patte...
Deep Learning-based Histological Segmentation Differentiates Cavitation Patte...
Ā 
Artificial Intelligence To Reduce Radiation-induced Cardiotoxicity In Lung Ca...
Artificial Intelligence To Reduce Radiation-induced Cardiotoxicity In Lung Ca...Artificial Intelligence To Reduce Radiation-induced Cardiotoxicity In Lung Ca...
Artificial Intelligence To Reduce Radiation-induced Cardiotoxicity In Lung Ca...
Ā 
Novel Functional Radiomics for Prediction of Cardiac PET Avidity in Lung Canc...
Novel Functional Radiomics for Prediction of Cardiac PET Avidity in Lung Canc...Novel Functional Radiomics for Prediction of Cardiac PET Avidity in Lung Canc...
Novel Functional Radiomics for Prediction of Cardiac PET Avidity in Lung Canc...
Ā 
Novel Deep Learning Segmentation Models for Accurate GTV and OAR Segmentation...
Novel Deep Learning Segmentation Models for Accurate GTV and OAR Segmentation...Novel Deep Learning Segmentation Models for Accurate GTV and OAR Segmentation...
Novel Deep Learning Segmentation Models for Accurate GTV and OAR Segmentation...
Ā 
Novel Functional Delta-Radiomics for Predicting Overall Survival in Lung Canc...
Novel Functional Delta-Radiomics for Predicting Overall Survival in Lung Canc...Novel Functional Delta-Radiomics for Predicting Overall Survival in Lung Canc...
Novel Functional Delta-Radiomics for Predicting Overall Survival in Lung Canc...
Ā 
CIRDataset: A large-scale Dataset for Clinically-Interpretable lung nodule Ra...
CIRDataset: A large-scale Dataset for Clinically-Interpretable lung nodule Ra...CIRDataset: A large-scale Dataset for Clinically-Interpretable lung nodule Ra...
CIRDataset: A large-scale Dataset for Clinically-Interpretable lung nodule Ra...
Ā 
Automatic motion tracking system for analysis of insect behavior
Automatic motion tracking system for analysis of insect behaviorAutomatic motion tracking system for analysis of insect behavior
Automatic motion tracking system for analysis of insect behavior
Ā 
Assessing the Dosimetric Links between Organ-At-Risk Delineation Variability ...
Assessing the Dosimetric Links between Organ-At-Risk Delineation Variability ...Assessing the Dosimetric Links between Organ-At-Risk Delineation Variability ...
Assessing the Dosimetric Links between Organ-At-Risk Delineation Variability ...
Ā 
Simulation of Realistic Organ-At-Risk Delineation Variability in Head and Nec...
Simulation of Realistic Organ-At-Risk Delineation Variability in Head and Nec...Simulation of Realistic Organ-At-Risk Delineation Variability in Head and Nec...
Simulation of Realistic Organ-At-Risk Delineation Variability in Head and Nec...
Ā 
Quantitative image analysis for cancer diagnosis and radiation therapy
Quantitative image analysis for cancer diagnosis and radiation therapyQuantitative image analysis for cancer diagnosis and radiation therapy
Quantitative image analysis for cancer diagnosis and radiation therapy
Ā 
Interpretable Spiculation Quantification for Lung Cancer Screening
Interpretable Spiculation Quantification for Lung Cancer ScreeningInterpretable Spiculation Quantification for Lung Cancer Screening
Interpretable Spiculation Quantification for Lung Cancer Screening
Ā 
Quantitative Image Analysis for Cancer Diagnosis and Radiation Therapy
Quantitative Image Analysis for Cancer Diagnosis and Radiation TherapyQuantitative Image Analysis for Cancer Diagnosis and Radiation Therapy
Quantitative Image Analysis for Cancer Diagnosis and Radiation Therapy
Ā 
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...
Ā 
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...
Robust Normal Lung CT Texture Features for the Prediction of Radiation-Induce...
Ā 
Aggressive Lung Adenocarcinoma Subtype Prediction Using FDG-PET/CT Radiomics
Aggressive Lung Adenocarcinoma Subtype Prediction Using FDG-PET/CT RadiomicsAggressive Lung Adenocarcinoma Subtype Prediction Using FDG-PET/CT Radiomics
Aggressive Lung Adenocarcinoma Subtype Prediction Using FDG-PET/CT Radiomics
Ā 
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...
Ā 
Current Projects - Wookjin Choi
Current Projects - Wookjin ChoiCurrent Projects - Wookjin Choi
Current Projects - Wookjin Choi
Ā 
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...
Individually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in...
Ā 
Identification of Robust Normal Lung CT Texture Features
Identification of Robust Normal Lung CT Texture FeaturesIdentification of Robust Normal Lung CT Texture Features
Identification of Robust Normal Lung CT Texture Features
Ā 
Robust breathing signal extraction from cone beam CT projections based on ada...
Robust breathing signal extraction from cone beam CT projections based on ada...Robust breathing signal extraction from cone beam CT projections based on ada...
Robust breathing signal extraction from cone beam CT projections based on ada...
Ā 

Recently uploaded

call girls in Connaught Place DELHI šŸ” >ą¼’9540349809 šŸ” genuine Escort Service ...
call girls in Connaught Place  DELHI šŸ” >ą¼’9540349809 šŸ” genuine Escort Service ...call girls in Connaught Place  DELHI šŸ” >ą¼’9540349809 šŸ” genuine Escort Service ...
call girls in Connaught Place DELHI šŸ” >ą¼’9540349809 šŸ” genuine Escort Service ...saminamagar
Ā 
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment Booking
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment BookingCall Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment Booking
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment BookingNehru place Escorts
Ā 
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...narwatsonia7
Ā 
Call Girl Lucknow Mallika 7001305949 Independent Escort Service Lucknow
Call Girl Lucknow Mallika 7001305949 Independent Escort Service LucknowCall Girl Lucknow Mallika 7001305949 Independent Escort Service Lucknow
Call Girl Lucknow Mallika 7001305949 Independent Escort Service Lucknownarwatsonia7
Ā 
Artifacts in Nuclear Medicine with Identifying and resolving artifacts.
