SlideShare a Scribd company logo
AGENDA
1. How to assess & select a sucessful computer vision
POC ?
2. What kind of data+label+annotation is required for
each type of models ?
3. The complexity vs time to develop the model
4. Inspirational ideas
– case study (inputs-vs-outputs of different type of
CV models)
SELECTING &
ASSESSING A GOOD
COMPUTER
VISION(=CV) POC
COMPUTER VISION POC ASSESSMENT
•Find ares of skin lesion <- define skin lesion
•Is it binary ? i.e is it normal vs skin lesion  Yes
•Define concrete logic flow to diagnose skin lesion vs
normal
•Scope for timeplan , model deliverables ...etc
Break down problems into
definitions
•Can skin lesion be identified by naked eyes(via logic above)
•Are the skin lesion areas annotated properly ?(show how)
•Data volume and pos vs neg per images(complexity of the
problem)
Naked-eye judgement +
annotate corretly
•Can I augment images like this (show how) ?
•Were the digitalized image always stored in jpeg format ?
•Are skin lesion alwaysbe rounded and browish-black ?
•Difference between normal tatoo vs skin lesion ?
•Do you need to understand how the model is built
(interpretability)
Subject matters
• What’s the value gain once you have the model ( time-to-diagnose ?)
• How is it making an impact (to others potentially)?
•Secure plans of model maintenance & lifecycle management
What’s the value add &
maintenance
•Is the existing datasets capable of answering the problem stated ?
•Is the data quality good ” enough ” to proceed ?
•Do we have enough time & budget to execute this ?
•Is GDPR mitigated on the dataset?
•Is the Azure environment set up properly for Data science activities ?
Is the problem solvable ( = goal
achivable )?
Assessment
(iterative evaluation)
skin lesion as an example POC steps
ASK YOURSELF (AND CUSTOMERS, AND
PARTNERS ) ...
You know what problem
you want to solve ?
Find areas of skin lesion
Are you able to show
data+labels+anotations
that answer to the
question ?
Yes, I do, and
”this (image +
logic)” is how the
diagnose is being
made !
Nope, I have no idea
Partially , I know the
class but I didn’t mark
the area of skin lesion
I have no idea, I just
want to find something
useful
Ideas brain storming &
prioritization workshop
Problem
statement
Assessing
data & labels
Type of algorithm
considered
Self-assessing
problem
statement
SupervisedUnsupervise
d
Acquire
annotations
WHAT KIND OF
DATA+LABEL
(+MASK/ANNOTATION
)
DO I NEED PER TYPE
OF COMPUTER
VISION MODEL ?
Skin lesion example continued ...
CLASSIFICATION MODEL
Melanoma
Class label =1LABE
L
DATA
No Melanoma
Class label = 0
images labels
1
1
1
images labels
0
0
0
SEGMENTATION MODEL
Melanoma
Class label =1LABE
L
DATA
No Melanoma
Class label = 0
images labels
=masks
images Labels=masks
OBJECT DETECTION MODEL
(BOUNDING BOX)
Melanoma
Class label =1LABE
L
DATA
No Melanoma
Class label = 0
images Labels +
annotation
Usually in xml format
images labels +
annotation
Usually in xml format
1
1
1
0
0
0
SKIPPING GENERAL ADVERSIAL KIND
OF MDOELS & OTHER TYPES OF
MODEL SUCH AS VISUAL QUESTIONS
AND ANSWERING
MODEL+DATA
COMPLEXITY VS
TIME-TO-DEVELOP
THE MODEL
Consideration for POC time planning and budgeting
MODEL COMPLEXITY VS TIME-
NEEDED
(GENERAL RULE-OF-THUMB , START WITH THE SIMPLEST
MODEL)
model + data
complexity
Time needed
To develop the
model
1. Data volume
2. good enough data
quality + annotation to
proceed
-------------------shortest -----
-----------------
(depends on num_of_classes )
--------------mid-
range-----------
(depends on num_of_classes + mask
-------- logest-------
(depends on num_of_classes + annotatiuon
