Artificial Neural Networks applications in Computer Aided Diagnosis. System design and use as an educational tool

Technological Ecosystems for Enhancing Multiculturality
Technological Ecosystems for Enhancing MulticulturalityTechnological Ecosystems for Enhancing Multiculturality
Artificial Neural Networks applications
in Computer Aided Diagnosis. System
design and use as an educational tool.
Jorge Hernández Rodríguez
Education in the Knowledge Society PhD Program
Hernández Rodríguez, Jorge; Rodríguez Conde, María José; Cabrero Fraile, Francisco Javier
Track 16: Doctoral Consortium
2
INDEX
● Context and motivation that drives the dissertation research
● State of the art
● Hypothesis and problem statement
● Research objective and goals
● Research approaches and methods
● Results to date and their validity. Dissertation status.
● Current and expected contributions
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
3
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
CONTEXT AND MOTIVATION THAT DRIVES THE DISSERTATION RESEARCH (I)
Computer Aided
Detection and
Diagnosis (CAD)
Software and algorithms (image
processing, lesion detection
and classification algorithms)
“Second opinion” for the
radiologist
CADe CADx
Software specialized in
lesion and pathological
features detection
Software specialized in
pathology characterization ,
classification or diagnosis
Radiological image
modalities
Computed Tomography
Mammography
Conventional Radiology
Magnetic Resonance Imaging
Ultrasound
…
4
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
CONTEXT AND MOTIVATION THAT DRIVES THE DISSERTATION RESEARCH (II)
CAD UTILITY
● Increasing number of high technology equipment in hospitals
and clinics
● Radiological examinations with a high number of images and
rising number of patients scanned (CT, MRI, screening,
interventional radiology…)
● Growth of clinical indications for different types of modalities
¡ Huge amount of
workload for the
radiologist!
CAD: very useful tool to ease
workload and improve detection
and diagnosis and therefore
posterior treatment prescription.
Reported in scientific and technical literature:
► The utility of these systems has been confirmed in numerous articles
► Reduction of inter-observer variability associated with image interpretation
► Sensitivity and specificity improvement associated with its use
► Improvement in specialists’ Receiver Operating Characteristic curves
► They provide supplementary information for managing a problem
► They have been successfully integrated into Picture Archiving and
Communication Systems and Radiology Information Systems
OBJECTIVE
■ Assisting radiologists in
detection and diagnosis
■ Great potential for medical
specialists’ training and as an
educational tool
5
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
STATE OF THE ART (I)
Artificial
Intelligence
Artificial
Neural
Networks
(ANNs)
6
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
STATE OF THE ART (II)
Evolution of ANNs in Medical Imaging
↓
Convolutional Neural Networks (CNNs)
Neural Networks based in Deep Learning Methods
↓
Highly non-linear systems
Designed to work directly with images
Deeper architectures (higher number of layers)
Reduction in the amount of adjustable network parameters
Limited size of training and validation datasets
↓
Calculation of empirical features of segmented lesions in images (manually or automatically)
Use of a classification algorithm
(for example, linear discriminant analysis, support vector machines, artificial neural networks
“Traditional CAD schemes”
7
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
STATE OF THE ART (III): Mammography
Arévalo et al. (2016).
Computers and Methods in Biomedicine
CNNs based in Deep learning techniques
Mammography database: BCDR
Preprocessing
• Image cropping
• Data augmentation
• Global Contrast Normalization
• Local Contrast Normalization
Sample of
marked
lesion
Supervised learning
through CNN and
classification
Architecture of the CNN that performed better
Evaluation of results
Comparison with other
state-of-the art
representations for
lesion classification and
image analysis
8
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
STATE OF THE ART (IV): Pulmonary node classification
Cheng et al. (2016).
Scientific reports - Nature
Deep learning
techniques
SDAE classifier achieved better results
Patterns from
classifier’s
hidden layers
Examples of
rated nodules
Shen et al. (2015). Information
processing in medical imaging Multi-scale CNNs
Classification accuracy of 86,4% on LIDC-IDRI database
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
STATE OF THE ART (V): Computer-Aided Detection
Roth et.al. (2016).
IEEE Transactions on Medical Imaging
Convolutional Neural Networks
● Hierarchical two-tiered CADe system
● New approach: 2,5 D image decomposition
● Wide range of applications:
Sclerotic spine metastases
ThoracoAbdominal Lymph nodes
Colonyc polyps
● Different candidate generation algorithms
● False positive reduction schemes
Detection of sclerotic metastases Detection of lymph nodes
Influence of Random View number on performance Performance on training and testing datasets of different
approaches. Lymph node detection.
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
STATE OF THE ART (VI): CAD applications in medical education
Mazurowski et al. (2010). Medical Physics Zhang et al. (2014). Medical Physics Ping et al. (2005). Academic Radiology
Hypermedia instructional program in
CAD-aided mammography training
Individually adapted computer-aided
educational system in mammography
Construction of user models.
Probability of making diagnostic errors by
radiology residents (difficulty of cases).
↓
Relation with image features
Computer models to predict locations of
false-positives in mammography
Based on previous annotations
11
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
HYPOTHESIS AND PROBLEM STATEMENT (I)
FIRST STAGE: CAD SYSTEM DEVELOPMENT
Creation of a CAD system for radiological image analysis
Analysis of system’s performance and efficiency in pathology detection and classification.
HYPOTHESES