Artifacts in Nuclear Medicine with Identifying and resolving artifacts.Artifacts in Nuclear Medicine with Identifying and resolving artifacts.
Artifacts in Nuclear Medicine with Identifying and resolving artifacts.MiadAlsulami
Ā 
Call Girls Hebbal Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Hebbal Just Call 7001305949 Top Class Call Girl Service AvailableCall Girls Hebbal Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Hebbal Just Call 7001305949 Top Class Call Girl Service Availablenarwatsonia7
Ā 
Call Girl Bangalore Nandini 7001305949 Independent Escort Service Bangalore
Call Girl Bangalore Nandini 7001305949 Independent Escort Service BangaloreCall Girl Bangalore Nandini 7001305949 Independent Escort Service Bangalore
Call Girl Bangalore Nandini 7001305949 Independent Escort Service Bangalorenarwatsonia7
Ā 
VIP Call Girls Lucknow Nandini 7001305949 Independent Escort Service Lucknow
VIP Call Girls Lucknow Nandini 7001305949 Independent Escort Service LucknowVIP Call Girls Lucknow Nandini 7001305949 Independent Escort Service Lucknow
VIP Call Girls Lucknow Nandini 7001305949 Independent Escort Service Lucknownarwatsonia7
Ā 
VIP Call Girls Pune Vrinda 9907093804 Short 1500 Night 6000 Best call girls S...
VIP Call Girls Pune Vrinda 9907093804 Short 1500 Night 6000 Best call girls S...VIP Call Girls Pune Vrinda 9907093804 Short 1500 Night 6000 Best call girls S...
VIP Call Girls Pune Vrinda 9907093804 Short 1500 Night 6000 Best call girls S...Miss joya
Ā 
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...narwatsonia7
Ā 
College Call Girls Vyasarpadi Whatsapp 7001305949 Independent Escort Service
College Call Girls Vyasarpadi Whatsapp 7001305949 Independent Escort ServiceCollege Call Girls Vyasarpadi Whatsapp 7001305949 Independent Escort Service
College Call Girls Vyasarpadi Whatsapp 7001305949 Independent Escort ServiceNehru place Escorts
Ā 
Sonagachi Call Girls Services 9907093804 @24x7 High Class Babes Here Call Now
Sonagachi Call Girls Services 9907093804 @24x7 High Class Babes Here Call NowSonagachi Call Girls Services 9907093804 @24x7 High Class Babes Here Call Now
Sonagachi Call Girls Services 9907093804 @24x7 High Class Babes Here Call NowRiya Pathan
Ā 
See the 2,456 pharmacies on the National E-Pharmacy Platform
See the 2,456 pharmacies on the National E-Pharmacy PlatformSee the 2,456 pharmacies on the National E-Pharmacy Platform
See the 2,456 pharmacies on the National E-Pharmacy PlatformKweku Zurek
Ā 
Glomerular Filtration and determinants of glomerular filtration .pptx
Glomerular Filtration and  determinants of glomerular filtration .pptxGlomerular Filtration and  determinants of glomerular filtration .pptx
Glomerular Filtration and determinants of glomerular filtration .pptxDr.Nusrat Tariq
Ā 
Call Girls Service In Shyam Nagar Whatsapp 8445551418 Independent Escort Service
Call Girls Service In Shyam Nagar Whatsapp 8445551418 Independent Escort ServiceCall Girls Service In Shyam Nagar Whatsapp 8445551418 Independent Escort Service
Call Girls Service In Shyam Nagar Whatsapp 8445551418 Independent Escort Serviceparulsinha
Ā 
Low Rate Call Girls Pune Esha 9907093804 Short 1500 Night 6000 Best call girl...
Low Rate Call Girls Pune Esha 9907093804 Short 1500 Night 6000 Best call girl...Low Rate Call Girls Pune Esha 9907093804 Short 1500 Night 6000 Best call girl...
Low Rate Call Girls Pune Esha 9907093804 Short 1500 Night 6000 Best call girl...Miss joya
Ā 
Call Girls In Andheri East Call 9920874524 Book Hot And Sexy Girls
Call Girls In Andheri East Call 9920874524 Book Hot And Sexy GirlsCall Girls In Andheri East Call 9920874524 Book Hot And Sexy Girls
Call Girls In Andheri East Call 9920874524 Book Hot And Sexy Girlsnehamumbai
Ā 
Call Girls Thane Just Call 9910780858 Get High Class Call Girls Service
Call Girls Thane Just Call 9910780858 Get High Class Call Girls ServiceCall Girls Thane Just Call 9910780858 Get High Class Call Girls Service
Call Girls Thane Just Call 9910780858 Get High Class Call Girls Servicesonalikaur4
Ā 
Call Girls Service Chennai Jiya 7001305949 Independent Escort Service Chennai
Call Girls Service Chennai Jiya 7001305949 Independent Escort Service ChennaiCall Girls Service Chennai Jiya 7001305949 Independent Escort Service Chennai
Call Girls Service Chennai Jiya 7001305949 Independent Escort Service ChennaiNehru place Escorts
Ā 

Recently uploaded (20)

call girls in Connaught Place DELHI šŸ” >ą¼’9540349809 šŸ” genuine Escort Service ...
call girls in Connaught Place  DELHI šŸ” >ą¼’9540349809 šŸ” genuine Escort Service ...call girls in Connaught Place  DELHI šŸ” >ą¼’9540349809 šŸ” genuine Escort Service ...
call girls in Connaught Place DELHI šŸ” >ą¼’9540349809 šŸ” genuine Escort Service ...
Ā 
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment Booking
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment BookingCall Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment Booking
Call Girls Service Nandiambakkam | 7001305949 At Low Cost Cash Payment Booking
Ā 
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...
Russian Call Girls Chickpet - 7001305949 Booking and charges genuine rate for...
