Supervised Segmentation of Visible Human Data with Image Analogies

Size: px
Start display at page:

Download "Supervised Segmentation of Visible Human Data with Image Analogies"

Transcription

1 Supervised Segmentation of Visible Human Data with Analogies James B. Lackey and Michael D. Colagrosso Department of Mathematical and Computer Sciences Colorado School of Mines Golden, CO Phone: Fax: Abstract We present a new application of the Analogies algorithm to be used for image segmentation. Our approach requires supervised training data, so we apply it to the domain of labeling human anatomical data. In the Visible Human Project, expert anatomists are overwhelmed with high-resolution images to analyze. We propose that the anatomist can work in conjunction with our approach, letting the machine segment 80% of the images, and requiring that the expert segment only every fifth image. Keywords: image analogies, image segmentation, visible human I. INTRODUCTION We consider image segmentation in a domain that facilitates supervised learning. Unlike unsupervised learning methods, such as spectral clustering [2], we are trying to directly emulate the segmentation of an expert anatomist. In the medical industry, far more data on the human body is being collected than can be properly analyzed. There is no reliable way to quickly segment large volumes of data, and even when this data is segmented, it may stack to form a poor 3D model. If we can partially automate segmentation, then medical data can be analyzed faster and at smaller cost. Furthermore, we can impose constraints on the segmentation process that will lead to the creation better 3D models of human structures. II. THE VISIBLE HUMAN PROJECT The Visible Human Project [3] has collected large amounts of data to be segmented. The anatomical images of the Visible Human Male were created with vertical and planar resolution of 1 mm, yielding approximately 1800 slices to be segmented. Each image contains dozens of structures which need to be segmented. Some, such as the colon or appendix are relatively large. But, others, such as cross sections of arteries or even blood vessels are much smaller. Anatomists segment slices one structure at a time, so segmentation of all the organs in a set of slices requires several passes. Currently, a human segments the images by hand with the aid of Photoshop-like tools. This is very time consuming. One goal of the visible human project is to build instructive 3D models of human organs, such as the kidneys, heart, and lungs. Taking the kidney as an example, the model is formed by stacking all the slices that contain the kidney and combining all the points in the slices that are members of the kidney segmentation. One problem with human segmentations they often do not form smooth 3D models. A method which imposes a curvature restriction on the structures within the model would perform much better. Segmentation Fig. 1. Sample visible human image and its human segmentation. Solid colors in the segmentation represent bone, muscle, fat, etc. III. PREVIOUS SEGMENTATION METHODS In the past, deformative models have had the most success automatically segmenting data. [4] These start with a human segmented image and deform the boundaries as proximate slices are presented. Deformative models perform well for contingous structures. These models attempt to identify boundaries and then segment on the basis of these boundaries. They typically do not incorporate knowledge about textures or textons. However, these deformative models are limited. They can only account for changes in shape of structures; they cannot correctly segment structures that were not present in the original slice. IV. IMAGE ANALOGIES Analogies (IA) [1] gives an algorithm to solve the problem A : A :: B : B where B is unknown. For English words, A relates to A in the same way B relates to B. Analogies (IA) can be used to map the texture of one image