► The system developed is a useful tool for assisting radiologists as a “second image reader”.
► System’s sensitivity and specificity is comparable to other published results.
► The number of false positives detected is kept in reasonable numbers, similar to published values.
► Detection and classification accuracy on different image datasets reaches reported values.
► Software is capable of extracting useful image features and information for the specialist.
12
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
SECOND STAGE: SYSTEM VALIDATION
Evaluation of system’s effectiveness as a clinical, educational and training tool for medicine
students and future medical specialists.
HYPOTHESES
► The CAD system designed and trained with validated image datasets produces positive effects in the
analysis and diagnosis of radiological images.
► The CAD scheme developed together with its functionalities will serve as a useful and complementary
educational tool for medicine students learning to interpret medical images and for specialists’ training in
the field of diagnostic radiology.
HYPOTHESIS AND PROBLEM STATEMENT (II)
13
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
RESEARCH OBJECTIVES AND GOALS (I)
MAIN OBJECTIVE
Development and validation of an ANN based CAD scheme
Extensible to different image modalities and pathologies
(pulmonary nodes in CT, polyps in CT colonography, mammography…)
With tools to access and classify the CAD generated information and
educational functionalities
14
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
RESEARCH OBJECTIVES AND GOALS (II)
SPECIFIC
OBJECTIVES
1. Revision of CAD schemes, models and algorithms.
2. CAD scheme design.
3. Download validated image databases and annotations.
4. Project development from neural network training.
5. Integration of ANNs in a CAD platform.
6. System validation prior to clinical use.
7. System validation for educational purposes and specialists training.
15
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
STAGE 1: Development and validation of an ANN based CAD scheme
CAD system’s training,
adjustment and validation
Public medical
image databases
Available for
educational, research
and software
development purposes
They contain clinical and
diagnostic information
related with the cases
and pathologies
considered
Information validated by
groups of experts and
with different methods
and techniques
Image repositories for
different modalities and
pathologies
RESEARCH APPROACHES AND METHODS (I)
16
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
MAMMOGRAPHY
Medical image databases
THORAX
RADIOGRAPHY
CANCER IMAGE DATASETS
(DIFFERENT IMAGE MODALITIES)
RESEARCH APPROACHES AND METHODS (II)
17
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
Calculation environment for
working with medical images
and neural networks
Convolutional Neural
Networks Toolbox for
artificial vision applications
Image
Processing
Toolbox
Computer vision
algorithm library
(open source code)
RESEARCH APPROACHES AND METHODS (III)
Software and calculations
18
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
General purpose
programming language
Package of functions dedicated for
scientific computing with PythonTM
Python library for defining, optimizing
and evaluating multidimensional array
expressions (suited for ANNs)
Python software
development environment
Possibility of performing
GPU accelerated
calculations with these
environments
RESEARCH APPROACHES AND METHODS (IV)
19
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
RESEARCH APPROACHES AND METHODS (V)
Validation of ANNs performance
● Split of datasets into three groups: training, validation and testing
● Monitoring network parameter’s during iterative processes
● Calculation of parameters to quantify network’s performance
● Statistical analysis of the results produced in the testing phase
● Cross-validation combining different ANNs in an ensemble
- Increase the testing samples to ensure optimal tuning (if needed)
20
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
RESEARCH APPROACHES AND METHODS (VI)
STAGE 2: System validation for clinical use and utility as an educational tool
● Use in clinical practice by specialists
● Training in software operation
● Analysis of cases with software and result interpretation
● Study of CAD influence on reader’s performance
● Statistical analysis of a sample of cases
● Analysis of clinical cases by radiology residents and
medicine students
● Training in software operation and educational tools
● Surveys to evaluate goals during training period
● Study of CAD influence on learning process
● Statistical analysis of surveys’ results
21
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
RESULTS TO DATE AND THEIR VALIDITY. DISSERTATION STATUS
● Bibliographic review in the areas of CAD and ANNs.
● Installation of software environments and function libraries.
● Download of image datasets and annotations.
● Formatting of images and preprocessing steps. Classification of cases.
● Initial design of CAD application. Initial routines and scripts.
● Selection of network architecture and layer types.
22
Track 16: Artificial Neural Networks application in Computer Aided Diagnosis.
System design and use as an educational tool
CURRENT AND EXPECTED CONTRIBUTIONS
● ANN training from parameter value optimization based on annotations
● Comprehensive analysis of ANNs performance. Analysis of the influence of several architecture parameters, learning
algorithms, sample sets size and type, preprocessing steps… on it.
● Implementation of tools to allow learning through CAD use. Design of specific validation tests.
● Validation of the CAD scheme in clinical practice: interaction with the use and clinical data generated
● Educational validation derived from the use of the system by medicine students and radiology residents.
● Optimization of the system from information extracted from practical use
● Future contributions published in the next TEEM conference and journals in the field of medical imaging, computers in
medicine or radiology and medical education
Thank you very much for your attention!
1 de 23