quality + kind of model + interpretability
CASES OF
COMPUTER
MODELS AND ITS
INPUTS-VS-
In - What you feed to the model ( data requirements)
Out – what the prediction looks like ( model
deliverables)
During Model training and model prediction, you
feed different data to the model
CLASSIFICATION
CASES
INPUTs :
(image + label)
Feed to
NN
(trainin
g)
predict
labels
CLASSIFYING PNEUMONIA/NORMAL
(X-RAY IMAGES)
The Neural Networks
Pneumonia Norma
l
Image 1 Image 2
label:
Image ID:
Image 1 Image 2Image ID:
input
Train the model
use trained model
Prediction
( only the
labels)
Image1 :
Pneumonia
Image2: Normal
Feed to
NN
(trainin
g)
predict
labels
CLASSIFYING PARKINSON/NORMAL
HAND-DRAWING
(HAND-DRAWING SPIRAL IMAGE)
The Neural Networks
norma
l
parkinso
n
Image 1 Image 2
label:
Image ID:
Image 1 Image 2Image ID:
input
Train the model
use trained model
Prediction
( only the
labels)
Image1 : normal
Image2 :
parkinson
Feed to
CNN
(trainin
g)
predict
labels
CLASSIFYING
CONTRAST/NO_CONTRAST CT SCAN
(DICOM IMAGES)
contra
st
No
contrast
Image 1 Image 2
label:
Image ID:
Image 1 Image 2Image ID:
input
Train the model
use trained model
Prediction
( only the
labels)
Image1 : contrast
Image2 : no
contrast
SEGMENTATION
CASES
Feed to
train
the
model
Model
predict
s
masks
SEGMENTING BRAIN LESION
Model Prediction
train the model
label:
(=mask)
Image ID:
Image ID:
Use trained model
input
Prediction
(the area of interest)
Feed to
train
the
model
Model
predict
s
masks
SEGMENTING ULTRASOUND
Model Prediction
train the model
label:
(=mask)
Image ID:
Image ID:
Use trained model
input
Prediction
(the area of interest)
Feed to
train
the
model
Model
predict
s
masks
SEGMENTING LUNGS
Model Prediction
train the model
label:
(=mask)
Image ID:
Image ID:
Use trained model
input
Prediction
(the area of interest)
Feed to
train
the
model
Model
predict
s
masks
SEGMENTING EYE VESSELS
Model Prediction
train the model
label:
(=mask)
Image ID:
Test Image ID:
Use trained model
input
Prediction
(the eye vessels)
REAL TIME
SEGMENTATION IN
VIDEO
REAL TIME ABDOMINAL ORGAN
SEGMENTATION
OBJECT DETECTION
CASES
Feed to
train
the
model
Model
predict
s
masks
OBJECT DETECTION - SKIN LESION
train the model
Image ID:
Image ID:
Use trained model
input
Prediction
(bounding box + class
)
rectangle: class:
28 37 160 156 1annotati
on
Detect type of blood cells -1. training
(annotated blood cells class + location (point-per-point )
Labels:
(1) Classes
(2) Location (points)
Image ID: bloodcell_img_0001.jpg
Train the model
Feed to
train
the
model
Input
image
Use trained model
Get
predicti
on
Input an image
Get prediction of
(1) Classes
(2) Bounding box of
location for each
class
Detect type of blood cells -2 prediction
(annotated blood cells class + location (point-per-point )
IMAGE TRANSLATION
CASES
Feed to
train
the
model
CYCLEGANS TRANSLATION EYE +
EYE VESSELS
- 1. DURING TRAINING Model Prediction
train the model
Paris of image A(=the eye) and B (= the vessel )
A B
Model
predict
s
masks
CYCLEGANS TRANSLATION EYE 
EYE VESSELS
- 2. DURING PREDICTION
Sent image A
Use trained
model
input
Prediction
direction: A B or B
A
Sent image B
A
B
B
A
Generate realistic looking ultrasound image
OTHER NOVEL
COMPUTER VISION
CASES
IMAGE TRANSLATION CASES
Tracking throat cancer in a video – used to extract interesting
frames
3D POINT-NET
Possible usage –
Instead of this furniture dataset, if we
could scan the entire body and obtain a
voxel grid kind of data , then we could
reconstruct a person’s 3D point cloud
similar to this demo in this slide.
The most common application is in VR/AR
domain, i.e used in Microsoft Hololens as
an example

