SlideShare a Scribd company logo
1 of 11
Visual Recognition of Tissue Patterns on
Pathology Slides
web-pathology.net
FOR HEALTHCARE AND DRUG DISCOVERY
SMART IMAGING TECHNOLOGIES
Personalized Cancer Therapy: Knowledge Path
Personalized
Therapy
Medical
History and
Personal
Information
Genetic
Information
Pathology
Information
(tumor
biomarkers)
Personalized cancer therapy is a treatment strategy centered on
the ability to predict which patients are more likely to respond to
specific cancer therapies.
This approach is founded upon the idea that tumor biomarkers
are associated with patient prognosis and tumor response to
therapy.
In addition, patient genetic factors can be associated with drug
metabolism, drug response and drug toxicity.
Personalized tumor molecular profiles, tumor disease site and
other patient characteristics are then potentially used for
determining optimum individualized therapy options.
Source: MD Anderson Cancer Center
Pathology, the “study of disease”,
is an essential component for
analysis of personalized cancer
therapy options
Needs for Healthcare
FOR HEALTHCARE
Diagnostic Pattern Library: Applications and Benefits
Classified cancer pattern library is valuable digital asset that can
be licensed to other parties to train visual recognition and image
analysis algorithms.
Visual recognition application can be used to automatically
annotate digital pathology slides and link them with the rest of
institutional cancer knowledge base. This application can be
licensed to third parties to use for the same purposes.
Research and Clinical Applications:
• Computer-assisted cancer diagnosis with pre-screening,
suggestive diagnosis options and contextual links to cancer
knowledge libraries (similar cases, experts, research, additional
tests etc.)
• Data mining and of advanced analytics of historic tissue
samples for cancer patients with known outcomes with
purpose of building predictive knowledge bases for cancer
care and drug discovery
FOR HEALTHCARE
Researchers need to measure drug effects on tissue.
They also need to analyze relationships between effect on tissue and
all other relevant data in the drug research project.
Drug effects on tissue are expressed via biomarker patterns on tissue
slides. Researchers need to turn visual pattern expression into usable
data.
They need technology to:
• Objectify pattern definitions to deal with subjective biases of human
observers
• Search for similar biomarker expression patterns across archives
• Aggregate, analyze and organize pattern expression information
• Link pattern information to other “omnics” and project research data
Needs for Drug Discovery
FOR DRUG DISCOVERY
Utilizing Pathology Knowledge: The Challenge
Traditionally, pathology diagnosis is presented in descriptive
natural language statements. Often It is verbose professional
opinion of human expert with little quantitative information
Statistical agreement between human experts is 75%-85%.
Pathology diagnosis is rendered by pathologist observing
patterns of cells on tissue slide under the microscope. In order
to be useful for comparison and analysis these observations
must be:
• Quantified
• Objectified
This can be achieved (in theory) by digitizing pathology slide
and applying image analysis algorithms to quantify cell pattern
expressions.
Analyzing Pathology Slides: The Real Life Issues
In practice traditional image analysis approach often fall
short of expectations for number of reasons:
• Target image features are complex and difficult for
engineers to formalize and code.
• Algorithm development process is inherently
complicated: only pathologists know that patterns
they need but only image analysis engineers may
know how to extract them
• Tissues have variability and algorithms developed on
one sample may not work on others. Algorithms have
to be redesigned by engineers every time new “out of
range” samples are encountered
• Digital tissue slides are very large and processing every
pixel is extremely computationally expensive, making
“online” analysis impossible or impractical
Analyzing Pathology Slides: Machine Learning
Machine Learning Neural Networks learn to recognize images in the
same way humans do – by example, rather than by formalized
“handcrafted features”.
Since 2012 major improvement in visual recognition was achieved
with so called deep learning neural networks. Latest generation of
Visual Recognition Neural Networks achieve accuracy of recognition of
natural objects similar to human observers. This area of technology is
experiencing explosive growth.
Using Machine Learning brings number of advantages to visual
recognition applications:
• No need to formalize complex “handcrafted features”, pathologist
can just point to patterns they need to recognize
• No dependency on image analysis engineers (almost)
• System can be trained on very large number of samples to achieve
robust recognition
• New data samples can be added to model easily to increase
accuracy
Machine Learning: Requirements
Machine Learning approach to pattern recognition creates
new functional requirements for digital pathology software
• Robust visual recognition models need large number of training images which requires more
time for annotating than single pathologist can provide. This problem can be solved by
utilizing number of pathologists creating annotations for training visual recognition models
Collaborative Model Training (Crowdsourcing)
• Digital Pathology system should have capability for extracting specially formatted image
data sets on demand for training neural networks
Training Data Extraction
• Training of Neural Network requires massive parallel GPU computing power for a short time.
This scalable computing power can be economically delivered by scalable cloud
infrastructures such as IBM, Amazon or Microsoft.
Cloud Deployment
• Digital pathology software should be able to send imaging data to neural network
application for recognition and visualize responses for user.
API Integration and Visualization Interface
The Solution: Pattern Recognition with Machine Learning
• Last generation deep learning convolution networks can identify target tissue patters with 95%
accuracy
Deep Learning
• Pathologists can train recognition solution by simply annotating target tissue patters on slides
in their workspace
• They can easily set up classes of patterns for identification
Easy Training
• Robust solutions can be trained from multiple slides to identify target tissue patterns reliably
across large variety of samples
• Recognition models can be retrained easily if new patterns or different samples should be
added
Robust Recognition Models
• Slides in digital archives can be processed automatically for pattern detection and labeled
based on findings
• New slides can be analyzed and classified on upload with suggestive classification available
when human expert opens the slide
Automatic Processing
• Visualization overlays help quickly locate and review target patterns
• Visualization layer provides quantitative information about patterns
Advanced Visualization
• All data is stored in the database and available for search, data mining and analytics
Powerful Analytics
Our software can train neural networks and utilize latest deep learning visual recognition solutions
from best in class solution providers
The Solution: How it works
• Human experts mark areas with target patterns on digital slides in the web interface
• Multiple people across geographic locations can work on the same slide library at the
same time
• Software automatically prepares and extracts imaging data for training visual recognition
model in appropriate format
• Massive amount of training data can be produced quickly and efficiently with
crowdsourcing approach
1. Preparation of Training Data
• We train visual recognition models using proven templates and best-in-class machine
learning service providers
2. Training Visual Recognition Model
• Tissues on new slides coming to server are automatically recognized against known
taxonomy and labeled according to application logic.
3. Automatic Recognition Processing
• Users can mark the area and search for similar patterns on the same slide or across entire
slide archive
4. Search by Example
• Software shows identified patterns on slides using interactive “smart” overlays and
computes statistics
5. Viewing Results
• Users can correct visual recognition results by marking area with appropriate label. New
data will be added to the training set to improve recognition
6. Continuous Learning
Integration: Data Mining and Discovery
• Non SQL flexible indexed database architecture allows integrated
storage of different data items across multiple locations
Distributed Database
• Flexible structure allows storing and integrating various data in the
single information store
• New data can be added to database structure at any time
Comprehensive Data
• Selection and navigation is possible for any data item in the database
• Global search on any data is instant even for millions of items
Instant Search and Navigation
• Data items can be linked with external data sources and knowledge
bases such as diagnostic codes, SNOMED classifications or
proprietary knowledge bases
Data Linking
We provide instant search, navigation and data mining ability across millions of slides