Recomendados

Artificial Intelligence in Medical Imaging por
Artificial Intelligence in Medical ImagingArtificial Intelligence in Medical Imaging
Artificial Intelligence in Medical ImagingZahidulIslamJewel2
281 vistas3 diapositivas
Prospects of Deep Learning in Medical Imaging por
Prospects of Deep Learning in Medical ImagingProspects of Deep Learning in Medical Imaging
Prospects of Deep Learning in Medical ImagingGodswll Egegwu
500 vistas12 diapositivas
Medical image analysis and big data evaluation infrastructures por
Medical image analysis and big data evaluation infrastructuresMedical image analysis and big data evaluation infrastructures
Medical image analysis and big data evaluation infrastructuresInstitute of Information Systems (HES-SO)
1.8K vistas40 diapositivas
Rapid COVID-19 Diagnosis Using Deep Learning of the Computerized Tomography ... por
Rapid COVID-19 Diagnosis Using Deep Learning  of the Computerized Tomography ...Rapid COVID-19 Diagnosis Using Deep Learning  of the Computerized Tomography ...
Rapid COVID-19 Diagnosis Using Deep Learning of the Computerized Tomography ...Dr. Amir Mosavi, PhD., P.Eng.
134 vistas8 diapositivas
Artificial intelligence in medical image processing por
Artificial intelligence in medical image processingArtificial intelligence in medical image processing
Artificial intelligence in medical image processingFarzad Jahedi
6.9K vistas17 diapositivas
MedGIFT projects in medical imaging por
MedGIFT projects in medical imagingMedGIFT projects in medical imaging
MedGIFT projects in medical imagingInstitute of Information Systems (HES-SO)
899 vistas37 diapositivas

Más contenido relacionado

La actualidad más candente

A Biomedical Information Retrieval System based on Clustering for Mobile Dev... por
A Biomedical Information Retrieval System  based on Clustering for Mobile Dev...A Biomedical Information Retrieval System  based on Clustering for Mobile Dev...
A Biomedical Information Retrieval System based on Clustering for Mobile Dev...Manuel de la Villa
640 vistas28 diapositivas
BreastScreening: On the Use of Multi-Modality in Medical Imaging Diagnosis por
BreastScreening: On the Use of Multi-Modality in Medical Imaging DiagnosisBreastScreening: On the Use of Multi-Modality in Medical Imaging Diagnosis
BreastScreening: On the Use of Multi-Modality in Medical Imaging DiagnosisInstituto Superior Técnico
77 vistas58 diapositivas
Machine Learning in Modern Medicine with Erin LeDell at Stanford Med por
Machine Learning in Modern Medicine with Erin LeDell at Stanford MedMachine Learning in Modern Medicine with Erin LeDell at Stanford Med
Machine Learning in Modern Medicine with Erin LeDell at Stanford MedSri Ambati
5.2K vistas46 diapositivas
Medical Multimedia Information Systems (ACMMM17 Tutorial) por
Medical Multimedia Information Systems (ACMMM17 Tutorial) Medical Multimedia Information Systems (ACMMM17 Tutorial)
Medical Multimedia Information Systems (ACMMM17 Tutorial) klschoef
2.6K vistas91 diapositivas
Transparency in ML and AI (humble views from a concerned academic) por
Transparency in ML and AI (humble views from a concerned academic)Transparency in ML and AI (humble views from a concerned academic)
Transparency in ML and AI (humble views from a concerned academic)Paolo Missier
579 vistas26 diapositivas
Slima explainable deep learning using fuzzy logic human ist u fribourg ver 17... por
Slima explainable deep learning using fuzzy logic human ist u fribourg ver 17...Slima explainable deep learning using fuzzy logic human ist u fribourg ver 17...
Slima explainable deep learning using fuzzy logic human ist u fribourg ver 17...Servio Fernando Lima Reina
156 vistas41 diapositivas

La actualidad más candente(10)

A Biomedical Information Retrieval System based on Clustering for Mobile Dev... por Manuel de la Villa
A Biomedical Information Retrieval System  based on Clustering for Mobile Dev...A Biomedical Information Retrieval System  based on Clustering for Mobile Dev...
A Biomedical Information Retrieval System based on Clustering for Mobile Dev...
Manuel de la Villa640 vistas
Machine Learning in Modern Medicine with Erin LeDell at Stanford Med por Sri Ambati
Machine Learning in Modern Medicine with Erin LeDell at Stanford MedMachine Learning in Modern Medicine with Erin LeDell at Stanford Med
Machine Learning in Modern Medicine with Erin LeDell at Stanford Med
Sri Ambati5.2K vistas
Medical Multimedia Information Systems (ACMMM17 Tutorial) por klschoef
Medical Multimedia Information Systems (ACMMM17 Tutorial) Medical Multimedia Information Systems (ACMMM17 Tutorial)
Medical Multimedia Information Systems (ACMMM17 Tutorial)
klschoef2.6K vistas
Transparency in ML and AI (humble views from a concerned academic) por Paolo Missier
Transparency in ML and AI (humble views from a concerned academic)Transparency in ML and AI (humble views from a concerned academic)
Transparency in ML and AI (humble views from a concerned academic)
Paolo Missier579 vistas
Slima explainable deep learning using fuzzy logic human ist u fribourg ver 17... por Servio Fernando Lima Reina
Slima explainable deep learning using fuzzy logic human ist u fribourg ver 17...Slima explainable deep learning using fuzzy logic human ist u fribourg ver 17...
Slima explainable deep learning using fuzzy logic human ist u fribourg ver 17...
Machine Learning in Pathology Diagnostics with Simagis Live por khvatkov
Machine Learning in Pathology Diagnostics with Simagis LiveMachine Learning in Pathology Diagnostics with Simagis Live
Machine Learning in Pathology Diagnostics with Simagis Live
khvatkov1.7K vistas
Simagis for healthcare por khvatkov
Simagis for healthcareSimagis for healthcare
Simagis for healthcare
khvatkov465 vistas
International Journal of Computational Engineering Research(IJCER) por ijceronline
International Journal of Computational Engineering Research(IJCER)International Journal of Computational Engineering Research(IJCER)
International Journal of Computational Engineering Research(IJCER)
ijceronline366 vistas
IRJET- Chest Abnormality Detection from X-Ray using Deep Learning por IRJET Journal
IRJET- Chest Abnormality Detection from X-Ray using Deep LearningIRJET- Chest Abnormality Detection from X-Ray using Deep Learning
IRJET- Chest Abnormality Detection from X-Ray using Deep Learning
IRJET Journal18 vistas