More Related Content

Similar to DeepLearning Experiments in Medical Image show case

GDSC Unsri - From Zero to Hero with TensorFlow.pptx
GDSC Unsri - From Zero to Hero with TensorFlow.pptxGDSC Unsri - From Zero to Hero with TensorFlow.pptx
GDSC Unsri - From Zero to Hero with TensorFlow.pptx
FellyciaHikmahwarani
 
De duplication of entities with-in a cluster using image matching
De duplication of entities with-in a cluster using image matchingDe duplication of entities with-in a cluster using image matching
De duplication of entities with-in a cluster using image matching
Saurabh Singh
 
photo detection in personal photo collection
photo detection in personal photo collectionphoto detection in personal photo collection
photo detection in personal photo collection
sonalijagtap15
 
Machine Learning vs Decision Optimization comparison
Machine Learning vs Decision Optimization comparisonMachine Learning vs Decision Optimization comparison
Machine Learning vs Decision Optimization comparison
Alain Chabrier
 
IRJET- Survey on Face Recognition using Biometrics
IRJET-  	  Survey on Face Recognition using BiometricsIRJET-  	  Survey on Face Recognition using Biometrics
IRJET- Survey on Face Recognition using Biometrics
IRJET Journal
 
Introduction to Machine Learning
Introduction to Machine LearningIntroduction to Machine Learning
Introduction to Machine Learning
AI Summary
 
Study and Analysis of Novel Face Recognition Techniques using PCA, LDA and Ge...
Study and Analysis of Novel Face Recognition Techniques using PCA, LDA and Ge...Study and Analysis of Novel Face Recognition Techniques using PCA, LDA and Ge...
Study and Analysis of Novel Face Recognition Techniques using PCA, LDA and Ge...
sadique_ghitm
 
Report Face Detection
Report Face DetectionReport Face Detection
Report Face Detection
Jugal Patel
 
Image–based face-detection-and-recognition-using-matlab
Image–based face-detection-and-recognition-using-matlabImage–based face-detection-and-recognition-using-matlab
Image–based face-detection-and-recognition-using-matlab
Ijcem Journal
 
AI PROJECT CYCLE.pptx
AI PROJECT CYCLE.pptxAI PROJECT CYCLE.pptx
AI PROJECT CYCLE.pptx
Jitendra Kumar Yadav
 
Facial Recognition Vinod
Facial Recognition VinodFacial Recognition Vinod
Facial Recognition Vinod
Vinod siragaon
 
Machine Learning Interview Questions
Machine Learning Interview QuestionsMachine Learning Interview Questions
Machine Learning Interview Questions
Rock Interview
 
A Robust & Fast Face Detection System
A Robust & Fast Face Detection SystemA Robust & Fast Face Detection System
A Robust & Fast Face Detection System
IDES Editor
 
COVID-19-Preventions-Control-System and Unconstrained Face-mask and Face-hand...
COVID-19-Preventions-Control-System and Unconstrained Face-mask and Face-hand...COVID-19-Preventions-Control-System and Unconstrained Face-mask and Face-hand...
COVID-19-Preventions-Control-System and Unconstrained Face-mask and Face-hand...
SaiPrakash106
 
Module 5: Decision Trees
Module 5: Decision TreesModule 5: Decision Trees
Module 5: Decision Trees
Sara Hooker
 
LR2. Summary Day 2
LR2. Summary Day 2LR2. Summary Day 2
LR2. Summary Day 2
Machine Learning Valencia
 
Facerecognition
FacerecognitionFacerecognition
Facerecognition
venkata Anil
 
Facerecognition
FacerecognitionFacerecognition
Facerecognition
venkata Anil
 
inam
inaminam
notes as .ppt
notes as .pptnotes as .ppt
notes as .ppt
butest
 

Similar to DeepLearning Experiments in Medical Image show case (20)

GDSC Unsri - From Zero to Hero with TensorFlow.pptx
GDSC Unsri - From Zero to Hero with TensorFlow.pptxGDSC Unsri - From Zero to Hero with TensorFlow.pptx
GDSC Unsri - From Zero to Hero with TensorFlow.pptx
 
De duplication of entities with-in a cluster using image matching
De duplication of entities with-in a cluster using image matchingDe duplication of entities with-in a cluster using image matching
De duplication of entities with-in a cluster using image matching
 
photo detection in personal photo collection
photo detection in personal photo collectionphoto detection in personal photo collection
photo detection in personal photo collection
 
Machine Learning vs Decision Optimization comparison
Machine Learning vs Decision Optimization comparisonMachine Learning vs Decision Optimization comparison
Machine Learning vs Decision Optimization comparison
 
IRJET- Survey on Face Recognition using Biometrics
IRJET-  	  Survey on Face Recognition using BiometricsIRJET-  	  Survey on Face Recognition using Biometrics
IRJET- Survey on Face Recognition using Biometrics
 
Introduction to Machine Learning
Introduction to Machine LearningIntroduction to Machine Learning
Introduction to Machine Learning
 
Study and Analysis of Novel Face Recognition Techniques using PCA, LDA and Ge...
Study and Analysis of Novel Face Recognition Techniques using PCA, LDA and Ge...Study and Analysis of Novel Face Recognition Techniques using PCA, LDA and Ge...
Study and Analysis of Novel Face Recognition Techniques using PCA, LDA and Ge...
 