More Related Content

Viewers also liked

A 3-D Riesz-Covariance Texture Model for the Prediction of Nodule Recurrence ...
A 3-D Riesz-Covariance Texture Model for the Prediction of Nodule Recurrence ...A 3-D Riesz-Covariance Texture Model for the Prediction of Nodule Recurrence ...
A 3-D Riesz-Covariance Texture Model for the Prediction of Nodule Recurrence ...Institute of Information Systems (HES-SO)
 
Computer aided detection of pulmonary nodules using genetic programming
Computer aided detection of pulmonary nodules using genetic programmingComputer aided detection of pulmonary nodules using genetic programming
Computer aided detection of pulmonary nodules using genetic programmingWookjin Choi
 
Radioterapi of lung cancer
Radioterapi of lung cancerRadioterapi of lung cancer
Radioterapi of lung cancerMulkan Fadhli
 
L.T.D second seminar
L.T.D second seminarL.T.D second seminar
L.T.D second seminarFatmaSamy
 
Conceptual indexing
Conceptual indexingConceptual indexing
Conceptual indexingSajan Sahu
 
Phrase Based Indexing and Information Retrivel
Phrase Based Indexing and Information RetrivelPhrase Based Indexing and Information Retrivel
Phrase Based Indexing and Information Retrivelbalaabirami
 