Destacado

Artificial neural network por
Artificial neural networkArtificial neural network
Artificial neural networkDEEPASHRI HK
186.8K vistas22 diapositivas
Learning ecosystems for knowledge management in educational institutions por
Learning ecosystems for knowledge management in educational institutions Learning ecosystems for knowledge management in educational institutions
Learning ecosystems for knowledge management in educational institutions Technological Ecosystems for Enhancing Multiculturality
190 vistas21 diapositivas
Neural network & its applications por
Neural network & its applications Neural network & its applications
Neural network & its applications Ahmed_hashmi
195.3K vistas50 diapositivas
Turing variations por
Turing variationsTuring variations
Turing variationsShiraz316
182 vistas60 diapositivas
Challenge-based gamification as a teaching’Open Educational Innovation strate... por
Challenge-based gamification as a teaching’Open Educational Innovation strate...Challenge-based gamification as a teaching’Open Educational Innovation strate...
Challenge-based gamification as a teaching’Open Educational Innovation strate...Technological Ecosystems for Enhancing Multiculturality
133 vistas17 diapositivas
Creating quality standards for scientific content in digital environments thr... por
Creating quality standards for scientific content in digital environments thr...Creating quality standards for scientific content in digital environments thr...
Creating quality standards for scientific content in digital environments thr...Technological Ecosystems for Enhancing Multiculturality
110 vistas11 diapositivas

Destacado(20)

Artificial neural network por DEEPASHRI HK
Artificial neural networkArtificial neural network
Artificial neural network
DEEPASHRI HK186.8K vistas
Neural network & its applications por Ahmed_hashmi
Neural network & its applications Neural network & its applications
Neural network & its applications
Ahmed_hashmi195.3K vistas
Turing variations por Shiraz316
Turing variationsTuring variations
Turing variations
Shiraz316182 vistas

Similar a Artificial Neural Networks applications in Computer Aided Diagnosis. System design and use as an educational tool

Artificial Intelligence in Medicine por
Artificial Intelligence in MedicineArtificial Intelligence in Medicine
Artificial Intelligence in MedicineNancy Gertrudiz
765 vistas19 diapositivas
Diabetic Retinopathy.pptx por
Diabetic Retinopathy.pptxDiabetic Retinopathy.pptx
Diabetic Retinopathy.pptxNGOKUL3
279 vistas34 diapositivas
Diabetic Retinopathy.pptx por
Diabetic Retinopathy.pptxDiabetic Retinopathy.pptx
Diabetic Retinopathy.pptxNGOKUL3
1.9K vistas34 diapositivas
IRJET- A Survey on Medical Image Interpretation for Predicting Pneumonia por
IRJET- A Survey on Medical Image Interpretation for Predicting PneumoniaIRJET- A Survey on Medical Image Interpretation for Predicting Pneumonia
IRJET- A Survey on Medical Image Interpretation for Predicting PneumoniaIRJET Journal
12 vistas3 diapositivas
RETINAL IMAGE CLASSIFICATION USING NEURAL NETWORK BASED ON A CNN METHODS por
RETINAL IMAGE CLASSIFICATION USING NEURAL NETWORK BASED ON A CNN METHODSRETINAL IMAGE CLASSIFICATION USING NEURAL NETWORK BASED ON A CNN METHODS
RETINAL IMAGE CLASSIFICATION USING NEURAL NETWORK BASED ON A CNN METHODSIRJET Journal
11 vistas5 diapositivas
Meta analysis of convolutional neural networks for radiological images - Pubrica por
Meta analysis of convolutional neural networks for radiological images - PubricaMeta analysis of convolutional neural networks for radiological images - Pubrica
Meta analysis of convolutional neural networks for radiological images - PubricaPubrica
16 vistas3 diapositivas

Similar a Artificial Neural Networks applications in Computer Aided Diagnosis. System design and use as an educational tool(20)