Ā 
Call Girl Lucknow Mallika 7001305949 Independent Escort Service Lucknow
Call Girl Lucknow Mallika 7001305949 Independent Escort Service LucknowCall Girl Lucknow Mallika 7001305949 Independent Escort Service Lucknow
Call Girl Lucknow Mallika 7001305949 Independent Escort Service Lucknow
Ā 
Artifacts in Nuclear Medicine with Identifying and resolving artifacts.
Artifacts in Nuclear Medicine with Identifying and resolving artifacts.Artifacts in Nuclear Medicine with Identifying and resolving artifacts.
Artifacts in Nuclear Medicine with Identifying and resolving artifacts.
Ā 
Call Girls Hebbal Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Hebbal Just Call 7001305949 Top Class Call Girl Service AvailableCall Girls Hebbal Just Call 7001305949 Top Class Call Girl Service Available
Call Girls Hebbal Just Call 7001305949 Top Class Call Girl Service Available
Ā 
Call Girl Bangalore Nandini 7001305949 Independent Escort Service Bangalore
Call Girl Bangalore Nandini 7001305949 Independent Escort Service BangaloreCall Girl Bangalore Nandini 7001305949 Independent Escort Service Bangalore
Call Girl Bangalore Nandini 7001305949 Independent Escort Service Bangalore
Ā 
VIP Call Girls Lucknow Nandini 7001305949 Independent Escort Service Lucknow
VIP Call Girls Lucknow Nandini 7001305949 Independent Escort Service LucknowVIP Call Girls Lucknow Nandini 7001305949 Independent Escort Service Lucknow
VIP Call Girls Lucknow Nandini 7001305949 Independent Escort Service Lucknow
Ā 
VIP Call Girls Pune Vrinda 9907093804 Short 1500 Night 6000 Best call girls S...
VIP Call Girls Pune Vrinda 9907093804 Short 1500 Night 6000 Best call girls S...VIP Call Girls Pune Vrinda 9907093804 Short 1500 Night 6000 Best call girls S...
VIP Call Girls Pune Vrinda 9907093804 Short 1500 Night 6000 Best call girls S...
Ā 
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...
Call Girls Electronic City Just Call 7001305949 Top Class Call Girl Service A...
Ā 
College Call Girls Vyasarpadi Whatsapp 7001305949 Independent Escort Service
College Call Girls Vyasarpadi Whatsapp 7001305949 Independent Escort ServiceCollege Call Girls Vyasarpadi Whatsapp 7001305949 Independent Escort Service
College Call Girls Vyasarpadi Whatsapp 7001305949 Independent Escort Service
Ā 
Sonagachi Call Girls Services 9907093804 @24x7 High Class Babes Here Call Now
Sonagachi Call Girls Services 9907093804 @24x7 High Class Babes Here Call NowSonagachi Call Girls Services 9907093804 @24x7 High Class Babes Here Call Now
Sonagachi Call Girls Services 9907093804 @24x7 High Class Babes Here Call Now
Ā 
See the 2,456 pharmacies on the National E-Pharmacy Platform
See the 2,456 pharmacies on the National E-Pharmacy PlatformSee the 2,456 pharmacies on the National E-Pharmacy Platform
See the 2,456 pharmacies on the National E-Pharmacy Platform
Ā 
Escort Service Call Girls In Sarita Vihar,, 99530Ā°56974 Delhi NCR
Escort Service Call Girls In Sarita Vihar,, 99530Ā°56974 Delhi NCREscort Service Call Girls In Sarita Vihar,, 99530Ā°56974 Delhi NCR
Escort Service Call Girls In Sarita Vihar,, 99530Ā°56974 Delhi NCR
Ā 
Glomerular Filtration and determinants of glomerular filtration .pptx
Glomerular Filtration and  determinants of glomerular filtration .pptxGlomerular Filtration and  determinants of glomerular filtration .pptx
Glomerular Filtration and determinants of glomerular filtration .pptx
Ā 
Call Girls Service In Shyam Nagar Whatsapp 8445551418 Independent Escort Service
Call Girls Service In Shyam Nagar Whatsapp 8445551418 Independent Escort ServiceCall Girls Service In Shyam Nagar Whatsapp 8445551418 Independent Escort Service
Call Girls Service In Shyam Nagar Whatsapp 8445551418 Independent Escort Service
Ā 
Low Rate Call Girls Pune Esha 9907093804 Short 1500 Night 6000 Best call girl...
Low Rate Call Girls Pune Esha 9907093804 Short 1500 Night 6000 Best call girl...Low Rate Call Girls Pune Esha 9907093804 Short 1500 Night 6000 Best call girl...
Low Rate Call Girls Pune Esha 9907093804 Short 1500 Night 6000 Best call girl...