Report Face Detection
Report Face DetectionReport Face Detection
Report Face Detection
 
Image–based face-detection-and-recognition-using-matlab
Image–based face-detection-and-recognition-using-matlabImage–based face-detection-and-recognition-using-matlab
Image–based face-detection-and-recognition-using-matlab
 
AI PROJECT CYCLE.pptx
AI PROJECT CYCLE.pptxAI PROJECT CYCLE.pptx
AI PROJECT CYCLE.pptx
 
Facial Recognition Vinod
Facial Recognition VinodFacial Recognition Vinod
Facial Recognition Vinod
 
Machine Learning Interview Questions
Machine Learning Interview QuestionsMachine Learning Interview Questions
Machine Learning Interview Questions
 
A Robust & Fast Face Detection System
A Robust & Fast Face Detection SystemA Robust & Fast Face Detection System
A Robust & Fast Face Detection System
 
COVID-19-Preventions-Control-System and Unconstrained Face-mask and Face-hand...
COVID-19-Preventions-Control-System and Unconstrained Face-mask and Face-hand...COVID-19-Preventions-Control-System and Unconstrained Face-mask and Face-hand...
COVID-19-Preventions-Control-System and Unconstrained Face-mask and Face-hand...
 
Module 5: Decision Trees
Module 5: Decision TreesModule 5: Decision Trees
Module 5: Decision Trees
 
LR2. Summary Day 2
LR2. Summary Day 2LR2. Summary Day 2
LR2. Summary Day 2
 
Facerecognition
FacerecognitionFacerecognition
Facerecognition
 
Facerecognition
FacerecognitionFacerecognition
Facerecognition
 
inam
inaminam
inam
 
notes as .ppt
notes as .pptnotes as .ppt
notes as .ppt
 

More from Zenodia Charpy

how to build a Length of Stay model for a ProofOfConcept project
how to build a Length of Stay model for a ProofOfConcept projecthow to build a Length of Stay model for a ProofOfConcept project
how to build a Length of Stay model for a ProofOfConcept project
Zenodia Charpy
 
Tech Day Kista Mässa Stockholm 2018
Tech Day Kista Mässa Stockholm 2018Tech Day Kista Mässa Stockholm 2018
Tech Day Kista Mässa Stockholm 2018
Zenodia Charpy
 
Aiday
AidayAiday
Data Science on Azure
Data Science on Azure Data Science on Azure
Data Science on Azure
Zenodia Charpy
 
Zenodia TechDays talks Oct 24-25 Stockholm Kistamässan
Zenodia TechDays talks Oct 24-25 Stockholm KistamässanZenodia TechDays talks Oct 24-25 Stockholm Kistamässan
Zenodia TechDays talks Oct 24-25 Stockholm Kistamässan
Zenodia Charpy
 
Datascience and Azure(v1.0)
Datascience and Azure(v1.0)Datascience and Azure(v1.0)
Datascience and Azure(v1.0)
Zenodia Charpy
 
Göteborg university(condensed)
Göteborg university(condensed)Göteborg university(condensed)
Göteborg university(condensed)
Zenodia Charpy
 
Meetup sthlm - introduction to Machine Learning with demo cases
Meetup sthlm - introduction to Machine Learning with demo casesMeetup sthlm - introduction to Machine Learning with demo cases
Meetup sthlm - introduction to Machine Learning with demo cases
Zenodia Charpy
 

More from Zenodia Charpy (8)

how to build a Length of Stay model for a ProofOfConcept project
how to build a Length of Stay model for a ProofOfConcept projecthow to build a Length of Stay model for a ProofOfConcept project
how to build a Length of Stay model for a ProofOfConcept project
 
Tech Day Kista Mässa Stockholm 2018
Tech Day Kista Mässa Stockholm 2018Tech Day Kista Mässa Stockholm 2018
Tech Day Kista Mässa Stockholm 2018
 
Aiday
AidayAiday
Aiday
 
Data Science on Azure
Data Science on Azure Data Science on Azure
Data Science on Azure
 
Zenodia TechDays talks Oct 24-25 Stockholm Kistamässan
Zenodia TechDays talks Oct 24-25 Stockholm KistamässanZenodia TechDays talks Oct 24-25 Stockholm Kistamässan
Zenodia TechDays talks Oct 24-25 Stockholm Kistamässan
 
Datascience and Azure(v1.0)
Datascience and Azure(v1.0)Datascience and Azure(v1.0)
Datascience and Azure(v1.0)
 
Göteborg university(condensed)
Göteborg university(condensed)Göteborg university(condensed)
Göteborg university(condensed)
 
Meetup sthlm - introduction to Machine Learning with demo cases
Meetup sthlm - introduction to Machine Learning with demo casesMeetup sthlm - introduction to Machine Learning with demo cases
Meetup sthlm - introduction to Machine Learning with demo cases
 

Recently uploaded

Palo Alto Cortex XDR presentation .......
Palo Alto Cortex XDR presentation .......Palo Alto Cortex XDR presentation .......
Palo Alto Cortex XDR presentation .......
Sachin Paul
 