Artificial Intelligence in Medicine por Nancy Gertrudiz
Artificial Intelligence in MedicineArtificial Intelligence in Medicine
Artificial Intelligence in Medicine
Nancy Gertrudiz765 vistas
Diabetic Retinopathy.pptx por NGOKUL3
Diabetic Retinopathy.pptxDiabetic Retinopathy.pptx
Diabetic Retinopathy.pptx
NGOKUL3279 vistas
Diabetic Retinopathy.pptx por NGOKUL3
Diabetic Retinopathy.pptxDiabetic Retinopathy.pptx
Diabetic Retinopathy.pptx
NGOKUL31.9K vistas
IRJET- A Survey on Medical Image Interpretation for Predicting Pneumonia por IRJET Journal
IRJET- A Survey on Medical Image Interpretation for Predicting PneumoniaIRJET- A Survey on Medical Image Interpretation for Predicting Pneumonia
IRJET- A Survey on Medical Image Interpretation for Predicting Pneumonia
IRJET Journal12 vistas
RETINAL IMAGE CLASSIFICATION USING NEURAL NETWORK BASED ON A CNN METHODS por IRJET Journal
RETINAL IMAGE CLASSIFICATION USING NEURAL NETWORK BASED ON A CNN METHODSRETINAL IMAGE CLASSIFICATION USING NEURAL NETWORK BASED ON A CNN METHODS
RETINAL IMAGE CLASSIFICATION USING NEURAL NETWORK BASED ON A CNN METHODS
IRJET Journal11 vistas
Meta analysis of convolutional neural networks for radiological images - Pubrica por Pubrica
Meta analysis of convolutional neural networks for radiological images - PubricaMeta analysis of convolutional neural networks for radiological images - Pubrica
Meta analysis of convolutional neural networks for radiological images - Pubrica
Pubrica 16 vistas
Basic image analysis(processing and classification) and visualization using m... por Vishwas N
Basic image analysis(processing and classification) and visualization using m...Basic image analysis(processing and classification) and visualization using m...
Basic image analysis(processing and classification) and visualization using m...
Vishwas N116 vistas
An Introduction To Artificial Intelligence And Its Applications In Biomedical... por Jill Brown
An Introduction To Artificial Intelligence And Its Applications In Biomedical...An Introduction To Artificial Intelligence And Its Applications In Biomedical...
An Introduction To Artificial Intelligence And Its Applications In Biomedical...
Jill Brown10 vistas
APPLICATION OF CNN MODEL ON MEDICAL IMAGE por IRJET Journal
APPLICATION OF CNN MODEL ON MEDICAL IMAGEAPPLICATION OF CNN MODEL ON MEDICAL IMAGE
APPLICATION OF CNN MODEL ON MEDICAL IMAGE
IRJET Journal14 vistas
A study of different tools for medical image processing por Swarada Kanap
A study of different tools for medical image processingA study of different tools for medical image processing
A study of different tools for medical image processing
Swarada Kanap153 vistas
Image recognition por Aseed Usmani
Image recognitionImage recognition
Image recognition
Aseed Usmani9.5K vistas
Development of Computational Tool for Lung Cancer Prediction Using Data Mining por Editor IJCATR
Development of Computational Tool for Lung Cancer Prediction Using Data MiningDevelopment of Computational Tool for Lung Cancer Prediction Using Data Mining
Development of Computational Tool for Lung Cancer Prediction Using Data Mining
Editor IJCATR175 vistas
Intelligent data analysis for medicinal diagnosis por IRJET Journal
Intelligent data analysis for medicinal diagnosisIntelligent data analysis for medicinal diagnosis
Intelligent data analysis for medicinal diagnosis
IRJET Journal34 vistas
diacon and pacs.pptx por alemu27
diacon and pacs.pptxdiacon and pacs.pptx
diacon and pacs.pptx
alemu2750 vistas

Más de Technological Ecosystems for Enhancing Multiculturality

A Preliminary Study of Proof of Concept Practices and their connection with I... por
A Preliminary Study of Proof of Concept Practices and their connection with I...A Preliminary Study of Proof of Concept Practices and their connection with I...
A Preliminary Study of Proof of Concept Practices and their connection with I...Technological Ecosystems for Enhancing Multiculturality
516 vistas12 diapositivas
Social networks as a promotional space for Spanish radio content. The case st... por
Social networks as a promotional space for Spanish radio content. The case st...Social networks as a promotional space for Spanish radio content. The case st...
Social networks as a promotional space for Spanish radio content. The case st...Technological Ecosystems for Enhancing Multiculturality
153 vistas22 diapositivas
Towards the study of sentiment in the public opinion of science in Spanish por
Towards the study of sentiment in the public opinion of science in SpanishTowards the study of sentiment in the public opinion of science in Spanish
Towards the study of sentiment in the public opinion of science in SpanishTechnological Ecosystems for Enhancing Multiculturality
116 vistas25 diapositivas
A Three-Step Data-Mining Analysis of Top-Ranked Higher Education Institutions... por
A Three-Step Data-Mining Analysis of Top-Ranked Higher Education Institutions...A Three-Step Data-Mining Analysis of Top-Ranked Higher Education Institutions...
A Three-Step Data-Mining Analysis of Top-Ranked Higher Education Institutions...Technological Ecosystems for Enhancing Multiculturality
154 vistas15 diapositivas
Specifics of multimedia texts in the context of social networks media aesthetics por
Specifics of multimedia texts in the context of social networks media aestheticsSpecifics of multimedia texts in the context of social networks media aesthetics
Specifics of multimedia texts in the context of social networks media aestheticsTechnological Ecosystems for Enhancing Multiculturality
149 vistas13 diapositivas
Combined Effects of Similarity and Imagined Contact on First-Person Testimoni... por
Combined Effects of Similarity and Imagined Contact on First-Person Testimoni...Combined Effects of Similarity and Imagined Contact on First-Person Testimoni...
Combined Effects of Similarity and Imagined Contact on First-Person Testimoni...Technological Ecosystems for Enhancing Multiculturality
126 vistas14 diapositivas