Ā 
Call Girls In Andheri East Call 9920874524 Book Hot And Sexy Girls
Call Girls In Andheri East Call 9920874524 Book Hot And Sexy GirlsCall Girls In Andheri East Call 9920874524 Book Hot And Sexy Girls
Call Girls In Andheri East Call 9920874524 Book Hot And Sexy Girls
Ā 
Call Girls Thane Just Call 9910780858 Get High Class Call Girls Service
Call Girls Thane Just Call 9910780858 Get High Class Call Girls ServiceCall Girls Thane Just Call 9910780858 Get High Class Call Girls Service
Call Girls Thane Just Call 9910780858 Get High Class Call Girls Service
Ā 
Call Girls Service Chennai Jiya 7001305949 Independent Escort Service Chennai
Call Girls Service Chennai Jiya 7001305949 Independent Escort Service ChennaiCall Girls Service Chennai Jiya 7001305949 Independent Escort Service Chennai
Call Girls Service Chennai Jiya 7001305949 Independent Escort Service Chennai
Ā 

Artificial Intelligence in Radiation Oncology

  • 1. Artificial Intelligence in Radiation Oncology Wookjin Choi, PhD Assistant Professor of Radiation Oncology Sidney Kimmel Medical College at Thomas Jefferson University Wookjin.Choi@Jefferson.edu Mar 11, 2022 @ Mayo Clinic
  • 2. Acknowledgements Memorial Sloan Kettering Cancer Center ā€¢ Wei Lu PhD ā€¢ Sadegh Riyahi, PhD ā€¢ Jung Hun Oh, PhD ā€¢ Saad Nadeem, PhD ā€¢ Eric Aliotta, PhD ā€¢ Joseph O. Deasy, PhD ā€¢ Andreas Rimner, MD ā€¢ Prasad Adusumilli, MD Stony Brook University ā€¢ Allen Tannenbaum, PhD University of Virginia School of Medicine ā€¢ Jeffrey Siebers, PhD ā€¢ Victor Gabriel Leandro Alves, PhD University of Maryland School of Medicine ā€¢ Howard Zhang, PhD ā€¢ Wengen Chen, MD, PhD ā€¢ Charles White, MD Thomas Jefferson University ā€¢ Yevgeniy Vinogradskiy, PhD ā€¢ Hamidreza Nourzadeh, PhD ā€¢ Adam P. Dicker, MD 2 NIH/NCI Grant R01 CA222216, R01 CA172638 and NIH/NCI Cancer Center Support Grant P30 CA008748 and 5P30 CA056036 The ESTRO Falcon project team, Scott Kaylor of EduCase, Benjamin Nelms of Proknow for the multi- delineator contour data presented in this work
  • 3. AI in Radiation Oncology 3 Huynh et al. Nat Rev Clin Oncol 2020
  • 4. 4 Netherton et al. Oncology 2021 Hype cycle for three major innovations in radiation oncology Automatable tasks in radiation oncology for the modern clinic
  • 5. Outline ā€¢ Radiomics - Decision Support Tools - Lung Cancer Screening - Tumor Response Prediction and Evaluation - Aggressive Lung ADC subtype prediction - Multimodal data: Pathology, Multiomics, etc. ā€¢ Auto Delineation and Variability Analysis - Delineation Variability Quantification - Dosimetric Consequences of Variabilities - OARNet, Voxel2Mesh 5
  • 6. Radiomics 6 ĀØ Controllable Feature Analysis ĀØ More Interpretable Lambin, et al. Eur J Cancer 2012 Aerts et al., Nature Communications, 2014
  • 7. Radiomics Framework 7 Image Registration ā€¢ Multi-level rigid ā€¢ Deformable ā€¢ Pre/Post-CT ā€¢ MSE, MI Tumor Segmentation ā€¢ Adaptive region growing ā€¢ Level set ā€¢ Grow cut ā€¢ Morphology filter ā€¢ Multi-modality image segmentation Feature Extraction ā€¢ Intensity distribution ā€¢ Spatial variations (texture) ā€¢ Geometric properties ā€¢ Jacobian feature from DVF ā€¢ Feature selection Predictive Model ā€¢ ROC analyses ā€¢ Prediction models ā€¢ Validation ā€¢ Tumor response ā€¢ Recurrence ā€¢ Survival Source codes: https://github.com/taznux/radiomics-tools ā€¢ Automated Workflow - Integrate all the radiomics components - 3D Slicer, ITK (C++), Matlab, R, and Python - Scalable: support multicore & GPU computing
  • 8. Radiomics Framework 8 Image Registration ā€¢ Multi-level rigid ā€¢ Deformable ā€¢ Pre/Post-CT ā€¢ MSE, MI Tumor Segmentation ā€¢ Adaptive region growing ā€¢ Level set ā€¢ Grow cut ā€¢ Morphology filter ā€¢ Multi-modality image segmentation Feature Extraction ā€¢ Intensity distribution ā€¢ Spatial variations (texture) ā€¢ Geometric properties ā€¢ Jacobian feature from DVF ā€¢ Feature selection Predictive Model ā€¢ ROC analyses ā€¢ Prediction models ā€¢ Validation ā€¢ Tumor response ā€¢ Recurrence ā€¢ Survival Source codes: https://github.com/taznux/radiomics-tools Deep Learning Model ā€¢ Automated Workflow - Integrate all the radiomics components - 3D Slicer, ITK (C++), Matlab, R, and Python - Scalable: support multicore & GPU computing
  • 9. Radiomics Framework 9 Image Registration ā€¢ Multi-level rigid ā€¢ Deformable ā€¢ Pre/Post-CT ā€¢ MSE, MI Tumor Segmentation ā€¢ Adaptive region growing ā€¢ Level set ā€¢ Grow cut ā€¢ Morphology filter ā€¢ Multi-modality image segmentation Feature Extraction ā€¢ Intensity distribution ā€¢ Spatial variations (texture) ā€¢ Geometric properties ā€¢ Jacobian feature from DVF ā€¢ Feature selection Predictive Model ā€¢ ROC analyses ā€¢ Prediction models ā€¢ Validation ā€¢ Tumor response ā€¢ Recurrence ā€¢ Survival Source codes: https://github.com/taznux/radiomics-tools Deep Learning Model ā€¢ Automated Workflow - Integrate all the radiomics components - 3D Slicer, ITK (C++), Matlab, R, and Python - Scalable: support multicore & GPU computing