一比一原版兰加拉学院毕业证(Langara毕业证书)学历如何办理
一比一原版兰加拉学院毕业证(Langara毕业证书)学历如何办理一比一原版兰加拉学院毕业证(Langara毕业证书)学历如何办理
一比一原版兰加拉学院毕业证(Langara毕业证书)学历如何办理
hyfjgavov
 
writing report business partner b1+ .pdf
writing report business partner b1+ .pdfwriting report business partner b1+ .pdf
writing report business partner b1+ .pdf
VyNguyen709676
 
Intelligence supported media monitoring in veterinary medicine
Intelligence supported media monitoring in veterinary medicineIntelligence supported media monitoring in veterinary medicine
Intelligence supported media monitoring in veterinary medicine
AndrzejJarynowski
 
一比一原版(UO毕业证)渥太华大学毕业证如何办理
一比一原版(UO毕业证)渥太华大学毕业证如何办理一比一原版(UO毕业证)渥太华大学毕业证如何办理
一比一原版(UO毕业证)渥太华大学毕业证如何办理
aqzctr7x
 
Predictably Improve Your B2B Tech Company's Performance by Leveraging Data
Predictably Improve Your B2B Tech Company's Performance by Leveraging DataPredictably Improve Your B2B Tech Company's Performance by Leveraging Data
Predictably Improve Your B2B Tech Company's Performance by Leveraging Data
Kiwi Creative
 
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
bopyb
 
办(uts毕业证书)悉尼科技大学毕业证学历证书原版一模一样
办(uts毕业证书)悉尼科技大学毕业证学历证书原版一模一样办(uts毕业证书)悉尼科技大学毕业证学历证书原版一模一样
办(uts毕业证书)悉尼科技大学毕业证学历证书原版一模一样
apvysm8
 
原版制作(unimelb毕业证书)墨尔本大学毕业证Offer一模一样
原版制作(unimelb毕业证书)墨尔本大学毕业证Offer一模一样原版制作(unimelb毕业证书)墨尔本大学毕业证Offer一模一样
原版制作(unimelb毕业证书)墨尔本大学毕业证Offer一模一样
ihavuls
 
Udemy_2024_Global_Learning_Skills_Trends_Report (1).pdf
Udemy_2024_Global_Learning_Skills_Trends_Report (1).pdfUdemy_2024_Global_Learning_Skills_Trends_Report (1).pdf
Udemy_2024_Global_Learning_Skills_Trends_Report (1).pdf
Fernanda Palhano
 
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
taqyea
 
Global Situational Awareness of A.I. and where its headed
Global Situational Awareness of A.I. and where its headedGlobal Situational Awareness of A.I. and where its headed
Global Situational Awareness of A.I. and where its headed
vikram sood
 
The Ipsos - AI - Monitor 2024 Report.pdf
The  Ipsos - AI - Monitor 2024 Report.pdfThe  Ipsos - AI - Monitor 2024 Report.pdf
The Ipsos - AI - Monitor 2024 Report.pdf
Social Samosa
 
Analysis insight about a Flyball dog competition team's performance
Analysis insight about a Flyball dog competition team's performanceAnalysis insight about a Flyball dog competition team's performance
Analysis insight about a Flyball dog competition team's performance
roli9797
 
Learn SQL from basic queries to Advance queries
Learn SQL from basic queries to Advance queriesLearn SQL from basic queries to Advance queries
Learn SQL from basic queries to Advance queries
manishkhaire30
 
DATA COMMS-NETWORKS YR2 lecture 08 NAT & CLOUD.docx
DATA COMMS-NETWORKS YR2 lecture 08 NAT & CLOUD.docxDATA COMMS-NETWORKS YR2 lecture 08 NAT & CLOUD.docx
DATA COMMS-NETWORKS YR2 lecture 08 NAT & CLOUD.docx
SaffaIbrahim1
 
"Financial Odyssey: Navigating Past Performance Through Diverse Analytical Lens"
"Financial Odyssey: Navigating Past Performance Through Diverse Analytical Lens""Financial Odyssey: Navigating Past Performance Through Diverse Analytical Lens"
"Financial Odyssey: Navigating Past Performance Through Diverse Analytical Lens"
sameer shah
 
06-12-2024-BudapestDataForum-BuildingReal-timePipelineswithFLaNK AIM
06-12-2024-BudapestDataForum-BuildingReal-timePipelineswithFLaNK AIM06-12-2024-BudapestDataForum-BuildingReal-timePipelineswithFLaNK AIM
06-12-2024-BudapestDataForum-BuildingReal-timePipelineswithFLaNK AIM
Timothy Spann
 
4th Modern Marketing Reckoner by MMA Global India & Group M: 60+ experts on W...
4th Modern Marketing Reckoner by MMA Global India & Group M: 60+ experts on W...4th Modern Marketing Reckoner by MMA Global India & Group M: 60+ experts on W...
4th Modern Marketing Reckoner by MMA Global India & Group M: 60+ experts on W...
Social Samosa
 