폐 CT영상에서 voxel classification을 이용한 폐 결절 검출
폐 CT영상에서 voxel classification을 이용한 폐 결절 검출폐 CT영상에서 voxel classification을 이용한 폐 결절 검출
폐 CT영상에서 voxel classification을 이용한 폐 결절 검출Wookjin Choi
 
A Comparison of Different Strategies for Automated Semantic Document Annotation
A Comparison of Different Strategies for Automated Semantic Document AnnotationA Comparison of Different Strategies for Automated Semantic Document Annotation
A Comparison of Different Strategies for Automated Semantic Document AnnotationAnsgar Scherp
 
Introduction to Machine Learning @ Mooncascade ML Camp
Introduction to Machine Learning @ Mooncascade ML CampIntroduction to Machine Learning @ Mooncascade ML Camp
Introduction to Machine Learning @ Mooncascade ML CampIlya Kuzovkin
 
Rule based approach to sentiment analysis at romip’11 slides
Rule based approach to sentiment analysis at romip’11 slidesRule based approach to sentiment analysis at romip’11 slides
Rule based approach to sentiment analysis at romip’11 slidesDmitry Kan
 
Semantic Analysis using Wikipedia Taxonomy
Semantic Analysis using Wikipedia TaxonomySemantic Analysis using Wikipedia Taxonomy
Semantic Analysis using Wikipedia TaxonomyPatrick Nicolas
 

Viewers also liked (13)

Sle 衛教講座 chen
Sle 衛教講座 chenSle 衛教講座 chen
Sle 衛教講座 chen
 
A 3-D Riesz-Covariance Texture Model for the Prediction of Nodule Recurrence ...
A 3-D Riesz-Covariance Texture Model for the Prediction of Nodule Recurrence ...A 3-D Riesz-Covariance Texture Model for the Prediction of Nodule Recurrence ...
A 3-D Riesz-Covariance Texture Model for the Prediction of Nodule Recurrence ...
 
Computer aided detection of pulmonary nodules using genetic programming
Computer aided detection of pulmonary nodules using genetic programmingComputer aided detection of pulmonary nodules using genetic programming
Computer aided detection of pulmonary nodules using genetic programming
 
Radioterapi of lung cancer
Radioterapi of lung cancerRadioterapi of lung cancer
Radioterapi of lung cancer
 
L.T.D second seminar
L.T.D second seminarL.T.D second seminar
L.T.D second seminar
 
Conceptual indexing
Conceptual indexingConceptual indexing
Conceptual indexing
 
Phrase Based Indexing and Information Retrivel
Phrase Based Indexing and Information RetrivelPhrase Based Indexing and Information Retrivel
Phrase Based Indexing and Information Retrivel
 
폐 CT영상에서 voxel classification을 이용한 폐 결절 검출
폐 CT영상에서 voxel classification을 이용한 폐 결절 검출폐 CT영상에서 voxel classification을 이용한 폐 결절 검출
폐 CT영상에서 voxel classification을 이용한 폐 결절 검출
 
A Comparison of Different Strategies for Automated Semantic Document Annotation
A Comparison of Different Strategies for Automated Semantic Document AnnotationA Comparison of Different Strategies for Automated Semantic Document Annotation
A Comparison of Different Strategies for Automated Semantic Document Annotation
 
FUNDAMENTALS OF TEXTURE PROCESSING FOR BIOMEDICAL IMAGE ANALYSIS
FUNDAMENTALS OF TEXTURE PROCESSING FOR BIOMEDICAL IMAGE ANALYSISFUNDAMENTALS OF TEXTURE PROCESSING FOR BIOMEDICAL IMAGE ANALYSIS
FUNDAMENTALS OF TEXTURE PROCESSING FOR BIOMEDICAL IMAGE ANALYSIS
 
Introduction to Machine Learning @ Mooncascade ML Camp
Introduction to Machine Learning @ Mooncascade ML CampIntroduction to Machine Learning @ Mooncascade ML Camp
Introduction to Machine Learning @ Mooncascade ML Camp
 
Rule based approach to sentiment analysis at romip’11 slides
Rule based approach to sentiment analysis at romip’11 slidesRule based approach to sentiment analysis at romip’11 slides
Rule based approach to sentiment analysis at romip’11 slides
 
Semantic Analysis using Wikipedia Taxonomy
Semantic Analysis using Wikipedia TaxonomySemantic Analysis using Wikipedia Taxonomy
Semantic Analysis using Wikipedia Taxonomy
 