  • 10. Radiomics Framework 10 Image Registration ā€¢ Multi-level rigid ā€¢ Deformable ā€¢ Pre/Post-CT ā€¢ MSE, MI Tumor Segmentation ā€¢ Adaptive region growing ā€¢ Level set ā€¢ Grow cut ā€¢ Morphology filter ā€¢ Multi-modality image segmentation Feature Extraction ā€¢ Intensity distribution ā€¢ Spatial variations (texture) ā€¢ Geometric properties ā€¢ Jacobian feature from DVF ā€¢ Feature selection Predictive Model ā€¢ ROC analyses ā€¢ Prediction models ā€¢ Validation ā€¢ Tumor response ā€¢ Recurrence ā€¢ Survival Source codes: https://github.com/taznux/radiomics-tools Deep Learning Model ā€¢ Automated Workflow - Integrate all the radiomics components - 3D Slicer, ITK (C++), Matlab, R, and Python - Scalable: support multicore & GPU computing
  • 11. Radiomics Framework 11 Image Registration ā€¢ Multi-level rigid ā€¢ Deformable ā€¢ Pre/Post-CT ā€¢ MSE, MI Tumor Segmentation ā€¢ Adaptive region growing ā€¢ Level set ā€¢ Grow cut ā€¢ Morphology filter ā€¢ Multi-modality image segmentation Feature Extraction ā€¢ Intensity distribution ā€¢ Spatial variations (texture) ā€¢ Geometric properties ā€¢ Jacobian feature from DVF ā€¢ Feature selection Predictive Model ā€¢ ROC analyses ā€¢ Prediction models ā€¢ Validation ā€¢ Tumor response ā€¢ Recurrence ā€¢ Survival Source codes: https://github.com/taznux/radiomics-tools Deep Learning Model ā€¢ Automated Workflow - Integrate all the radiomics components - 3D Slicer, ITK (C++), Matlab, R, and Python - Scalable: support multicore & GPU computing
  • 12. Radiomics Framework 12 Image Registration ā€¢ Deep Learning Optical fl ow ā€¢ Action like flow ā€¢ Differential warp ā€¢ Dynamic filtering ā€¢ ā€¦ Tumor Segmentation ā€¢ U-Net ā€¢ Prob. U-Net ā€¢ UANet ā€¢ ā€¦ Feature Extraction ā€¢ AlexNet ā€¢ ResNet ā€¢ VGG ā€¢ LeNet ā€¢ ā€¦. Predictive Model ā€¢ ROC analyses ā€¢ Prediction models ā€¢ Validation ā€¢ Tumor response ā€¢ Recurrence ā€¢ Survival ā€¢ Automated Workflow - Integrate all the radiomics components - 3D Slicer, ITK (C++), Matlab, R, and Python - Scalable: support multicore & GPU computing
  • 13. Lung Cancer Screening 13 ĀØ Early detection of lung cancer by LDCT can reduce mortality ĀØ Known features correlated with PN malignancy Ā¤ Size, growth rate (Lung-RADS) Ā¤ Calcification, enhancement, solidity ā†’ texture features Ā¤ Boundary margins (spiculation, lobulation), attachment ā†’ shape and appearance features Malignant nodules Benign nodules Size Total Malignancy < 4mm 2038 0% 4-7 mm 1034 1% 8-20 mm 268 15% > 20 mm 16 75%
  • 14. ACR Lung-RADS 1.0 Category Baseline Screening Malignancy 1 No PNs; PNs with calcification Negative <1% chance of malignancy 2 Solid/part-solid: <6 mm GGN: <20 mm Benign appearance <1% chance of malignancy 3 Solid: ā‰„6 to <8 mm Part-solid: ā‰„6 mm with solid component <6 mm GGN: ā‰„20 mm Probably benign 1-2% chance of malignancy 4A Solid: ā‰„8 to <15 mm Part-solid: ā‰„8 mm with solid component ā‰„6 and <8 mm Suspicious 5-15% chance of malignancy 4B Solid: ā‰„15 mm Part-solid: Solid component ā‰„8 mm >15% chance of malignancy 4X Category 3 or 4 PNs with suspicious features (e.g., enlarged lymph nodes) or suspicious imaging findings (e.g., spiculation) >15% chance of malignancy 14 ACR: American College of Radiology Lung-RADS: Lung CT Screening Reporting and Data System
  • 15. Lung Cancer Screening (Methodology) ā€¢ TCIA LIDC-IDRI public data set (n=1,010) - Multi-institutional data - 72 cases evaluated (31 benign and 41 malignant cases) ā€¢ Consensus contour 15 GLCM GLRM Texture features Intensity features 2D Shape features 3D
  • 16. Lung Cancer Screening (SVM-LASSO Model ) 16 SVM classification Distinctive feature identification Malignant? Predicted malignancy Feature extraction Yes 10x10-fold CV 10-fold CV LASSO feature selection ā€¢ Size (BB_AP) : Highly correlated with the axial longest diameter and its perpendicular diameter (r = 0.96, larger ā€“ more malignant) ā€¢ Texture (SD_IDM) : Tumor heterogeneity (smaller ā€“ more malignant)
  • 17. Lung Screening (Results: Comparison) Sensitivity Specificity Accuracy AUC Lung-RADS Clinical guideline 73.3% 70.4% 72.2% 0.74 Hawkins et al. (2016) Radiomics ā€“ 23 features 51.7 % 92.9% 80.0% 0.83 Ma et al. (2016) Radiomics ā€“ 583 features 80.0% 85.5% 82.7% Buty et al. (2016) DL ā€“ 400 SH and 4096 AlexNet features 82.4% Kumar et al. (2015) DL: 5000 features 79.1% 76.1% 77.5% Proposed Radiomics: two features (Size and Texture) 87.2% 81.2% 84.6% 0.89 17 DL: Deep Learning, SH: Spherical Harmonics Choi et al., Medical Physics, 2018.