在线办理(英国UCA毕业证书)创意艺术大学毕业证在读证明一模一样
在线办理(英国UCA毕业证书)创意艺术大学毕业证在读证明一模一样在线办理(英国UCA毕业证书)创意艺术大学毕业证在读证明一模一样
在线办理(英国UCA毕业证书)创意艺术大学毕业证在读证明一模一样
v7oacc3l
 

Recently uploaded (20)

Palo Alto Cortex XDR presentation .......
Palo Alto Cortex XDR presentation .......Palo Alto Cortex XDR presentation .......
Palo Alto Cortex XDR presentation .......
 
一比一原版兰加拉学院毕业证(Langara毕业证书)学历如何办理
一比一原版兰加拉学院毕业证(Langara毕业证书)学历如何办理一比一原版兰加拉学院毕业证(Langara毕业证书)学历如何办理
一比一原版兰加拉学院毕业证(Langara毕业证书)学历如何办理
 
writing report business partner b1+ .pdf
writing report business partner b1+ .pdfwriting report business partner b1+ .pdf
writing report business partner b1+ .pdf
 
Intelligence supported media monitoring in veterinary medicine
Intelligence supported media monitoring in veterinary medicineIntelligence supported media monitoring in veterinary medicine
Intelligence supported media monitoring in veterinary medicine
 
一比一原版(UO毕业证)渥太华大学毕业证如何办理
一比一原版(UO毕业证)渥太华大学毕业证如何办理一比一原版(UO毕业证)渥太华大学毕业证如何办理
一比一原版(UO毕业证)渥太华大学毕业证如何办理
 
Predictably Improve Your B2B Tech Company's Performance by Leveraging Data
Predictably Improve Your B2B Tech Company's Performance by Leveraging DataPredictably Improve Your B2B Tech Company's Performance by Leveraging Data
Predictably Improve Your B2B Tech Company's Performance by Leveraging Data
 
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
 
办(uts毕业证书)悉尼科技大学毕业证学历证书原版一模一样
办(uts毕业证书)悉尼科技大学毕业证学历证书原版一模一样办(uts毕业证书)悉尼科技大学毕业证学历证书原版一模一样
办(uts毕业证书)悉尼科技大学毕业证学历证书原版一模一样
 
原版制作(unimelb毕业证书)墨尔本大学毕业证Offer一模一样
原版制作(unimelb毕业证书)墨尔本大学毕业证Offer一模一样原版制作(unimelb毕业证书)墨尔本大学毕业证Offer一模一样
原版制作(unimelb毕业证书)墨尔本大学毕业证Offer一模一样
 
Udemy_2024_Global_Learning_Skills_Trends_Report (1).pdf
Udemy_2024_Global_Learning_Skills_Trends_Report (1).pdfUdemy_2024_Global_Learning_Skills_Trends_Report (1).pdf
Udemy_2024_Global_Learning_Skills_Trends_Report (1).pdf
 
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
一比一原版(harvard毕业证书)哈佛大学毕业证如何办理
 
Global Situational Awareness of A.I. and where its headed
Global Situational Awareness of A.I. and where its headedGlobal Situational Awareness of A.I. and where its headed
Global Situational Awareness of A.I. and where its headed
 
The Ipsos - AI - Monitor 2024 Report.pdf
The  Ipsos - AI - Monitor 2024 Report.pdfThe  Ipsos - AI - Monitor 2024 Report.pdf
The Ipsos - AI - Monitor 2024 Report.pdf
 
Analysis insight about a Flyball dog competition team's performance
Analysis insight about a Flyball dog competition team's performanceAnalysis insight about a Flyball dog competition team's performance
Analysis insight about a Flyball dog competition team's performance
 
Learn SQL from basic queries to Advance queries
Learn SQL from basic queries to Advance queriesLearn SQL from basic queries to Advance queries
Learn SQL from basic queries to Advance queries
 
DATA COMMS-NETWORKS YR2 lecture 08 NAT & CLOUD.docx
DATA COMMS-NETWORKS YR2 lecture 08 NAT & CLOUD.docxDATA COMMS-NETWORKS YR2 lecture 08 NAT & CLOUD.docx
DATA COMMS-NETWORKS YR2 lecture 08 NAT & CLOUD.docx
 
"Financial Odyssey: Navigating Past Performance Through Diverse Analytical Lens"
"Financial Odyssey: Navigating Past Performance Through Diverse Analytical Lens""Financial Odyssey: Navigating Past Performance Through Diverse Analytical Lens"
"Financial Odyssey: Navigating Past Performance Through Diverse Analytical Lens"
 
06-12-2024-BudapestDataForum-BuildingReal-timePipelineswithFLaNK AIM
06-12-2024-BudapestDataForum-BuildingReal-timePipelineswithFLaNK AIM06-12-2024-BudapestDataForum-BuildingReal-timePipelineswithFLaNK AIM
06-12-2024-BudapestDataForum-BuildingReal-timePipelineswithFLaNK AIM
 
4th Modern Marketing Reckoner by MMA Global India & Group M: 60+ experts on W...
4th Modern Marketing Reckoner by MMA Global India & Group M: 60+ experts on W...4th Modern Marketing Reckoner by MMA Global India & Group M: 60+ experts on W...
4th Modern Marketing Reckoner by MMA Global India & Group M: 60+ experts on W...
 