Recently uploaded

Call Girls Service In Goa 💋 9316020077💋 Goa Call Girls By Russian Call Girl...
Call Girls Service In Goa  💋 9316020077💋 Goa Call Girls  By Russian Call Girl...Call Girls Service In Goa  💋 9316020077💋 Goa Call Girls  By Russian Call Girl...
Call Girls Service In Goa 💋 9316020077💋 Goa Call Girls By Russian Call Girl...russian goa call girl and escorts service
 
❤️♀️@ Jaipur Call Girls ❤️♀️@ Jaispreet Call Girl Services in Jaipur QRYPCF ...
❤️♀️@ Jaipur Call Girls ❤️♀️@ Jaispreet Call Girl Services in Jaipur QRYPCF  ...❤️♀️@ Jaipur Call Girls ❤️♀️@ Jaispreet Call Girl Services in Jaipur QRYPCF  ...
❤️♀️@ Jaipur Call Girls ❤️♀️@ Jaispreet Call Girl Services in Jaipur QRYPCF ...Gfnyt.com
 
VIP Call Girl Sector 10 Noida Call Me: 9711199171
VIP Call Girl Sector 10 Noida Call Me: 9711199171VIP Call Girl Sector 10 Noida Call Me: 9711199171
VIP Call Girl Sector 10 Noida Call Me: 9711199171Call Girls Service Gurgaon
 
(Sonam Bajaj) Call Girl in Jaipur- 09257276172 Escorts Service 50% Off with C...
(Sonam Bajaj) Call Girl in Jaipur- 09257276172 Escorts Service 50% Off with C...(Sonam Bajaj) Call Girl in Jaipur- 09257276172 Escorts Service 50% Off with C...
(Sonam Bajaj) Call Girl in Jaipur- 09257276172 Escorts Service 50% Off with C...indiancallgirl4rent
 
VIP Call Girls Noida Sia 9711199171 High Class Call Girl Near Me
VIP Call Girls Noida Sia 9711199171 High Class Call Girl Near MeVIP Call Girls Noida Sia 9711199171 High Class Call Girl Near Me
VIP Call Girls Noida Sia 9711199171 High Class Call Girl Near Memriyagarg453
 
VIP Call Girl Sector 88 Gurgaon Delhi Just Call Me 9899900591
VIP Call Girl Sector 88 Gurgaon Delhi Just Call Me 9899900591VIP Call Girl Sector 88 Gurgaon Delhi Just Call Me 9899900591
VIP Call Girl Sector 88 Gurgaon Delhi Just Call Me 9899900591adityaroy0215
 
raisen Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meet
raisen Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meetraisen Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meet
raisen Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real MeetCall Girls Service
 
Krishnagiri call girls Tamil aunty 7877702510
Krishnagiri call girls Tamil aunty 7877702510Krishnagiri call girls Tamil aunty 7877702510
Krishnagiri call girls Tamil aunty 7877702510Vipesco
 
(Ajay) Call Girls in Dehradun- 8854095900 Escorts Service 50% Off with Cash O...
(Ajay) Call Girls in Dehradun- 8854095900 Escorts Service 50% Off with Cash O...(Ajay) Call Girls in Dehradun- 8854095900 Escorts Service 50% Off with Cash O...
(Ajay) Call Girls in Dehradun- 8854095900 Escorts Service 50% Off with Cash O...indiancallgirl4rent
 
Bareilly Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meet
Bareilly Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real MeetBareilly Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meet
Bareilly Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real MeetCall Girls Service
 
(Deeksha) 💓 9920725232 💓High Profile Call Girls Navi Mumbai You Can Get The S...
(Deeksha) 💓 9920725232 💓High Profile Call Girls Navi Mumbai You Can Get The S...(Deeksha) 💓 9920725232 💓High Profile Call Girls Navi Mumbai You Can Get The S...
(Deeksha) 💓 9920725232 💓High Profile Call Girls Navi Mumbai You Can Get The S...Ahmedabad Call Girls
 
Enjoyment ★ 8854095900 Indian Call Girls In Dehradun 🍆🍌 By Dehradun Call Girl ★
Enjoyment ★ 8854095900 Indian Call Girls In Dehradun 🍆🍌 By Dehradun Call Girl ★Enjoyment ★ 8854095900 Indian Call Girls In Dehradun 🍆🍌 By Dehradun Call Girl ★
Enjoyment ★ 8854095900 Indian Call Girls In Dehradun 🍆🍌 By Dehradun Call Girl ★indiancallgirl4rent
 