  • 18. ACR Lung-RADS 1.0 Category Baseline Screening Malignancy 1 No PNs; PNs with calcification Negative <1% chance of malignancy 2 Solid/part-solid: <6 mm GGN: <20 mm Benign appearance <1% chance of malignancy 3 Solid: ā‰„6 to <8 mm Part-solid: ā‰„6 mm with solid component <6 mm GGN: ā‰„20 mm Probably benign 1-2% chance of malignancy 4A Solid: ā‰„8 to <15 mm Part-solid: ā‰„8 mm with solid component ā‰„6 and <8 mm Suspicious 5-15% chance of malignancy 4B Solid: ā‰„15 mm Part-solid: Solid component ā‰„8 mm >15% chance of malignancy 4X Category 3 or 4 PNs with suspicious features (e.g., enlarged lymph nodes) or suspicious imaging findings (e.g., spiculation) >15% chance of malignancy 18 ACR: American College of Radiology Lung-RADS: Lung CT Screening Reporting and Data System
  • 19. Spiculation Quantification (Motivation) ā€¢ Semantic Features ā€¢ Semi-automatic Segmentation - GrowCut and LevelSet 19 Radiologists spiculation score (RS) for different pulmonary nodules 1 2 3 4 5 Choi et al. in CMPB 2021
  • 20. Spiculation Quantification (Methodology) 20 !!: = log āˆ‘",$ ( ([+(,!), +(,"), +(,$)]) āˆ‘",$ ( ([,!, ,", ,$]) Area Distortion Map Spherical Mapping Eigenfunction
  • 21. Spiculation Quantification (Results) 21 Number of Spiculations and Radiologists spiculation score (RS) for different pulmonary nodules 1 2 3 4 5 Choi et al. in CMPB 2021 0 1 4 8 14
  • 23. Spiculation Quantification (Results: Comparison) 23 Choi et al. in CMPB 2021
  • 24. Progression-free survival Prediction after SBRT for early-stage NSCLC 24 Thor, Choi et al. ASTRO 2020 ā€¢ 412 patients treated between 2006 and 2017 ā€¢ PETs and CTs within three months prior to SBRT start. ā€¢ The median prescription dose was 50Gy in 5 fractions.
  • 25. Progression-free survival Prediction (Results) ā€¢ PET entropy, CT number of peaks, CT major axis, and gender. ā€¢ The most frequently selected model included PET entropy and CT number of peaks - The c-index in the validation subset was 0.77 - The prediction-stratified survival indicated a clear separation between the observed HR and LR - e.g. a PFS of 60% was observed at 12 months in HR vs. 22 months in LR. 25 Thor, Choi et al. ASTRO 2020
  • 26. Local tumor morphological changes 26 Jacobian Map - Jacobian matrix: calculates rate of displacement change in each direction. - Determinant indicates volumetric ratio of shrinkage/expansion. 012 3 = 4 012 3 > 1 volume expansion 012 3 = 1 no volume change 012 3 < 1 volume shrinkage 012 3 = 1.2 = 20% expansion 012 3 = 0.8 = 20% shrinkage (-20%) Riyahi, Choi et al., PMB 2018
  • 27. Local tumor morphological changes (Results) 27
  • 28. Local tumor morphological changes (Results) Features P-value AUC Correlation to responders Minimum Jacobian 0.009 0.98 -0.79 Median Jacobian 0.046 0.95 -0.72 The P-value, AUC and correlation to responders for all significant features in univariate analysis 28 Riyahi, Choi et al., PMB 2018 SVM-LASSO: AUC 0.91
  • 29. Local Metabolic Tumor Volume Changes 29 Riyahi, Choi et al., DATRA@MICCAI 2018 AUC=0.81
  • 30. Aggressive Lung ADC Subtype Prediction (Motivation) 30 CT MIP PET/CT Soild CT PET/CT Five classifications of lung ADC Travis et al. JTO 2011 ĀØ Solid and MIP components: poor surgery/SBRT prognosis factor Ā¤ Benefit from lobectomy rather than limited resection ĀØ Core biopsy (Leeman et al. IJROBP 2017) Ā¤ Minimally invasive, not routinely performed, sampling error (about 60% agreement with pathology) ĀØ Preoperative diagnostic CT and FDG PET/CT radiomics Ā¤ Non-invasive and routinely performed
  • 31. Aggressive Lung ADC Subtype Prediction (Method & Results) ā€¢ Retrospectively enrolled 120 patients - Stage I lung ADC, ā‰¤2cm - Preoperative diagnostic CT and FDG PET/CT ā€¢ Histopathologic endpoint - Aggressiveness (Solid : 18 cases, MIP : 5 cases) ā€¢ 206 radiomic features & 14 clinical parameters ā€¢ SVM-LASSO model 31 Performance of the SVM-LASSO model to predict aggressive lung ADC Choi et al. Manuscript under review Box plots of SUVmax (FDR q=0.004) and PET Mean of Cluster Shade (q=0.002) Feature Sensitivity Specificity PPV NPV Accuracy AUC Conventional SUVmax 57.8Ā±4.6% 78.5Ā±1.4% 39.2Ā±2.3% 88.6Ā±1.1% 74.5Ā±1.4% 0.64Ā±0.01 SVM-LASSO PET Mean of Cluster Shade 67.4Ā±3.1% 86.0Ā±1.1% 53.7Ā±2.1% 91.7Ā±1.0% 82.4Ā±1.0% 0.78Ā±0.01 p-value SUVmax vs. SVM-LASSO 0.002 1e-5 7e-8 3e-5 5e-8 0.03
  • 32. Unsupervised Learning of Deep Learned Features from Breast Cancer Images 32 Lee, Choi et al. IEEE BIBE 2020
  • 33. Unsupervised Learning of Deep Learned Features 33 Slide name Silhouette optimal number Cluster set Accuracy F1-score TCGA-A7-A0DA 29 [25, 22, 6, 2, 24, 14, 0, 20, 10] 0.8829 0.8929 TCGA-A2-A0YM 20 [7, 6, 1, 5] 0.8360 0.8863 TCGA-A2-A3XT 19 [13, 10, 5, 1, 2] 0.9316 0.9514 TCGA-BH-A0BG 8 [1, 5] 0.7828 0.6857 TCGA-E2-A1LS 26 [18, 5] 0.8495 0.8680 TCGA-OL-A66I 25 [7, 18, 12, 20, 1, 0] 0.7761 0.7091 TCGA-C8-A26Y 21 [9, 18, 12, 16] 0.8594 0.8122 Lee, Choi et al. IEEE BIBE 2020
  • 34. PathCNN: interpretable convolutional neural networks for survival prediction and pathway analysis applied to glioblastoma 34 ā€¢ CNNs have achieved great success ā€¢ A lack of interpretability remains a key barrier ā€¢ Moreover, because biological array data are generally represented in a non-grid structured format ā€¢ PathCNN An interpretable CNN model on integrated multi-omics data using a newly defined pathway image. Oh, Choi et al. Bioinformatics 2021, joint first author Source code: https://github.com/mskspi/PathCNN.