2 onto another image, synthesize analogous texture, or compute texture by numbers. The synthesis of texture by numbers is most pertinent to our work. In texture by numbers, IA is given segmented images A and B, and an unsegmented image A. IA works by constructing Gaussian pyramid representations 1 4 resolution representations of A, A and B. For each pixel in B, a corresponding region in A is selected using the Gaussian pyramids. For that pixel, B is assigned the same value as A. This 1 4 resolution image is then sampled to produce a 1 2 resolution Gaussian pyramid. The remainder of the pyramid is computed. Finally, a full resolution pyramid is constructed using the 1 2 resolution pyramid and additional computation. V. APPLICATION Up to now, IA has been applied to do texture by numbers. That is, given a segmented image, IA can compute the unsegmented image. We want to do the opposite, compute a segmented image given an unsegmented image. That is, we want to use IA as a texture recognizer. We have a set of consecutive Visible Human slices: {m 0, m 1,..., m n } and we have a human segmented image s 0. The most naive application of IA would be to compute the analogies: m 0 : s 0 :: m 1 : s 1 m 0 : s 0 :: m 2 : s 2... m 0 : s 0 :: m n : s n However, this gives poor results. As the image being segmented gets farther and farther from m 0, accuracy drops off. Performance can be improved significantly by progressively computing analogies. That is, we compute m 0 : s 0 :: m 1 : s 1 m 1 : s 1 :: m 2 : s 2... m n 1 : s n 1 :: m n : s n In the results below, we use the anatomist segmentations as our basis for estimating performance in terms of accuracy, precision, and recall. Additionally, we desire a method that will provide segmentations that provide smoothness and therefore yield useful 3D models. Discrepancies between layers generated by IA will typically be small, so the model will be smooth. VI. RESULTS We present three sets of results. First, the results of IA segmented all structes in Visible Human slices. Second, we present segmentations of muscle tissue, and finally segmenations of the colon. A. Full Segmentation The naive method of segmentation consistently performs better than the progressive method on accuracy, precision and recall. However, these results are misleading. The progressively segmented images are more smooth within the slice and form a more smooth 3D model when stacked. For this reason, we need a different measure of segmentation performance. The original goal is to build smooth 3D models. Accuracy, precision and recall say nothing about smoothness, so we must devise our own smoothness metric. Say we want a normalized measure of smoothness. We define the smoothness of a pixel at coordinates (x, y, z) to be 1 if the pixels (x, y, z), (x, y, z 1) and (x, y, z + 1) have 1 the same value, 2 if (x, y, z) or (x, y, z 1) has the same value as (x, y, z) and 0 otherwise. We can then take the average smoothness over a segmentation by taking the average of smoothness across all pixels in each slice of the segmentation. TABLE I AVERAGE MEASURES OF ACCURACY ±SD FOR 9 SIMULATIONS Naive Progressive 0.450± ± ± ± ± ± ± ± The only significant errors in the segmentation are in images 2,3 and 4 of Figure 2. The mislabeled region is the psoas major right that is labeled as psoas major left. The psoas major left and right are muscle tissue on either side of the spine. They are symetric. Our experts segmented psoas major left, erector spinae left and quadratus lumborum left as one texture, but decided their symmetric counterparts would be a different texture. Clearly this choice is somewhat arbitrary, as there is no functional difference in the textures representing the muscle tissue. For this reason, it is understandable, and perhaps even acceptable, that IA was unable to distinguish between these two textures. B. Muscle Tissue Segmentation The full segmentation images in the previous section have some structures labeled in a way that is not conducive to learning. For example, in Figure 2, there are several muscle textures in the image, but they are labeled differently in the expert segmentation depending on whether they are on the left or right side. The IA algorithms cannot learn this anatomical distinction; the Progressive IA algorithm, for example, labels the left muscle texture the same as the right in s 2, 3, and 4 in Figure 2, resulting in large errors. Based on this result, we focus on correctly labeling one single texture, which is a similar to the anatomist s method of labeling one structure and making several passes. Once we

3 Expert Naive IA Progressive IA Prog. IA Error Fig. 2. Results of Full Segmentation. The Progressive IA approach labels most of the textures correctly, but labels the left muscle texture the same as the right, resulting in the large black regions in the Error column. have labeled all the muscle textures in an image, an anatomist can further subdivide the them based on what muscle groups they belong to. In Figure 3, we created the expert image by labeling any type of muscle white, and everything else black. As expected, the resulting errors are much smaller in this limited task. C. Colon Segmentation The labels of muscle texture in Figure 3 are especially sensitive to the small amounts of fat that are intermingled. In fact, most of the errors in the Progressive IA approach are in the interior of the large muscle group, which would be tedious for an anatomist to correct. Instead, we desire a texture that our method can label accurately and require the least postprocessing by the anatomist, even if it is on a more-limited scope than the full segmentation. To this end, we applied our method to the colon structure in the same manner as the muscle above. The results in Figure 4 show an accurate segmentation, and the errors are on the boundaries of the region, not the interior. TABLE II AVERAGE M EASURES OF ACCURACY FOR 5 NAIVE C OLON S EGMENTATION ±SD Accuracy 0.986± ± ± ±0.035 Precision 0.955± ± ± ±0.149 Recall 0.956± ± ± ±0.017 VII. C ONCLUSIONS AND F UTURE W ORK By progressively segmenting, we can produce reasonably accurate segmentation four slices out. In short, we can reduce the anatomist s work by 80%. One limitation of the human and machine learning segmentations is that a pixel can be a part of only one segment. Thin structures are sensitive the consequences, and discontinuities can arise. For the human segmentation, this is a consequence of the tools used to decompose the image, but the machine learning approach has no such limitation.

