指導老師 : 李麗華 教授 報告者 : 廖偉丞. Catalog  Author  Abstract  Keywords  Introduction  Artificial neural network ensembles  Lung cancer diagnosis system 

Slides:



Advertisements
Similar presentations
Chapter 4 Pattern Recognition Concepts: Introduction & ROC Analysis.
Advertisements

-Artificial Neural Network- Chapter 2 Basic Model
Rapid Object Detection using a Boosted Cascade of Simple Features Paul Viola, Michael Jones Conference on Computer Vision and Pattern Recognition 2001.
Rapid Object Detection using a Boosted Cascade of Simple Features Paul Viola, Michael Jones Conference on Computer Vision and Pattern Recognition 2001.
Probabilistic Classification using Fuzzy Support Vector Machines (PFSVM) Marzieh Parandehgheibi ORC - MIT INFORMS DM-HI11/12/20111.
Face detection Many slides adapted from P. Viola.
EBUS-TBNA reduces the number of mediastinoscopies for the staging of lung cancer with more than fifty percent Background Methods Results Conclusion Niels.
1 Learning to Detect Objects in Images via a Sparse, Part-Based Representation S. Agarwal, A. Awan and D. Roth IEEE Transactions on Pattern Analysis and.
Dynamic Face Recognition Committee Machine Presented by Sunny Tang.
Lazy Learning k-Nearest Neighbour Motivation: availability of large amounts of processing power improves our ability to tune k-NN classifiers.
An Approach to Korean License Plate Recognition Based on Vertical Edge Matching Mei Yu and Yong Deak Kim Ajou University Suwon, , Korea 指導教授 張元翔.
-Artificial Neural Network- Adaptive Resonance Theory(ART) 朝陽科技大學 資訊管理系 李麗華 教授.
A Neural Network-Based Approach for Statistical Probability Distribution Recognition 指導教授:陳茂生 教授 報告學生:周家任 d IEEM 7103 Topics in Operations Research.
1 Diagnosing Breast Cancer with Ensemble Strategies for a Medical Diagnostic Decision Support System David West East Carolina University Paul Mangiameli.
Diagnosis of Ovarian Cancer Based on Mass Spectrum of Blood Samples Committee: Eugene Fink Lihua Li Dmitry B. Goldgof Hong Tang.
A Computer Aided Detection System For Mammograms Based on Asymmetry and Feature Extraction Techniques By Mohammed Jirari Benidorm, Spain Sept 9th, 2005.
Rotation Forest: A New Classifier Ensemble Method 交通大學 電子所 蕭晴駿 Juan J. Rodríguez and Ludmila I. Kuncheva.
Oral Defense by Sunny Tang 15 Aug 2003
-Artificial Neural Network- Chapter 3 Perceptron 朝陽科技大學 資訊管理系 李麗華教授.
Digital Pathology Diagnostic Accuracy, Viewing Behavior and Image Characterization Linda Shapiro University of Washington Computer Science and Engineering.
Foundations of Computer Vision Rapid object / face detection using a Boosted Cascade of Simple features Presented by Christos Stoilas Rapid object / face.
JAVED KHAN ET AL. NATURE MEDICINE – Volume 7 – Number 6 – JUNE 2001
FACE DETECTION AND RECOGNITION By: Paranjith Singh Lohiya Ravi Babu Lavu.
Selective Sampling on Probabilistic Labels Peng Peng, Raymond Chi-Wing Wong CSE, HKUST 1.
Ranga Rodrigo April 5, 2014 Most of the sides are from the Matlab tutorial. 1.
1 LUNG. 2 Equivalent Terms, Def, Charts, Tables, Illustrations.
Early methods of diagnostic in oral medicine
Bayesian Network for Predicting Invasive and In-situ Breast Cancer using Mammographic Findings Jagpreet Chhatwal1 O. Alagoz1, E.S. Burnside1, H. Nassif1,
Classification of multiple cancer types by multicategory support vector machines using gene expression data.
-Artificial Neural Network- Hopfield Neural Network(HNN) 朝陽科技大學 資訊管理系 李麗華 教授.
WELCOME. Malay Mitra Lecturer in Computer Science & Application Jalpaiguri Polytechnic West Bengal.
ARTIFICIAL NEURAL NETWORKS. Overview EdGeneral concepts Areej:Learning and Training Wesley:Limitations and optimization of ANNs Cora:Applications and.