  • 35. PathCNN: interpretable CNNs (results) 35 Cancer PathCNN Logistic regression SVM with RBF Neural network MiNet GBM 0.755ā€‰Ā±ā€‰0.009 0.668ā€‰Ā±ā€‰0.039 0.685ā€‰Ā±ā€‰0.037 0.692ā€‰Ā±ā€‰0.030 0.690ā€‰Ā±ā€‰0.032 LGG 0.877ā€‰Ā±ā€‰0.007 0.816ā€‰Ā±ā€‰0.036 0.884ā€‰Ā±ā€‰0.017 0.791ā€‰Ā±ā€‰0.031 0.854ā€‰Ā±ā€‰0.027 LUAD 0.637ā€‰Ā±ā€‰0.014 0.581ā€‰Ā±ā€‰0.028 0.624ā€‰Ā±ā€‰0.034 0.573ā€‰Ā±ā€‰0.031 0.597ā€‰Ā±ā€‰0.042 KIRC 0.709ā€‰Ā±ā€‰0.009 0.654ā€‰Ā±ā€‰0.034 0.684ā€‰Ā±ā€‰0.027 0.702ā€‰Ā±ā€‰0.028 0.659ā€‰Ā±ā€‰0.030 Comparison of predictive performance with benchmark methods in terms of the area under the curve (AUC: mean Ā± standard deviation) over 30 iterations of the 5-fold cross validation Note: AUCs for PathCNN were obtained with three principal components. Bold = Highest AUC for each dataset. SVM, support vector machine; RBF, radial basis function; MiNet, Multi-omics Integrative Net; GBM, glioblastoma multiforme; LGG, low-grade glioma; LUAD, lung adenocarcinoma; KIRC, kidney cancer. Oh, Choi et al. Bioinformatics 2021, joint first author Source code: https://github.com/mskspi/PathCNN.
  • 36. 36 A matrix of adjusted P-values. The row represents the 146 KEGG pathways ordered on pathway images. The columns represent the first two principal components of each omics type. The red color indicates key pathways with adjusted P-values < 0.001
  • 37. Outline ā€¢ Radiomics - Decision Support Tools - Lung Cancer Screening - Tumor Response Prediction and Evaluation - Aggressive Lung ADC subtype prediction - Multimodal data: Pathology, Multiomics, etc. ā€¢ Auto Delineation and Variability Analysis - Delineation Variability Quantification - Dosimetric Consequences of Variabilities - OARNet, Voxel2Mesh 37
  • 38. Delineation Variability Quantification and Simulation A framework for radiation therapy variability analysis 38 RT plan Structure Set CT Image Dose Distribution Structure Sets DV simulation ASSD GrowCut RW Other delineators SV analysis DV analysis Geometric Dosimetric Variability analysis Human DV Simulated DV Consensus SS OARNet Choi et al., AAPM, 2019.
  • 39. Delineation Variability Quantification and Simulation ā€¢ ESTRO Falcon contour workshop (EduCase) - A HNC case, Larynx, 70 Gy and 35 fractions - 14 independent manually delineated (MD) OAR structure sets (SS) - BrainStem, Esophagus, OralCavity, Parotid_L, Parotid_R, SpinalCord, and Thyroid ā€¢ Consensus MD SS - The simultaneous truth and performance level estimation (STAPLE) 39 Choi, Nourzadeh et al., AAPM, 2019.
  • 40. Delineation Variability Quantification and Simulation (Methods) ā€¢ Geometric analysis - Similarity: Dice coefficient (Volumetric, Surface) - Distance: Hausdorff distance (HD), Actual Average Surface Distance (AASD) - Reference: STAPLE SS ā€¢ Dosimetric analysis - Single dose distribution planned from a human SS - DVH confidence bands (90%tile) - !!"#$, !!#%, !!&$, !'( 40 Choi, Nourzadeh et al., AAPM, 2019.
  • 41. Delineation Variability Quantification and Simulation (Results) ā€¢ DVH variability not predicted by geometric measures ā€¢ Large human variability 41 100% 50% 0% 100% 50% 0% Human ASSD GrowCut RW Right Parotid Left Parotid Choi, Nourzadeh et al., AAPM, 2019.
  • 42. Human ASSD GrowCut RW 100% 50% 0% Education 100% 50% 0% Clinic 42
  • 43. ā€¢ Plan Competition Data Set: A HNC case, Nasopharynx , 70 Gy ā€¢ PV: 409 plans for IMRT, VMAT, and Tomotherapy on various TPS Eclipse, Monaco , Pinnacle, RayStation, Tomotherapy ā€¢ SV: Setup error simulation using Radiation Therapy Robustness Analyzer (RTRA) - 1000 simulations: 3mm translation and 5-degree rotation ā€¢ DV: 5 manually delineated (MD) SSs - BrainStem, Esophagus, OralCavity, Parotid_L, Parotid_R, and SpinalCord - 3x5 Simulated DVs using Radiation Therapy Variability Analyzer (RTVA) Variability Analysis Plan Variability, Setup Variability, Delineation Variability 43 Choi, Nourzadeh et al., AAPM, 2020.
  • 44. Dose Volume Coverage Map (DVCM) 44 Plan Variability Setup Variability Delineation Variability Choi, Nourzadeh et al., AAPM, 2020.