Más de Technological Ecosystems for Enhancing Multiculturality(20)

Último

INT-244 Topic 6b Confucianism por
INT-244 Topic 6b ConfucianismINT-244 Topic 6b Confucianism
INT-244 Topic 6b ConfucianismS Meyer
49 vistas77 diapositivas
Gross Anatomy of the Liver por
Gross Anatomy of the LiverGross Anatomy of the Liver
Gross Anatomy of the Liverobaje godwin sunday
89 vistas12 diapositivas
UNIDAD 3 6º C.MEDIO.pptx por
UNIDAD 3 6º C.MEDIO.pptxUNIDAD 3 6º C.MEDIO.pptx
UNIDAD 3 6º C.MEDIO.pptxMarcosRodriguezUcedo
150 vistas32 diapositivas
Six Sigma Concept by Sahil Srivastava.pptx por
Six Sigma Concept by Sahil Srivastava.pptxSix Sigma Concept by Sahil Srivastava.pptx
Six Sigma Concept by Sahil Srivastava.pptxSahil Srivastava
51 vistas11 diapositivas
12.5.23 Poverty and Precarity.pptx por
12.5.23 Poverty and Precarity.pptx12.5.23 Poverty and Precarity.pptx
12.5.23 Poverty and Precarity.pptxmary850239
514 vistas30 diapositivas
Nelson_RecordStore.pdf por
Nelson_RecordStore.pdfNelson_RecordStore.pdf
Nelson_RecordStore.pdfBrynNelson5
50 vistas10 diapositivas

Último(20)

INT-244 Topic 6b Confucianism por S Meyer
INT-244 Topic 6b ConfucianismINT-244 Topic 6b Confucianism
INT-244 Topic 6b Confucianism
S Meyer49 vistas
Six Sigma Concept by Sahil Srivastava.pptx por Sahil Srivastava
Six Sigma Concept by Sahil Srivastava.pptxSix Sigma Concept by Sahil Srivastava.pptx
Six Sigma Concept by Sahil Srivastava.pptx
Sahil Srivastava51 vistas
12.5.23 Poverty and Precarity.pptx por mary850239
12.5.23 Poverty and Precarity.pptx12.5.23 Poverty and Precarity.pptx
12.5.23 Poverty and Precarity.pptx
mary850239514 vistas
Nelson_RecordStore.pdf por BrynNelson5
Nelson_RecordStore.pdfNelson_RecordStore.pdf
Nelson_RecordStore.pdf
BrynNelson550 vistas
Pharmaceutical Analysis PPT (BP 102T) por yakshpharmacy009
Pharmaceutical Analysis PPT (BP 102T) Pharmaceutical Analysis PPT (BP 102T)
Pharmaceutical Analysis PPT (BP 102T)
yakshpharmacy009116 vistas
Creative Restart 2023: Leonard Savage - The Permanent Brief: Unearthing unobv... por Taste
Creative Restart 2023: Leonard Savage - The Permanent Brief: Unearthing unobv...Creative Restart 2023: Leonard Savage - The Permanent Brief: Unearthing unobv...
Creative Restart 2023: Leonard Savage - The Permanent Brief: Unearthing unobv...
Taste62 vistas
JQUERY.pdf por ArthyR3
JQUERY.pdfJQUERY.pdf
JQUERY.pdf
ArthyR3107 vistas
Ask The Expert! Nonprofit Website Tools, Tips, and Technology.pdf por TechSoup
 Ask The Expert! Nonprofit Website Tools, Tips, and Technology.pdf Ask The Expert! Nonprofit Website Tools, Tips, and Technology.pdf
Ask The Expert! Nonprofit Website Tools, Tips, and Technology.pdf
TechSoup 62 vistas
Guess Papers ADC 1, Karachi University por Khalid Aziz
Guess Papers ADC 1, Karachi UniversityGuess Papers ADC 1, Karachi University
Guess Papers ADC 1, Karachi University
Khalid Aziz105 vistas
NodeJS and ExpressJS.pdf por ArthyR3
NodeJS and ExpressJS.pdfNodeJS and ExpressJS.pdf
NodeJS and ExpressJS.pdf
ArthyR350 vistas
Monthly Information Session for MV Asterix (November) por Esquimalt MFRC
Monthly Information Session for MV Asterix (November)Monthly Information Session for MV Asterix (November)
Monthly Information Session for MV Asterix (November)
Esquimalt MFRC213 vistas

Artificial Neural Networks applications in Computer Aided Diagnosis. System design and use as an educational tool