4 Expert Naive IA Progressive IA Prog. IA Error Fig. 3. Results of Muscle Tissue Segmentation. By labeling all the muscle textures white and everything else black, both approaches perform better, but still make several one-pixel errors that would require review by an anatomist. TABLE III AVERAGE M EASURES OF ACCURACY FOR 5 P ROGRESSIVE C OLON S EGMENTATION ±SD Accuracy 0.990± ± ± ± Precision 0.970± ± ± ±0.008 Recall 0.967± ± ± ±0.002 In the future, we will generalize IA to higher dimensionality. The obvious way to do this is to compute a Gaussian hyperpyramid over some set of image {m0, m1,..., mn } for which we have segmented images {s0, s1,..., sn }. After computing the pyramid we could compute matches for the images {sn+1, sn+2,..., sm } One advantage of this methodology is that it would impose smoothness on the image. Another is that it incorporates information about consecutive layers. For instance, in the 2D case, the machine might know that the backbone occurs next to other pixels of backbone, but it would not know that long sections of backbone occur vertical. The Gaussian hyperpyramid would incorporate this knowledge. We believe we can further improve performance by using a Canny Filter as a post-processor. Given our machine segmented images {sn, sn+1,..., sm } and a set of closed regions defined by the Canny filter, we can define a new segmentation with the boundaries from the Canny filter. The regions defined by those boundaries will be segmented according to whatever texture was dominant within the segmentation {sn, sn+1,..., sm }. This gives a new segmentation. ACKNOWLEDGMENTS Thanks to Elizabeth Prince from the CU Health Sciences Center for insight into the Visible Human data and thoughtful explanations. R EFERENCES [1] Aaron Hertzmann, Charles E. Jacobs, Nuria Oliver, Brian Curless, David H. Salesin. Analogies. Proc. SIGGRAPH, 2001.

5 Expert Naive IA Progressive IA Prog. IA Error Fig. 4. Results of Colon Segmentation. In this simplified texture, the Progressive IA approach makes errors only on the boundaries of the texture. [2] A. Ng, M. Jordan and Y. Weiss. On spectral clustering: Analysis and an algorithm. Advances in Neural Information Processing Systems 14, [3] human.html [4] A. Singh, L. von Kurowski, and M. Y. Chiu. Cardiac MR Segmentation Using Deformable Models. Proc. SPIE Medical Imaging, New Port Beach, 1993.

Facebook Friend Suggestion Eytan Daniyalzade and Tim Lipus

Facebook Friend Suggestion Eytan Daniyalzade and Tim Lipus Facebook Friend Suggestion Eytan Daniyalzade and Tim Lipus 1. Introduction Facebook is a social networking website with an open platform that enables developers to extract and utilize user information

More information

In this presentation, you will be introduced to data mining and the relationship with meaningful use.

In this presentation, you will be introduced to data mining and the relationship with meaningful use. In this presentation, you will be introduced to data mining and the relationship with meaningful use. Data mining refers to the art and science of intelligent data analysis. It is the application of machine

More information

OUTLIER ANALYSIS. Data Mining 1

OUTLIER ANALYSIS. Data Mining 1 OUTLIER ANALYSIS Data Mining 1 What Are Outliers? Outlier: A data object that deviates significantly from the normal objects as if it were generated by a different mechanism Ex.: Unusual credit card purchase,

More information

Environmental Remote Sensing GEOG 2021

Environmental Remote Sensing GEOG 2021 Environmental Remote Sensing GEOG 2021 Lecture 4 Image classification 2 Purpose categorising data data abstraction / simplification data interpretation mapping for land cover mapping use land cover class

More information

Colour Image Segmentation Technique for Screen Printing

Colour Image Segmentation Technique for Screen Printing 60 R.U. Hewage and D.U.J. Sonnadara Department of Physics, University of Colombo, Sri Lanka ABSTRACT Screen-printing is an industry with a large number of applications ranging from printing mobile phone

More information

This puzzle is based on the following anecdote concerning a Hungarian sociologist and his observations of circles of friends among children.