  • 45. DVCM Analysis 45 Fractional Volume Dose OAR Dose Constraint Probability Threshold 5% 0.3
  • 46. DVCM Summary 46 TV SVƗPV SVƗDV PVƗDV SV PV DV Average BrainStem 0.15 0.11 0.10 0.07 0.03 0.01 0.02 0.07 >0.95 Chiasm 0.68 0.00 0.65 0.56 0.00 0.00 0.49 0.34 Eye_L 0.07 0.05 0.04 0.01 0.02 0.01 0.01 0.03 Eye_R 0.02 0.01 0.01 0.01 0.00 0.00 0.01 0.01 >0.5 Lens_L 1.00 1.00 1.00 1.00 0.99 1.00 0.01 0.86 Lens_R 1.00 1.00 1.00 1.00 1.00 1.00 0.01 0.86 >0.3 Mandible 0.04 0.04 0.00 0.07 0.00 0.08 0.01 0.03 >0.2 OpticNerve_L 0.51 0.42 0.43 0.36 0.17 0.00 0.34 0.32 >0.05 OpticNerve_R 0.56 0.38 0.61 0.38 0.20 0.00 0.39 0.36 ā‰¤0.05 SpinalCord 0.13 0.12 0.05 0.05 0.04 0.01 0.02 0.06 =0.00 Average 0.42 0.31 0.39 0.35 0.25 0.21 0.13 0.29 Fractional volume affected by different variations when constraint failing probability > 5% (worst-case scenario) Choi, Nourzadeh et al., AAPM, 2020.
  • 47. OARNet: auto-delineate organs-at-risk (OARs) in head and neck (H&N) CT image 47 Soomro, Nourzadeh, Choi et al., arXiv:2108.13987.
  • 48. OARNet results 48 Soomro, Nourzadeh, Choi et al., arXiv:2108.13987.
  • 49. Interpretable Radiomics Toolkit End-to-End Deep Learning Model for Malignancy Prediction 49 Input Ground Truth Voxel2Mesh Choi et al., Manuscript in Preparation Network AUC Accuracy Sensitivity Specificity F1 LIDC-PM Mesh Only 0.937 83.33 77.78 88.89 82.35 Mesh+Encoder 0.903 88.89 91.67 86.11 89.19 LUNGx Mesh Only 0.711 63.33 73.33 53.33 66.67 Mesh+Encoder 0.687 53.3 83.3 23.33 64.11
  • 50. Summary ā€¢ Radiomics - Decision Support Tools - Lung Cancer Screening - Tumor Response Prediction and Evaluation - Aggressive Lung ADC subtype prediction - Multimodal data: Pathology, Multiomics, etc. ā€¢ Auto Delineation and Variability Analysis - Delineation Variability Quantification - Dosimetric Consequences of Variabilities 50
  • 51. Short-term Future Works ā€¢ Develop interpretable radiomic features - Improve spiculation quantification and multi-institution validation - Multimodal data integration ā€¢ Human-Variability aware auto-delineation - Variability quantification and simulation using generative models - AI-guided interactive delineation editing ā€¢ Integrate the radiomics framework into TPS - Eclipse (C#) and MIM (Python) 51
  • 52. Long-term Future Works ā€¢ Comprehensive Framework for Cancer Imaging - Multi-modal imaging - Response prediction and evaluation (Pre, Mid, and Post) - Longitudinal analysis of tumor change during treatment (MRgRT) - Shape analysis (e.g., Spiculation) - Deep learning models ā€¢ Automation of Clinical Workflow - Big Data Analytics: EMR, PACS, ROIS, Genomics, etc. - Provide an informatics platform for comprehensive cancer therapy 53
  • 53. Selected Publications 1. Jung Hun Oh*, Wookjin Choi* et al., ā€œPathCNN: interpretable convolutional neural networks for survival prediction and pathway analysis applied to glioblastomaā€, Bioinformatics, 2021, *joint first author 2. Wookjin Choi et al., ā€œ Reproducible and Interpretable Spiculation Quantification for Lung Cancer Screeningā€, Computer Methods and Programs in Biomedicineā€, 2021 3. Noemi Garau, Wookjin Choi, et al., ā€œ External validation of radiomics-based predictive models in low-dose CT screening for early lung cancer diagnosisā€, Medical Physics, 2020 4. Jiahui Wang, Wookjin Choi et al., ā€œPrediction of anal cancer recurrence after chemoradiotherapy using quantitative image features extracted from serial 18F-FDG PET/CTā€, Frontiers in oncology, 2019 5. Wookjin Choi et al., ā€œRadiomics analysis of pulmonary nodules in low-dose CT for early detection of lung cancerā€, Medical Physics, 2018 6. Sadegh Riyahi, Wookjin Choi, et al., ā€œQuantifying local tumor morphological changes with Jacobian map for prediction of pathologic tumor response to chemo-radiotherapy in locally advanced esophageal cancerā€, Physics in Medicine and Biology, 2018 7. Shan Tan, Laquan Li, Wookjin Choi, et al., ā€œAdaptive region-growing with maximum curvature strategy for tumor segmentation in 18F-FDG PETā€, Physics in Medicine and Biology, 2017 8. Wookjin Choi et al., ā€œIndividually Optimized Contrast-Enhanced 4D-CT for Radiotherapy Simulation in Pancreatic Ductal Adenocarcinomaā€, Medical Physics, 2016 9. Wookjin Choi, Tae-Sun Choi, ā€œAutomated Pulmonary Nodule Detection based on Three-dimensional Shape-based Feature Descriptorā€, Computer Methods and Programs in Biomedicine, 2014 54 Complete list of publications: https://scholar.google.com/citations?user=iHgsGLUAAAAJ
  • 54. Post-Doctoral Research Fellow Developing Interpretable Predictive Models for Radiation Therapy ā€¢ PI: Wookjin Choi, PhD - Wookjin.Choi@Jefferson.edu ā€¢ 2 Years ā€¢ Machine Learning/Deep Learning: Radiomics (PET/CT & MR) and Bioinformatics ā€¢ Computational Medical Physics: Development of Predictive Models and Automated Workflows, and Improve Clinical Workflow ā€¢ Internal or Extramural Research Funding Opportunities Qualifications ā€¢ Ph.D. in Computer Science, Electrical Engineering, Medical Physics, or related field required