  • 1. Artificial Neural Networks applications in Computer Aided Diagnosis. System design and use as an educational tool. Jorge Hernández Rodríguez Education in the Knowledge Society PhD Program Hernández Rodríguez, Jorge; Rodríguez Conde, María José; Cabrero Fraile, Francisco Javier Track 16: Doctoral Consortium
  • 2. 2 INDEX ● Context and motivation that drives the dissertation research ● State of the art ● Hypothesis and problem statement ● Research objective and goals ● Research approaches and methods ● Results to date and their validity. Dissertation status. ● Current and expected contributions Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool
  • 3. 3 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool CONTEXT AND MOTIVATION THAT DRIVES THE DISSERTATION RESEARCH (I) Computer Aided Detection and Diagnosis (CAD) Software and algorithms (image processing, lesion detection and classification algorithms) “Second opinion” for the radiologist CADe CADx Software specialized in lesion and pathological features detection Software specialized in pathology characterization , classification or diagnosis Radiological image modalities Computed Tomography Mammography Conventional Radiology Magnetic Resonance Imaging Ultrasound …
  • 4. 4 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool CONTEXT AND MOTIVATION THAT DRIVES THE DISSERTATION RESEARCH (II) CAD UTILITY ● Increasing number of high technology equipment in hospitals and clinics ● Radiological examinations with a high number of images and rising number of patients scanned (CT, MRI, screening, interventional radiology…) ● Growth of clinical indications for different types of modalities ¡ Huge amount of workload for the radiologist! CAD: very useful tool to ease workload and improve detection and diagnosis and therefore posterior treatment prescription. Reported in scientific and technical literature: ► The utility of these systems has been confirmed in numerous articles ► Reduction of inter-observer variability associated with image interpretation ► Sensitivity and specificity improvement associated with its use ► Improvement in specialists’ Receiver Operating Characteristic curves ► They provide supplementary information for managing a problem ► They have been successfully integrated into Picture Archiving and Communication Systems and Radiology Information Systems OBJECTIVE ■ Assisting radiologists in detection and diagnosis ■ Great potential for medical specialists’ training and as an educational tool
  • 5. 5 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool STATE OF THE ART (I) Artificial Intelligence Artificial Neural Networks (ANNs)
  • 6. 6 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool STATE OF THE ART (II) Evolution of ANNs in Medical Imaging ↓ Convolutional Neural Networks (CNNs) Neural Networks based in Deep Learning Methods ↓ Highly non-linear systems Designed to work directly with images Deeper architectures (higher number of layers) Reduction in the amount of adjustable network parameters Limited size of training and validation datasets ↓ Calculation of empirical features of segmented lesions in images (manually or automatically) Use of a classification algorithm (for example, linear discriminant analysis, support vector machines, artificial neural networks “Traditional CAD schemes”
  • 7. 7 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool STATE OF THE ART (III): Mammography Arévalo et al. (2016). Computers and Methods in Biomedicine CNNs based in Deep learning techniques Mammography database: BCDR Preprocessing • Image cropping • Data augmentation • Global Contrast Normalization • Local Contrast Normalization Sample of marked lesion Supervised learning through CNN and classification Architecture of the CNN that performed better Evaluation of results Comparison with other state-of-the art representations for lesion classification and image analysis
  • 8. 8 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool STATE OF THE ART (IV): Pulmonary node classification Cheng et al. (2016). Scientific reports - Nature Deep learning techniques SDAE classifier achieved better results Patterns from classifier’s hidden layers Examples of rated nodules Shen et al. (2015). Information processing in medical imaging Multi-scale CNNs Classification accuracy of 86,4% on LIDC-IDRI database
  • 9. Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool STATE OF THE ART (V): Computer-Aided Detection Roth et.al. (2016). IEEE Transactions on Medical Imaging Convolutional Neural Networks ● Hierarchical two-tiered CADe system ● New approach: 2,5 D image decomposition ● Wide range of applications: Sclerotic spine metastases ThoracoAbdominal Lymph nodes Colonyc polyps ● Different candidate generation algorithms ● False positive reduction schemes Detection of sclerotic metastases Detection of lymph nodes Influence of Random View number on performance Performance on training and testing datasets of different approaches. Lymph node detection.
  • 10. Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool STATE OF THE ART (VI): CAD applications in medical education Mazurowski et al. (2010). Medical Physics Zhang et al. (2014). Medical Physics Ping et al. (2005). Academic Radiology Hypermedia instructional program in CAD-aided mammography training Individually adapted computer-aided educational system in mammography Construction of user models. Probability of making diagnostic errors by radiology residents (difficulty of cases). ↓ Relation with image features Computer models to predict locations of false-positives in mammography Based on previous annotations
  • 11. 11 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool HYPOTHESIS AND PROBLEM STATEMENT (I) FIRST STAGE: CAD SYSTEM DEVELOPMENT Creation of a CAD system for radiological image analysis Analysis of system’s performance and efficiency in pathology detection and classification. HYPOTHESES ► The system developed is a useful tool for assisting radiologists as a “second image reader”. ► System’s sensitivity and specificity is comparable to other published results. ► The number of false positives detected is kept in reasonable numbers, similar to published values. ► Detection and classification accuracy on different image datasets reaches reported values. ► Software is capable of extracting useful image features and information for the specialist.
  • 12. 12 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool SECOND STAGE: SYSTEM VALIDATION Evaluation of system’s effectiveness as a clinical, educational and training tool for medicine students and future medical specialists. HYPOTHESES ► The CAD system designed and trained with validated image datasets produces positive effects in the analysis and diagnosis of radiological images. ► The CAD scheme developed together with its functionalities will serve as a useful and complementary educational tool for medicine students learning to interpret medical images and for specialists’ training in the field of diagnostic radiology. HYPOTHESIS AND PROBLEM STATEMENT (II)