VIP Call Girl DLF Phase 2 Gurgaon (Noida) Just Meet Me@ 9711199012
VIP Call Girl DLF Phase 2 Gurgaon (Noida) Just Meet Me@ 9711199012VIP Call Girl DLF Phase 2 Gurgaon (Noida) Just Meet Me@ 9711199012
VIP Call Girl DLF Phase 2 Gurgaon (Noida) Just Meet Me@ 9711199012adityaroy0215
 
❤️Call girls in Jalandhar ☎️9876848877☎️ Call Girl service in Jalandhar☎️ Jal...
❤️Call girls in Jalandhar ☎️9876848877☎️ Call Girl service in Jalandhar☎️ Jal...❤️Call girls in Jalandhar ☎️9876848877☎️ Call Girl service in Jalandhar☎️ Jal...
❤️Call girls in Jalandhar ☎️9876848877☎️ Call Girl service in Jalandhar☎️ Jal...chandigarhentertainm
 
Call Girls Chandigarh 👙 7001035870 👙 Genuine WhatsApp Number for Real Meet
Call Girls Chandigarh 👙 7001035870 👙 Genuine WhatsApp Number for Real MeetCall Girls Chandigarh 👙 7001035870 👙 Genuine WhatsApp Number for Real Meet
Call Girls Chandigarh 👙 7001035870 👙 Genuine WhatsApp Number for Real Meetpriyashah722354
 
Call Now ☎ 9999965857 !! Call Girls in Hauz Khas Escort Service Delhi N.C.R.
Call Now ☎ 9999965857 !! Call Girls in Hauz Khas Escort Service Delhi N.C.R.Call Now ☎ 9999965857 !! Call Girls in Hauz Khas Escort Service Delhi N.C.R.
Call Now ☎ 9999965857 !! Call Girls in Hauz Khas Escort Service Delhi N.C.R.ktanvi103
 
Call Girl Raipur 📲 9999965857 ヅ10k NiGhT Call Girls In Raipur
Call Girl Raipur 📲 9999965857 ヅ10k NiGhT Call Girls In RaipurCall Girl Raipur 📲 9999965857 ヅ10k NiGhT Call Girls In Raipur
Call Girl Raipur 📲 9999965857 ヅ10k NiGhT Call Girls In Raipurgragmanisha42
 
VIP Call Girl Sector 32 Noida Just Book Me 9711199171
VIP Call Girl Sector 32 Noida Just Book Me 9711199171VIP Call Girl Sector 32 Noida Just Book Me 9711199171
VIP Call Girl Sector 32 Noida Just Book Me 9711199171Call Girls Service Gurgaon
 
Call Girls Hyderabad Just Call 9907093804 Top Class Call Girl Service Available
Call Girls Hyderabad Just Call 9907093804 Top Class Call Girl Service AvailableCall Girls Hyderabad Just Call 9907093804 Top Class Call Girl Service Available
Call Girls Hyderabad Just Call 9907093804 Top Class Call Girl Service AvailableDipal Arora
 

Recently uploaded (20)

Call Girls Service In Goa 💋 9316020077💋 Goa Call Girls By Russian Call Girl...
Call Girls Service In Goa  💋 9316020077💋 Goa Call Girls  By Russian Call Girl...Call Girls Service In Goa  💋 9316020077💋 Goa Call Girls  By Russian Call Girl...
Call Girls Service In Goa 💋 9316020077💋 Goa Call Girls By Russian Call Girl...
 
❤️♀️@ Jaipur Call Girls ❤️♀️@ Jaispreet Call Girl Services in Jaipur QRYPCF ...
❤️♀️@ Jaipur Call Girls ❤️♀️@ Jaispreet Call Girl Services in Jaipur QRYPCF  ...❤️♀️@ Jaipur Call Girls ❤️♀️@ Jaispreet Call Girl Services in Jaipur QRYPCF  ...
❤️♀️@ Jaipur Call Girls ❤️♀️@ Jaispreet Call Girl Services in Jaipur QRYPCF ...
 
VIP Call Girl Sector 10 Noida Call Me: 9711199171
VIP Call Girl Sector 10 Noida Call Me: 9711199171VIP Call Girl Sector 10 Noida Call Me: 9711199171
VIP Call Girl Sector 10 Noida Call Me: 9711199171
 
(Sonam Bajaj) Call Girl in Jaipur- 09257276172 Escorts Service 50% Off with C...
(Sonam Bajaj) Call Girl in Jaipur- 09257276172 Escorts Service 50% Off with C...(Sonam Bajaj) Call Girl in Jaipur- 09257276172 Escorts Service 50% Off with C...
(Sonam Bajaj) Call Girl in Jaipur- 09257276172 Escorts Service 50% Off with C...
 