This puzzle is based on the following anecdote concerning a Hungarian sociologist and his observations of circles of friends among children. 0.1 Friend Trends This puzzle is based on the following anecdote concerning a Hungarian sociologist and his observations of circles of friends among children. In the 1950s, a Hungarian sociologist S. Szalai

More information

Big Data: Rethinking Text Visualization

Big Data: Rethinking Text Visualization Big Data: Rethinking Text Visualization Dr. Anton Heijs anton.heijs@treparel.com Treparel April 8, 2013 Abstract In this white paper we discuss text visualization approaches and how these are important

More information

RUN-LENGTH ENCODING FOR VOLUMETRIC TEXTURE

RUN-LENGTH ENCODING FOR VOLUMETRIC TEXTURE RUN-LENGTH ENCODING FOR VOLUMETRIC TEXTURE Dong-Hui Xu, Arati S. Kurani, Jacob D. Furst, Daniela S. Raicu Intelligent Multimedia Processing Laboratory, School of Computer Science, Telecommunications, and

More information

Constrained Tetrahedral Mesh Generation of Human Organs on Segmented Volume *

Constrained Tetrahedral Mesh Generation of Human Organs on Segmented Volume * Constrained Tetrahedral Mesh Generation of Human Organs on Segmented Volume * Xiaosong Yang 1, Pheng Ann Heng 2, Zesheng Tang 3 1 Department of Computer Science and Technology, Tsinghua University, Beijing

More information

Modelling, Extraction and Description of Intrinsic Cues of High Resolution Satellite Images: Independent Component Analysis based approaches

Modelling, Extraction and Description of Intrinsic Cues of High Resolution Satellite Images: Independent Component Analysis based approaches Modelling, Extraction and Description of Intrinsic Cues of High Resolution Satellite Images: Independent Component Analysis based approaches PhD Thesis by Payam Birjandi Director: Prof. Mihai Datcu Problematic

More information

BIOL 1108 Vertebrate Anatomy Lab

BIOL 1108 Vertebrate Anatomy Lab BIOL 1108 Vertebrate Anatomy Lab This lab explores major organs associated with the circulatory, excretory, and nervous systems of mammals. Circulatory System Vertebrates are among the organisms that have

More information

QUALITY TESTING OF WATER PUMP PULLEY USING IMAGE PROCESSING

QUALITY TESTING OF WATER PUMP PULLEY USING IMAGE PROCESSING QUALITY TESTING OF WATER PUMP PULLEY USING IMAGE PROCESSING MRS. A H. TIRMARE 1, MS.R.N.KULKARNI 2, MR. A R. BHOSALE 3 MR. C.S. MORE 4 MR.A.G.NIMBALKAR 5 1, 2 Assistant professor Bharati Vidyapeeth s college

More information

Our Human Body On-site student activities Years 5 6

Our Human Body On-site student activities Years 5 6 Our Human Body On-site student activities Years 5 6 Our Human Body On-site student activities: Years 5-6 Student activity (and record) sheets have been developed with alternative themes for students to

More information

Doppler. Doppler. Doppler shift. Doppler Frequency. Doppler shift. Doppler shift. Chapter 19

Doppler. Doppler. Doppler shift. Doppler Frequency. Doppler shift. Doppler shift. Chapter 19 Doppler Doppler Chapter 19 A moving train with a trumpet player holding the same tone for a very long time travels from your left to your right. The tone changes relative the motion of you (receiver) and

More information

Chapter 6. The stacking ensemble approach

Chapter 6. The stacking ensemble approach 82 This chapter proposes the stacking ensemble approach for combining different data mining classifiers to get better performance. Other combination techniques like voting, bagging etc are also described

More information

COMPARISON OF OBJECT BASED AND PIXEL BASED CLASSIFICATION OF HIGH RESOLUTION SATELLITE IMAGES USING ARTIFICIAL NEURAL NETWORKS

COMPARISON OF OBJECT BASED AND PIXEL BASED CLASSIFICATION OF HIGH RESOLUTION SATELLITE IMAGES USING ARTIFICIAL NEURAL NETWORKS COMPARISON OF OBJECT BASED AND PIXEL BASED CLASSIFICATION OF HIGH RESOLUTION SATELLITE IMAGES USING ARTIFICIAL NEURAL NETWORKS B.K. Mohan and S. N. Ladha Centre for Studies in Resources Engineering IIT

More information

2.2 Creaseness operator

2.2 Creaseness operator 2.2. Creaseness operator 31 2.2 Creaseness operator Antonio López, a member of our group, has studied for his PhD dissertation the differential operators described in this section [72]. He has compared

More information

Decimal Notations for Fractions Number and Operations Fractions /4.NF

Decimal Notations for Fractions Number and Operations Fractions /4.NF Decimal Notations for Fractions Number and Operations Fractions /4.NF Domain: Cluster: Standard: 4.NF Number and Operations Fractions Understand decimal notation for fractions, and compare decimal fractions.