  • 13. 13 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool RESEARCH OBJECTIVES AND GOALS (I) MAIN OBJECTIVE Development and validation of an ANN based CAD scheme Extensible to different image modalities and pathologies (pulmonary nodes in CT, polyps in CT colonography, mammography…) With tools to access and classify the CAD generated information and educational functionalities
  • 14. 14 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool RESEARCH OBJECTIVES AND GOALS (II) SPECIFIC OBJECTIVES 1. Revision of CAD schemes, models and algorithms. 2. CAD scheme design. 3. Download validated image databases and annotations. 4. Project development from neural network training. 5. Integration of ANNs in a CAD platform. 6. System validation prior to clinical use. 7. System validation for educational purposes and specialists training.
  • 15. 15 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool STAGE 1: Development and validation of an ANN based CAD scheme CAD system’s training, adjustment and validation Public medical image databases Available for educational, research and software development purposes They contain clinical and diagnostic information related with the cases and pathologies considered Information validated by groups of experts and with different methods and techniques Image repositories for different modalities and pathologies RESEARCH APPROACHES AND METHODS (I)
  • 16. 16 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool MAMMOGRAPHY Medical image databases THORAX RADIOGRAPHY CANCER IMAGE DATASETS (DIFFERENT IMAGE MODALITIES) RESEARCH APPROACHES AND METHODS (II)
  • 17. 17 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool Calculation environment for working with medical images and neural networks Convolutional Neural Networks Toolbox for artificial vision applications Image Processing Toolbox Computer vision algorithm library (open source code) RESEARCH APPROACHES AND METHODS (III) Software and calculations
  • 18. 18 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool General purpose programming language Package of functions dedicated for scientific computing with PythonTM Python library for defining, optimizing and evaluating multidimensional array expressions (suited for ANNs) Python software development environment Possibility of performing GPU accelerated calculations with these environments RESEARCH APPROACHES AND METHODS (IV)
  • 19. 19 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool RESEARCH APPROACHES AND METHODS (V) Validation of ANNs performance ● Split of datasets into three groups: training, validation and testing ● Monitoring network parameter’s during iterative processes ● Calculation of parameters to quantify network’s performance ● Statistical analysis of the results produced in the testing phase ● Cross-validation combining different ANNs in an ensemble - Increase the testing samples to ensure optimal tuning (if needed)
  • 20. 20 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool RESEARCH APPROACHES AND METHODS (VI) STAGE 2: System validation for clinical use and utility as an educational tool ● Use in clinical practice by specialists ● Training in software operation ● Analysis of cases with software and result interpretation ● Study of CAD influence on reader’s performance ● Statistical analysis of a sample of cases ● Analysis of clinical cases by radiology residents and medicine students ● Training in software operation and educational tools ● Surveys to evaluate goals during training period ● Study of CAD influence on learning process ● Statistical analysis of surveys’ results
  • 21. 21 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool RESULTS TO DATE AND THEIR VALIDITY. DISSERTATION STATUS ● Bibliographic review in the areas of CAD and ANNs. ● Installation of software environments and function libraries. ● Download of image datasets and annotations. ● Formatting of images and preprocessing steps. Classification of cases. ● Initial design of CAD application. Initial routines and scripts. ● Selection of network architecture and layer types.
  • 22. 22 Track 16: Artificial Neural Networks application in Computer Aided Diagnosis. System design and use as an educational tool CURRENT AND EXPECTED CONTRIBUTIONS ● ANN training from parameter value optimization based on annotations ● Comprehensive analysis of ANNs performance. Analysis of the influence of several architecture parameters, learning algorithms, sample sets size and type, preprocessing steps… on it. ● Implementation of tools to allow learning through CAD use. Design of specific validation tests. ● Validation of the CAD scheme in clinical practice: interaction with the use and clinical data generated ● Educational validation derived from the use of the system by medicine students and radiology residents. ● Optimization of the system from information extracted from practical use ● Future contributions published in the next TEEM conference and journals in the field of medical imaging, computers in medicine or radiology and medical education
  • 23. Thank you very much for your attention!

Notas del editor

  1. Good morning. My name is Jorge Hernández, and I am student at the Knowledge Society Doctoral Program. I am here to present my dissertation work, which started one year ago. It’s title is Artificial Neural Networks applications in Computer Aided Diagnosis. System design and use as an educational tool.
  2. This talk will be distributed as you can see in this slide. I will start with the Context and Motivation, following with a brief state of the art presentation. Afterwards, I will expose the problem statement and the Hypotheses that will be tested, together with the research objectives and investigation methodologies. To conclude, the dissertation status and the current and expected contributions derived from this work will be commented.
  3. Computer Aided Diagnosis and Detection, from now on CAD, is defined as the computer algorithms and software designed to help radiologists in lesion detection and classification. Thay can be categorized in two main groups: CADe, which is specialized in detection of lesions and pathological features, and CADx, software created for pathology characterizatio, classification or diagnosis. Their applications extend to a wide range of image modalities, as for example, computed tomography, mammography, magentic resonance imaging, and conventional and interventional radiology.