VIP Call Girls Noida Sia 9711199171 High Class Call Girl Near Me
VIP Call Girls Noida Sia 9711199171 High Class Call Girl Near MeVIP Call Girls Noida Sia 9711199171 High Class Call Girl Near Me
VIP Call Girls Noida Sia 9711199171 High Class Call Girl Near Me
 
VIP Call Girl Sector 88 Gurgaon Delhi Just Call Me 9899900591
VIP Call Girl Sector 88 Gurgaon Delhi Just Call Me 9899900591VIP Call Girl Sector 88 Gurgaon Delhi Just Call Me 9899900591
VIP Call Girl Sector 88 Gurgaon Delhi Just Call Me 9899900591
 
raisen Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meet
raisen Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meetraisen Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meet
raisen Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meet
 
Krishnagiri call girls Tamil aunty 7877702510
Krishnagiri call girls Tamil aunty 7877702510Krishnagiri call girls Tamil aunty 7877702510
Krishnagiri call girls Tamil aunty 7877702510
 
(Ajay) Call Girls in Dehradun- 8854095900 Escorts Service 50% Off with Cash O...
(Ajay) Call Girls in Dehradun- 8854095900 Escorts Service 50% Off with Cash O...(Ajay) Call Girls in Dehradun- 8854095900 Escorts Service 50% Off with Cash O...
(Ajay) Call Girls in Dehradun- 8854095900 Escorts Service 50% Off with Cash O...
 
9316020077📞Goa Call Girls Numbers, Call Girls Whatsapp Numbers Goa
9316020077📞Goa  Call Girls  Numbers, Call Girls  Whatsapp Numbers Goa9316020077📞Goa  Call Girls  Numbers, Call Girls  Whatsapp Numbers Goa
9316020077📞Goa Call Girls Numbers, Call Girls Whatsapp Numbers Goa
 
Bareilly Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meet
Bareilly Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real MeetBareilly Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meet
Bareilly Call Girls 👙 6297143586 👙 Genuine WhatsApp Number for Real Meet
 
(Deeksha) 💓 9920725232 💓High Profile Call Girls Navi Mumbai You Can Get The S...
(Deeksha) 💓 9920725232 💓High Profile Call Girls Navi Mumbai You Can Get The S...(Deeksha) 💓 9920725232 💓High Profile Call Girls Navi Mumbai You Can Get The S...
(Deeksha) 💓 9920725232 💓High Profile Call Girls Navi Mumbai You Can Get The S...
 
Enjoyment ★ 8854095900 Indian Call Girls In Dehradun 🍆🍌 By Dehradun Call Girl ★
Enjoyment ★ 8854095900 Indian Call Girls In Dehradun 🍆🍌 By Dehradun Call Girl ★Enjoyment ★ 8854095900 Indian Call Girls In Dehradun 🍆🍌 By Dehradun Call Girl ★
Enjoyment ★ 8854095900 Indian Call Girls In Dehradun 🍆🍌 By Dehradun Call Girl ★
 
VIP Call Girl DLF Phase 2 Gurgaon (Noida) Just Meet Me@ 9711199012
VIP Call Girl DLF Phase 2 Gurgaon (Noida) Just Meet Me@ 9711199012VIP Call Girl DLF Phase 2 Gurgaon (Noida) Just Meet Me@ 9711199012
VIP Call Girl DLF Phase 2 Gurgaon (Noida) Just Meet Me@ 9711199012
 
❤️Call girls in Jalandhar ☎️9876848877☎️ Call Girl service in Jalandhar☎️ Jal...
❤️Call girls in Jalandhar ☎️9876848877☎️ Call Girl service in Jalandhar☎️ Jal...❤️Call girls in Jalandhar ☎️9876848877☎️ Call Girl service in Jalandhar☎️ Jal...
❤️Call girls in Jalandhar ☎️9876848877☎️ Call Girl service in Jalandhar☎️ Jal...
 