BOOSTING David Kauchak CS451 – Fall Admin Final project.
1 Learning Chapter 18 and Parts of Chapter 20 AI systems are complex and may have many parameters. It is impractical and often impossible to encode all.
Multiple parallel hidden layers and other improvements to recurrent neural network language modeling ICASSP 2013 Diamantino Caseiro, Andrej Ljolje AT&T.
指導老師 : 蔡亮宙 報告者 : 黃柏愷 A new method of vehicle license plate location under complex scenes.
BAGGING ALGORITHM, ONLINE BOOSTING AND VISION Se – Hoon Park.
A Face processing system Based on Committee Machine: The Approach and Experimental Results Presented by: Harvest Jang 29 Jan 2003.
Cytologic and DNA- Cytometric Early Diagnosis of Oral Cancer Torsten W. Remmerbach, Horst Weidenbach, Natalja Pomjanski, Kristiane Knops, Stefanie Mathes,
Evolutionary Algorithms for Finding Optimal Gene Sets in Micro array Prediction. J. M. Deutsch Presented by: Shruti Sharma.
A.N.N.C.R.I.P.S The Artificial Neural Networks for Cancer Research in Prediction & Survival A CSI – VESIT PRESENTATION Presented By Karan Kamdar Amit.
The Viola/Jones Face Detector A “paradigmatic” method for real-time object detection Training is slow, but detection is very fast Key ideas Integral images.
Measures of Teacher Stages of Technology Integration and Their Correlates with Student Achievement Christensen, R., Griffin, D., & Knezek, G. (2001,March).
9.913 Pattern Recognition for Vision Class9 - Object Detection and Recognition Bernd Heisele.
An Effective Hybridized Classifier for Breast Cancer Diagnosis DISHANT MITTAL, DEV GAURAV & SANJIBAN SEKHAR ROY VIT University, India.
TEMPLATE DESIGN © Classification of Magnetic Resonance Brain Images Using Feature Extraction and Adaptive Neuro-Fuzzy.
Cell Segmentation in Microscopy Imagery Using a Bag of Local Bayesian Classifiers Zhaozheng Yin RI/CMU, Fall 2009.
THIRD CLASSIFICATION OF MICROCALCIFICATION STAGES IN MAMMOGRAPHIC IMAGES THIRD REVIEW Supervisor: Mrs.P.Valarmathi HOD/CSE Project Members: M.HamsaPriya( )
Face detection Many slides adapted from P. Viola.
國立雲林科技大學 National Yunlin University of Science and Technology Intelligent Database Systems Lab 1 Self-organizing map for cluster analysis of a breast cancer.
A New Generation of Artificial Neural Networks.  Support Vector Machines (SVM) appeared in the early nineties in the COLT92 ACM Conference.  SVM have.
Automatic Classification for Pathological Prostate Images Based on Fractal Analysis Source: IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 28, NO. 7, JULY.
Brain imaging prior to lung cancer resection
Big data classification using neural network
Lung cancer cell identification based on artificial neural network ensembles 指導老師: 李麗華 教授 報告者: 廖偉丞.
報告者:聶家駿 指導老師:陳卉瑄老師 指導助教:葉庭禎 陳耀傑
Robust Lung Nodule Classification using 2
A New Classification Mechanism for Retinal Images
RECURRENT NEURAL NETWORKS FOR VOICE ACTIVITY DETECTION
CS 698 | Current Topics in Data Science
By: Behrouz Rostami, Zeyun Yu Electrical Engineering Department
THALASSEMIA MINOR DIAGNOSTICS BY A COMPUTATIONAL METHOD
Somi Jacob and Christian Bach
Volume 150, Issue 1, Pages (July 2016)
Ensembles An ensemble is a set of classifiers whose combined results give the final decision. test feature vector classifier 1 classifier 2 classifier.
Evolutionary Ensembles with Negative Correlation Learning
Introduction Face detection and alignment are essential to many applications such as face recognition, facial expression recognition, age identification,
Advisor: Dr.vahidipour Zahra salimian Shaghayegh jalali Dec 2017
Tschandl P1,2, Argenziano G3, Razmara M4, Yap J4
Presented By: Firas Gerges (fg92)
Presentation transcript:

指導老師 : 李麗華 教授 報告者 : 廖偉丞

Catalog  Author  Abstract  Keywords  Introduction  Artificial neural network ensembles  Lung cancer diagnosis system  Lung cancer cell identification  Results of single artificial neural networks  Results of two kinds of ensemble  Neural ensemble based detection  Conclusion

Author(1/1)  Nanjing University  Zhi-Hua Zhou 、 Yuan Jiang 、 Yu-Bin Yang 、 Shi-Fu Chen

Abstract(1/3)  Neural Ensemble-based Detection (NED) is automatic pathological diagnosis procedure  Built on a two-level ensemble architecture

Abstract(2/3)  First-level output Normal and Cancer cell  Second-level output adenocarcinoma, squamous cell carcinoma small cell carcinoma large cell carcinoma norma

Abstract(3/3)  Use NED identification rate↑ Type I error↓  reducing missing diagnoses and will save the lives of cancer patients

Keywords(1/1)  Artificial neural networks  Pattern recognition  Image processing  Computer-aided medical diagnosis  Expert system

Introduction(1/1)  Lung cancer is one of the most common and deadly diseases in the world.  Detection of lung cancer in its early stage is the key of its cure  senior pathologists are rare

Artificial neural network ensembles(1/2)  Two kinds of methods Individual artificial neural networks Combining individual predictions

Artificial neural network ensembles(2/2) Simple averaging and weighted averaging Output Plurality voting Final output

Lung cancer diagnosis system(1/4)  An early stage lung cancer diagnosis system named LCDS  LCDS could not be clearly diagnosed by X-ray chest films  The proportion of positives, i.e. lung cancer patients, is about 70–80%

Lung cancer diagnosis system(2/4)  LCDS system is depicted in Fig

Lung cancer diagnosis system(3/4)

Lung cancer diagnosis system(4/4)  morphologic features include the perimeter, area, roundness, and rectangleness.  colorimetric features include the red component, green component, blue component, illumination, saturatio.

Lung cancer cell identification(1/1)  552 cell images(75% belong to cancer cells)  Five subsets with similar size  Union of four subsets as training set to train.

Results of single artificial neural networks(1/2)  FANNC Learning↑ High accuracy↑

Results of single artificial neural networks(2/2)  Records the average value of those five experiments. Exp1Exp2Exp3Exp4Exp5Ave. Err (%) Err fn (%) Err fp (%)

Results of two kinds of ensemble(1/4)  The first kind of ensemble approach

Results of two kinds of ensemble(2/4)  Records the average value of those five experiments. identification↑ Exp1Exp2Exp3Exp4Exp5Ave. Err (%) Err fn (%) Err fp (%)

Results of two kinds of ensemble(3/4)  The second kind of ensemble approach  two artificial neural network ensembles were trained, among which one ensemble was trained biased to benign diagnoses by letting negative examples dominating the training sets  other ensemble was trained biased to malign diagnoses by letting positive examples dominating the training sets.

Results of two kinds of ensemble(4/4)  Records the average value of those five experiments. Exp1Exp2Exp3Exp4Exp5Ave. Err (%) Err fn (%) Err fp (%)

Neural ensemble based detection(1/3)  NED employs a specific two-level ensemble architecture.  First-level judged cancer cells  Second-level Responsible to report the type of the cells.

Neural ensemble based detection(2/3)  Experimental results of NED Exp1Exp2Exp3Exp4Exp5Ave. Err (%) Err fn (%) Err fp (%)

Neural ensemble based detection(3/3)  the flowchart of NED

Conclusion(1/1)  Through adopting those techniques, NED achieves not only high rate of overall identification, but also low rate of false negative identification.

END