More information

Development of Simulation Tools Software

Development of Simulation Tools Software Development of Simulation Tools Software Vincent Luboz Department of Biosurgery and Surgical Technology Imperial College London BSc VR Surgical Simulation Software Slide 1 Contents Virtual Reality VR Surgical

More information

LABORATORY TWO GEOLOGIC STRUCTURES

LABORATORY TWO GEOLOGIC STRUCTURES EARTH AND ENVIRONMENT THROUGH TIME LABORATORY- EES 1005 LABORATORY TWO GEOLOGIC STRUCTURES Introduction Structural geology is the study of the ways in which rocks or sediments are arranged and deformed

More information

Writing learning objectives

Writing learning objectives Writing learning objectives This material was excerpted and adapted from the following web site: http://www.utexas.edu/academic/diia/assessment/iar/students/plan/objectives/ What is a learning objective?

More information

Canny Edge Detection

Canny Edge Detection Canny Edge Detection 09gr820 March 23, 2009 1 Introduction The purpose of edge detection in general is to significantly reduce the amount of data in an image, while preserving the structural properties

More information

3 An Illustrative Example

3 An Illustrative Example Objectives An Illustrative Example Objectives - Theory and Examples -2 Problem Statement -2 Perceptron - Two-Input Case -4 Pattern Recognition Example -5 Hamming Network -8 Feedforward Layer -8 Recurrent

More information

E XPLORING QUADRILATERALS

E XPLORING QUADRILATERALS E XPLORING QUADRILATERALS E 1 Geometry State Goal 9: Use geometric methods to analyze, categorize and draw conclusions about points, lines, planes and space. Statement of Purpose: The activities in this

More information

Problem of the Month: Cutting a Cube

Problem of the Month: Cutting a Cube Problem of the Month: The Problems of the Month (POM) are used in a variety of ways to promote problem solving and to foster the first standard of mathematical practice from the Common Core State Standards:

More information

Customer Training Material. Lecture 4. Meshing in Mechanical. Mechanical. ANSYS, Inc. Proprietary 2010 ANSYS, Inc. All rights reserved.

Customer Training Material. Lecture 4. Meshing in Mechanical. Mechanical. ANSYS, Inc. Proprietary 2010 ANSYS, Inc. All rights reserved. Lecture 4 Meshing in Mechanical Introduction to ANSYS Mechanical L4-1 Chapter Overview In this chapter controlling meshing operations is described. Topics: A. Global Meshing Controls B. Local Meshing Controls

More information

CHECKLIST FOR THE DEGREE PROJECT REPORT

CHECKLIST FOR THE DEGREE PROJECT REPORT Kerstin Frenckner, kfrenck@csc.kth.se Copyright CSC 25 mars 2009 CHECKLIST FOR THE DEGREE PROJECT REPORT This checklist has been written to help you check that your report matches the demands that are

More information

Position Classification Standard for Management and Program Clerical and Assistance Series, GS-0344

Position Classification Standard for Management and Program Clerical and Assistance Series, GS-0344 Position Classification Standard for Management and Program Clerical and Assistance Series, GS-0344 Table of Contents SERIES DEFINITION... 2 EXCLUSIONS... 2 OCCUPATIONAL INFORMATION... 3 TITLES... 6 EVALUATING

More information

The Scientific Data Mining Process

The Scientific Data Mining Process Chapter 4 The Scientific Data Mining Process When I use a word, Humpty Dumpty said, in rather a scornful tone, it means just what I choose it to mean neither more nor less. Lewis Carroll [87, p. 214] In

More information

A Fishy Tale. Observing the Circulatory System of a Goldfish with a Compound Light Microscope

A Fishy Tale. Observing the Circulatory System of a Goldfish with a Compound Light Microscope A Fishy Tale Observing the Circulatory System of a Goldfish with a Compound Light Microscope A Fishy Tale About this Lesson In this lesson, students will explore a computer animation of the human body