Call Girls Chandigarh 👙 7001035870 👙 Genuine WhatsApp Number for Real Meet
Call Girls Chandigarh 👙 7001035870 👙 Genuine WhatsApp Number for Real MeetCall Girls Chandigarh 👙 7001035870 👙 Genuine WhatsApp Number for Real Meet
Call Girls Chandigarh 👙 7001035870 👙 Genuine WhatsApp Number for Real Meet
 
Call Now ☎ 9999965857 !! Call Girls in Hauz Khas Escort Service Delhi N.C.R.
Call Now ☎ 9999965857 !! Call Girls in Hauz Khas Escort Service Delhi N.C.R.Call Now ☎ 9999965857 !! Call Girls in Hauz Khas Escort Service Delhi N.C.R.
Call Now ☎ 9999965857 !! Call Girls in Hauz Khas Escort Service Delhi N.C.R.
 
Call Girl Raipur 📲 9999965857 ヅ10k NiGhT Call Girls In Raipur
Call Girl Raipur 📲 9999965857 ヅ10k NiGhT Call Girls In RaipurCall Girl Raipur 📲 9999965857 ヅ10k NiGhT Call Girls In Raipur
Call Girl Raipur 📲 9999965857 ヅ10k NiGhT Call Girls In Raipur
 
VIP Call Girl Sector 32 Noida Just Book Me 9711199171
VIP Call Girl Sector 32 Noida Just Book Me 9711199171VIP Call Girl Sector 32 Noida Just Book Me 9711199171
VIP Call Girl Sector 32 Noida Just Book Me 9711199171
 
Call Girls Hyderabad Just Call 9907093804 Top Class Call Girl Service Available
Call Girls Hyderabad Just Call 9907093804 Top Class Call Girl Service AvailableCall Girls Hyderabad Just Call 9907093804 Top Class Call Girl Service Available
Call Girls Hyderabad Just Call 9907093804 Top Class Call Girl Service Available
 