在线办理(英国UCA毕业证书)创意艺术大学毕业证在读证明一模一样
在线办理(英国UCA毕业证书)创意艺术大学毕业证在读证明一模一样在线办理(英国UCA毕业证书)创意艺术大学毕业证在读证明一模一样
在线办理(英国UCA毕业证书)创意艺术大学毕业证在读证明一模一样
 

DeepLearning Experiments in Medical Image show case

  • 1. AGENDA 1. How to assess & select a sucessful computer vision POC ? 2. What kind of data+label+annotation is required for each type of models ? 3. The complexity vs time to develop the model 4. Inspirational ideas – case study (inputs-vs-outputs of different type of CV models)
  • 2. SELECTING & ASSESSING A GOOD COMPUTER VISION(=CV) POC
  • 3. COMPUTER VISION POC ASSESSMENT •Find ares of skin lesion <- define skin lesion •Is it binary ? i.e is it normal vs skin lesion  Yes •Define concrete logic flow to diagnose skin lesion vs normal •Scope for timeplan , model deliverables ...etc Break down problems into definitions •Can skin lesion be identified by naked eyes(via logic above) •Are the skin lesion areas annotated properly ?(show how) •Data volume and pos vs neg per images(complexity of the problem) Naked-eye judgement + annotate corretly •Can I augment images like this (show how) ? •Were the digitalized image always stored in jpeg format ? •Are skin lesion alwaysbe rounded and browish-black ? •Difference between normal tatoo vs skin lesion ? •Do you need to understand how the model is built (interpretability) Subject matters • What’s the value gain once you have the model ( time-to-diagnose ?) • How is it making an impact (to others potentially)? •Secure plans of model maintenance & lifecycle management What’s the value add & maintenance •Is the existing datasets capable of answering the problem stated ? •Is the data quality good ” enough ” to proceed ? •Do we have enough time & budget to execute this ? •Is GDPR mitigated on the dataset? •Is the Azure environment set up properly for Data science activities ? Is the problem solvable ( = goal achivable )? Assessment (iterative evaluation) skin lesion as an example POC steps
  • 4. ASK YOURSELF (AND CUSTOMERS, AND PARTNERS ) ... You know what problem you want to solve ? Find areas of skin lesion Are you able to show data+labels+anotations that answer to the question ? Yes, I do, and ”this (image + logic)” is how the diagnose is being made ! Nope, I have no idea Partially , I know the class but I didn’t mark the area of skin lesion I have no idea, I just want to find something useful Ideas brain storming & prioritization workshop Problem statement Assessing data & labels Type of algorithm considered Self-assessing problem statement SupervisedUnsupervise d Acquire annotations
  • 5. WHAT KIND OF DATA+LABEL (+MASK/ANNOTATION ) DO I NEED PER TYPE OF COMPUTER VISION MODEL ? Skin lesion example continued ...
  • 6. CLASSIFICATION MODEL Melanoma Class label =1LABE L DATA No Melanoma Class label = 0 images labels 1 1 1 images labels 0 0 0
  • 7. SEGMENTATION MODEL Melanoma Class label =1LABE L DATA No Melanoma Class label = 0 images labels =masks images Labels=masks
  • 8. OBJECT DETECTION MODEL (BOUNDING BOX) Melanoma Class label =1LABE L DATA No Melanoma Class label = 0 images Labels + annotation Usually in xml format images labels + annotation Usually in xml format 1 1 1 0 0 0
  • 9. SKIPPING GENERAL ADVERSIAL KIND OF MDOELS & OTHER TYPES OF MODEL SUCH AS VISUAL QUESTIONS AND ANSWERING
  • 11. MODEL COMPLEXITY VS TIME- NEEDED (GENERAL RULE-OF-THUMB , START WITH THE SIMPLEST MODEL) model + data complexity Time needed To develop the model 1. Data volume 2. good enough data quality + annotation to proceed -------------------shortest ----- ----------------- (depends on num_of_classes ) --------------mid- range----------- (depends on num_of_classes + mask -------- logest------- (depends on num_of_classes + annotatiuon quality + kind of model + interpretability
  • 12. CASES OF COMPUTER MODELS AND ITS INPUTS-VS- In - What you feed to the model ( data requirements) Out – what the prediction looks like ( model deliverables) During Model training and model prediction, you feed different data to the model