More information

CHAPTER 1: THE LUNGS AND RESPIRATORY SYSTEM

CHAPTER 1: THE LUNGS AND RESPIRATORY SYSTEM CHAPTER 1: THE LUNGS AND RESPIRATORY SYSTEM INTRODUCTION Lung cancer affects a life-sustaining system of the body, the respiratory system. The respiratory system is responsible for one of the essential

More information

Introduction to Pattern Recognition

Introduction to Pattern Recognition Introduction to Pattern Recognition Selim Aksoy Department of Computer Engineering Bilkent University saksoy@cs.bilkent.edu.tr CS 551, Spring 2009 CS 551, Spring 2009 c 2009, Selim Aksoy (Bilkent University)

More information

Volume visualization I Elvins

Volume visualization I Elvins Volume visualization I Elvins 1 surface fitting algorithms marching cubes dividing cubes direct volume rendering algorithms ray casting, integration methods voxel projection, projected tetrahedra, splatting

More information

1. Abstract 2. Introduction 3. Algorithms and Techniques

1. Abstract 2. Introduction 3. Algorithms and Techniques MS PROJECT Virtual Surgery Piyush Soni under the guidance of Dr. Jarek Rossignac, Brian Whited Georgia Institute of Technology, Graphics, Visualization and Usability Center Atlanta, GA piyush_soni@gatech.edu,

More information

An Extremely Inexpensive Multisampling Scheme

An Extremely Inexpensive Multisampling Scheme An Extremely Inexpensive Multisampling Scheme Tomas Akenine-Möller Ericsson Mobile Platforms AB Chalmers University of Technology Technical Report No. 03-14 Note that this technical report will be extended

More information

Turing Machines: An Introduction

Turing Machines: An Introduction CIT 596 Theory of Computation 1 We have seen several abstract models of computing devices: Deterministic Finite Automata, Nondeterministic Finite Automata, Nondeterministic Finite Automata with ɛ-transitions,

More information

VisualCalc AdWords Dashboard Indicator Whitepaper Rev 3.2

VisualCalc AdWords Dashboard Indicator Whitepaper Rev 3.2 VisualCalc AdWords Dashboard Indicator Whitepaper Rev 3.2 873 Embarcadero Drive, Suite 3 El Dorado Hills, California 95762 916.939.2020 www.visualcalc.com Introduction The VisualCalc AdWords Dashboard

More information

Analecta Vol. 8, No. 2 ISSN 2064-7964

Analecta Vol. 8, No. 2 ISSN 2064-7964 EXPERIMENTAL APPLICATIONS OF ARTIFICIAL NEURAL NETWORKS IN ENGINEERING PROCESSING SYSTEM S. Dadvandipour Institute of Information Engineering, University of Miskolc, Egyetemváros, 3515, Miskolc, Hungary,

More information

1.2 Solving a System of Linear Equations

1.2 Solving a System of Linear Equations 1.. SOLVING A SYSTEM OF LINEAR EQUATIONS 1. Solving a System of Linear Equations 1..1 Simple Systems - Basic De nitions As noticed above, the general form of a linear system of m equations in n variables

More information

Solving simultaneous equations using the inverse matrix

Solving simultaneous equations using the inverse matrix Solving simultaneous equations using the inverse matrix 8.2 Introduction The power of matrix algebra is seen in the representation of a system of simultaneous linear equations as a matrix equation. Matrix

More information

N Q.3 Choose a level of accuracy appropriate to limitations on measurement when reporting quantities.

N Q.3 Choose a level of accuracy appropriate to limitations on measurement when reporting quantities. Performance Assessment Task Swimming Pool Grade 9 The task challenges a student to demonstrate understanding of the concept of quantities. A student must understand the attributes of trapezoids, how to

More information

Robust Real-Time Face Detection

Robust Real-Time Face Detection Robust Real-Time Face Detection International Journal of Computer Vision 57(2), 137 154, 2004 Paul Viola, Michael Jones 授 課 教 授 : 林 信 志 博 士 報 告 者 : 林 宸 宇 報 告 日 期 :96.12.18 Outline Introduction The Boost

More information

What role does the nucleolus have in cell functioning? Glial cells

What role does the nucleolus have in cell functioning? Glial cells Nervous System Lab The nervous system of vertebrates can be divided into the central nervous system, which consists of the brain and spinal cord, and the peripheral nervous system, which contains nerves,