Deep Learning Histology Pattern Recognition for Healthcare and Drug Discovery

  • 1. Visual Recognition of Tissue Patterns on Pathology Slides web-pathology.net FOR HEALTHCARE AND DRUG DISCOVERY SMART IMAGING TECHNOLOGIES
  • 2. Personalized Cancer Therapy: Knowledge Path Personalized Therapy Medical History and Personal Information Genetic Information Pathology Information (tumor biomarkers) Personalized cancer therapy is a treatment strategy centered on the ability to predict which patients are more likely to respond to specific cancer therapies. This approach is founded upon the idea that tumor biomarkers are associated with patient prognosis and tumor response to therapy. In addition, patient genetic factors can be associated with drug metabolism, drug response and drug toxicity. Personalized tumor molecular profiles, tumor disease site and other patient characteristics are then potentially used for determining optimum individualized therapy options. Source: MD Anderson Cancer Center Pathology, the “study of disease”, is an essential component for analysis of personalized cancer therapy options Needs for Healthcare FOR HEALTHCARE
  • 3. Diagnostic Pattern Library: Applications and Benefits Classified cancer pattern library is valuable digital asset that can be licensed to other parties to train visual recognition and image analysis algorithms. Visual recognition application can be used to automatically annotate digital pathology slides and link them with the rest of institutional cancer knowledge base. This application can be licensed to third parties to use for the same purposes. Research and Clinical Applications: • Computer-assisted cancer diagnosis with pre-screening, suggestive diagnosis options and contextual links to cancer knowledge libraries (similar cases, experts, research, additional tests etc.) • Data mining and of advanced analytics of historic tissue samples for cancer patients with known outcomes with purpose of building predictive knowledge bases for cancer care and drug discovery FOR HEALTHCARE
  • 4. Researchers need to measure drug effects on tissue. They also need to analyze relationships between effect on tissue and all other relevant data in the drug research project. Drug effects on tissue are expressed via biomarker patterns on tissue slides. Researchers need to turn visual pattern expression into usable data. They need technology to: • Objectify pattern definitions to deal with subjective biases of human observers • Search for similar biomarker expression patterns across archives • Aggregate, analyze and organize pattern expression information • Link pattern information to other “omnics” and project research data Needs for Drug Discovery FOR DRUG DISCOVERY
  • 5. Utilizing Pathology Knowledge: The Challenge Traditionally, pathology diagnosis is presented in descriptive natural language statements. Often It is verbose professional opinion of human expert with little quantitative information Statistical agreement between human experts is 75%-85%. Pathology diagnosis is rendered by pathologist observing patterns of cells on tissue slide under the microscope. In order to be useful for comparison and analysis these observations must be: • Quantified • Objectified This can be achieved (in theory) by digitizing pathology slide and applying image analysis algorithms to quantify cell pattern expressions.
  • 6. Analyzing Pathology Slides: The Real Life Issues In practice traditional image analysis approach often fall short of expectations for number of reasons: • Target image features are complex and difficult for engineers to formalize and code. • Algorithm development process is inherently complicated: only pathologists know that patterns they need but only image analysis engineers may know how to extract them • Tissues have variability and algorithms developed on one sample may not work on others. Algorithms have to be redesigned by engineers every time new “out of range” samples are encountered • Digital tissue slides are very large and processing every pixel is extremely computationally expensive, making “online” analysis impossible or impractical
  • 7. Analyzing Pathology Slides: Machine Learning Machine Learning Neural Networks learn to recognize images in the same way humans do – by example, rather than by formalized “handcrafted features”. Since 2012 major improvement in visual recognition was achieved with so called deep learning neural networks. Latest generation of Visual Recognition Neural Networks achieve accuracy of recognition of natural objects similar to human observers. This area of technology is experiencing explosive growth. Using Machine Learning brings number of advantages to visual recognition applications: • No need to formalize complex “handcrafted features”, pathologist can just point to patterns they need to recognize • No dependency on image analysis engineers (almost) • System can be trained on very large number of samples to achieve robust recognition • New data samples can be added to model easily to increase accuracy
  • 8. Machine Learning: Requirements Machine Learning approach to pattern recognition creates new functional requirements for digital pathology software • Robust visual recognition models need large number of training images which requires more time for annotating than single pathologist can provide. This problem can be solved by utilizing number of pathologists creating annotations for training visual recognition models Collaborative Model Training (Crowdsourcing) • Digital Pathology system should have capability for extracting specially formatted image data sets on demand for training neural networks Training Data Extraction • Training of Neural Network requires massive parallel GPU computing power for a short time. This scalable computing power can be economically delivered by scalable cloud infrastructures such as IBM, Amazon or Microsoft. Cloud Deployment • Digital pathology software should be able to send imaging data to neural network application for recognition and visualize responses for user. API Integration and Visualization Interface
  • 9. The Solution: Pattern Recognition with Machine Learning • Last generation deep learning convolution networks can identify target tissue patters with 95% accuracy Deep Learning • Pathologists can train recognition solution by simply annotating target tissue patters on slides in their workspace • They can easily set up classes of patterns for identification Easy Training • Robust solutions can be trained from multiple slides to identify target tissue patterns reliably across large variety of samples • Recognition models can be retrained easily if new patterns or different samples should be added Robust Recognition Models • Slides in digital archives can be processed automatically for pattern detection and labeled based on findings • New slides can be analyzed and classified on upload with suggestive classification available when human expert opens the slide Automatic Processing • Visualization overlays help quickly locate and review target patterns • Visualization layer provides quantitative information about patterns Advanced Visualization • All data is stored in the database and available for search, data mining and analytics Powerful Analytics Our software can train neural networks and utilize latest deep learning visual recognition solutions from best in class solution providers
  • 10. The Solution: How it works • Human experts mark areas with target patterns on digital slides in the web interface • Multiple people across geographic locations can work on the same slide library at the same time • Software automatically prepares and extracts imaging data for training visual recognition model in appropriate format • Massive amount of training data can be produced quickly and efficiently with crowdsourcing approach 1. Preparation of Training Data • We train visual recognition models using proven templates and best-in-class machine learning service providers 2. Training Visual Recognition Model • Tissues on new slides coming to server are automatically recognized against known taxonomy and labeled according to application logic. 3. Automatic Recognition Processing • Users can mark the area and search for similar patterns on the same slide or across entire slide archive 4. Search by Example • Software shows identified patterns on slides using interactive “smart” overlays and computes statistics 5. Viewing Results • Users can correct visual recognition results by marking area with appropriate label. New data will be added to the training set to improve recognition 6. Continuous Learning
  • 11. Integration: Data Mining and Discovery • Non SQL flexible indexed database architecture allows integrated storage of different data items across multiple locations Distributed Database • Flexible structure allows storing and integrating various data in the single information store • New data can be added to database structure at any time Comprehensive Data • Selection and navigation is possible for any data item in the database • Global search on any data is instant even for millions of items Instant Search and Navigation • Data items can be linked with external data sources and knowledge bases such as diagnostic codes, SNOMED classifications or proprietary knowledge bases Data Linking We provide instant search, navigation and data mining ability across millions of slides