  • 14. INPUTs : (image + label) Feed to NN (trainin g) predict labels CLASSIFYING PNEUMONIA/NORMAL (X-RAY IMAGES) The Neural Networks Pneumonia Norma l Image 1 Image 2 label: Image ID: Image 1 Image 2Image ID: input Train the model use trained model Prediction ( only the labels) Image1 : Pneumonia Image2: Normal
  • 15. Feed to NN (trainin g) predict labels CLASSIFYING PARKINSON/NORMAL HAND-DRAWING (HAND-DRAWING SPIRAL IMAGE) The Neural Networks norma l parkinso n Image 1 Image 2 label: Image ID: Image 1 Image 2Image ID: input Train the model use trained model Prediction ( only the labels) Image1 : normal Image2 : parkinson
  • 16. Feed to CNN (trainin g) predict labels CLASSIFYING CONTRAST/NO_CONTRAST CT SCAN (DICOM IMAGES) contra st No contrast Image 1 Image 2 label: Image ID: Image 1 Image 2Image ID: input Train the model use trained model Prediction ( only the labels) Image1 : contrast Image2 : no contrast
  • 18. Feed to train the model Model predict s masks SEGMENTING BRAIN LESION Model Prediction train the model label: (=mask) Image ID: Image ID: Use trained model input Prediction (the area of interest)
  • 19. Feed to train the model Model predict s masks SEGMENTING ULTRASOUND Model Prediction train the model label: (=mask) Image ID: Image ID: Use trained model input Prediction (the area of interest)
  • 20. Feed to train the model Model predict s masks SEGMENTING LUNGS Model Prediction train the model label: (=mask) Image ID: Image ID: Use trained model input Prediction (the area of interest)
  • 21. Feed to train the model Model predict s masks SEGMENTING EYE VESSELS Model Prediction train the model label: (=mask) Image ID: Test Image ID: Use trained model input Prediction (the eye vessels)
  • 23. REAL TIME ABDOMINAL ORGAN SEGMENTATION
  • 25. Feed to train the model Model predict s masks OBJECT DETECTION - SKIN LESION train the model Image ID: Image ID: Use trained model input Prediction (bounding box + class ) rectangle: class: 28 37 160 156 1annotati on
  • 26. Detect type of blood cells -1. training (annotated blood cells class + location (point-per-point ) Labels: (1) Classes (2) Location (points) Image ID: bloodcell_img_0001.jpg Train the model Feed to train the model
  • 27. Input image Use trained model Get predicti on Input an image Get prediction of (1) Classes (2) Bounding box of location for each class Detect type of blood cells -2 prediction (annotated blood cells class + location (point-per-point )
  • 29. Feed to train the model CYCLEGANS TRANSLATION EYE + EYE VESSELS - 1. DURING TRAINING Model Prediction train the model Paris of image A(=the eye) and B (= the vessel ) A B
  • 30. Model predict s masks CYCLEGANS TRANSLATION EYE  EYE VESSELS - 2. DURING PREDICTION Sent image A Use trained model input Prediction direction: A B or B A Sent image B A B B A
  • 31. Generate realistic looking ultrasound image
  • 33. IMAGE TRANSLATION CASES Tracking throat cancer in a video – used to extract interesting frames
  • 34. 3D POINT-NET Possible usage – Instead of this furniture dataset, if we could scan the entire body and obtain a voxel grid kind of data , then we could reconstruct a person’s 3D point cloud similar to this demo in this slide. The most common application is in VR/AR domain, i.e used in Microsoft Hololens as an example

Editor's Notes

  1. https://www.kaggle.com/kmader/electron-microscopy-3d-segmentation/home
  2. https://www.kaggle.com/kmader/electron-microscopy-3d-segmentation/home
  3. https://www.kaggle.com/kmader/electron-microscopy-3d-segmentation/home
  4. https://www.kaggle.com/kmader/electron-microscopy-3d-segmentation/home
  5. https://www.kaggle.com/kmader/electron-microscopy-3d-segmentation/home
  6. https://www.kaggle.com/kmader/electron-microscopy-3d-segmentation/home
  7. https://www.kaggle.com/kmader/electron-microscopy-3d-segmentation/home
  8. https://www.kaggle.com/kmader/electron-microscopy-3d-segmentation/home
  9. https://www.kaggle.com/kmader/electron-microscopy-3d-segmentation/home
  10. https://www.kaggle.com/kmader/electron-microscopy-3d-segmentation/home
  11. Niftynet demo on windows