More information

Extend Table Lens for High-Dimensional Data Visualization and Classification Mining

Extend Table Lens for High-Dimensional Data Visualization and Classification Mining Extend Table Lens for High-Dimensional Data Visualization and Classification Mining CPSC 533c, Information Visualization Course Project, Term 2 2003 Fengdong Du fdu@cs.ubc.ca University of British Columbia

More information

A Review of Anomaly Detection Techniques in Network Intrusion Detection System

A Review of Anomaly Detection Techniques in Network Intrusion Detection System A Review of Anomaly Detection Techniques in Network Intrusion Detection System Dr.D.V.S.S.Subrahmanyam Professor, Dept. of CSE, Sreyas Institute of Engineering & Technology, Hyderabad, India ABSTRACT:In

More information

Learning Example. Machine learning and our focus. Another Example. An example: data (loan application) The data and the goal

Learning Example. Machine learning and our focus. Another Example. An example: data (loan application) The data and the goal Learning Example Chapter 18: Learning from Examples 22c:145 An emergency room in a hospital measures 17 variables (e.g., blood pressure, age, etc) of newly admitted patients. A decision is needed: whether

More information

INTRODUCTION. The principles of which are to:

INTRODUCTION. The principles of which are to: Taking the Pain Out of Chromatographic Peak Integration Shaun Quinn, 1 Peter Sauter, 1 Andreas Brunner, 1 Shawn Anderson, 2 Fraser McLeod 1 1 Dionex Corporation, Germering, Germany; 2 Dionex Corporation,

More information

Creation of an Unlimited Database of Virtual Bone. Validation and Exploitation for Orthopedic Devices

Creation of an Unlimited Database of Virtual Bone. Validation and Exploitation for Orthopedic Devices Creation of an Unlimited Database of Virtual Bone Population using Mesh Morphing: Validation and Exploitation for Orthopedic Devices Najah Hraiech 1, Christelle Boichon 1, Michel Rochette 1, 2 Thierry

More information

Potential of face area data for predicting sharpness of natural images

Potential of face area data for predicting sharpness of natural images Potential of face area data for predicting sharpness of natural images Mikko Nuutinen a, Olli Orenius b, Timo Säämänen b, Pirkko Oittinen a a Dept. of Media Technology, Aalto University School of Science

More information

Reflection and Refraction

Reflection and Refraction Equipment Reflection and Refraction Acrylic block set, plane-concave-convex universal mirror, cork board, cork board stand, pins, flashlight, protractor, ruler, mirror worksheet, rectangular block worksheet,

More information

Specific Usage of Visual Data Analysis Techniques

Specific Usage of Visual Data Analysis Techniques Specific Usage of Visual Data Analysis Techniques Snezana Savoska 1 and Suzana Loskovska 2 1 Faculty of Administration and Management of Information systems, Partizanska bb, 7000, Bitola, Republic of Macedonia

More information

Review of Biomedical Image Processing

Review of Biomedical Image Processing BOOK REVIEW Open Access Review of Biomedical Image Processing Edward J Ciaccio Correspondence: ciaccio@columbia. edu Department of Medicine, Columbia University, New York, USA Abstract This article is

More information

www.studymafia.org Seminar report Rapid Prototyping Submitted in partial fulfillment of the requirement for the award of degree Of Mechanical

www.studymafia.org Seminar report Rapid Prototyping Submitted in partial fulfillment of the requirement for the award of degree Of Mechanical A Seminar report On Rapid Prototyping Submitted in partial fulfillment of the requirement for the award of degree Of Mechanical SUBMITTED TO: SUBMITTED BY: www.studymafia.org www.studymafia.org Preface

More information

ORGANIZATIONAL DESIGN AND ADAPTATION IN RESPONSE TO CRISES: THEORY AND PRACTICE

ORGANIZATIONAL DESIGN AND ADAPTATION IN RESPONSE TO CRISES: THEORY AND PRACTICE ORGANIZATIONAL DESIGN AND ADAPTATION IN RESPONSE TO CRISES: THEORY AND PRACTICE ZHIANG ("JOHN") LIN School of Management University of Texas at Dallas Richardson, TX 75083 KATHLEEN M. CARLEY Carnegie Mellon

More information

T O B C A T C A S E G E O V I S A T DETECTIE E N B L U R R I N G V A N P E R S O N E N IN P A N O R A MISCHE BEELDEN

T O B C A T C A S E G E O V I S A T DETECTIE E N B L U R R I N G V A N P E R S O N E N IN P A N O R A MISCHE BEELDEN T O B C A T C A S E G E O V I S A T DETECTIE E N B L U R R I N G V A N P E R S O N E N IN P A N O R A MISCHE BEELDEN Goal is to process 360 degree images and detect two object categories 1. Pedestrians,

More information

Predicting Flight Delays

Predicting Flight Delays Predicting Flight Delays Dieterich Lawson jdlawson@stanford.edu William Castillo will.castillo@stanford.edu Introduction Every year approximately 20% of airline flights are delayed or cancelled, costing

More information

The Bending Strength of Pasta

The Bending Strength of Pasta The Bending Strength of Pasta 1.105 Lab #1 Louis L. Bucciarelli 9 September, 2003 Lab Partners: [Name1] [Name2] Data File: Tgroup3.txt On the cover page, include your name, the names of your lab partners,

More information

Linear Programming Problems

Linear Programming Problems Linear Programming Problems Linear programming problems come up in many applications. In a linear programming problem, we have a function, called the objective function, which depends linearly on a number

More information

TIBCO Spotfire Guided Analytics. Transferring Best Practice Analytics from Experts to Everyone

TIBCO Spotfire Guided Analytics. Transferring Best Practice Analytics from Experts to Everyone TIBCO Spotfire Guided Analytics Transferring Best Practice Analytics from Experts to Everyone Introduction Business professionals need powerful and easy-to-use data analysis applications in order to make

More information

Building 3D anatomical scenes on the Web

Building 3D anatomical scenes on the Web Building 3D anatomical scenes on the Web F. Evesque, S. Gerlach, R. D. Hersch Ecole Polytechnique Fédérale de Lausanne (EPFL) http://visiblehuman.epfl.ch Abstract We propose a new service for building

More information

2x + y = 3. Since the second equation is precisely the same as the first equation, it is enough to find x and y satisfying the system

2x + y = 3. Since the second equation is precisely the same as the first equation, it is enough to find x and y satisfying the system 1. Systems of linear equations We are interested in the solutions to systems of linear equations. A linear equation is of the form 3x 5y + 2z + w = 3. The key thing is that we don t multiply the variables

More information

Investigation of Color Aliasing of High Spatial Frequencies and Edges for Bayer-Pattern Sensors and Foveon X3 Direct Image Sensors

Investigation of Color Aliasing of High Spatial Frequencies and Edges for Bayer-Pattern Sensors and Foveon X3 Direct Image Sensors Investigation of Color Aliasing of High Spatial Frequencies and Edges for Bayer-Pattern Sensors and Foveon X3 Direct Image Sensors Rudolph J. Guttosch Foveon, Inc. Santa Clara, CA Abstract The reproduction

More information

Test Automation Architectures: Planning for Test Automation

Test Automation Architectures: Planning for Test Automation Test Automation Architectures: Planning for Test Automation Douglas Hoffman Software Quality Methods, LLC. 24646 Heather Heights Place Saratoga, California 95070-9710 Phone 408-741-4830 Fax 408-867-4550

More information

Geometry Solve real life and mathematical problems involving angle measure, area, surface area and volume.

Geometry Solve real life and mathematical problems involving angle measure, area, surface area and volume. Performance Assessment Task Pizza Crusts Grade 7 This task challenges a student to calculate area and perimeters of squares and rectangles and find circumference and area of a circle. Students must find

More information

Introduction to nonparametric regression: Least squares vs. Nearest neighbors

Introduction to nonparametric regression: Least squares vs. Nearest neighbors Introduction to nonparametric regression: Least squares vs. Nearest neighbors Patrick Breheny October 30 Patrick Breheny STA 621: Nonparametric Statistics 1/16 Introduction For the remainder of the course,

More information

Session 7 Fractions and Decimals

Session 7 Fractions and Decimals Key Terms in This Session Session 7 Fractions and Decimals Previously Introduced prime number rational numbers New in This Session period repeating decimal terminating decimal Introduction In this session,

More information

2) Write in detail the issues in the design of code generator.

2) Write in detail the issues in the design of code generator. COMPUTER SCIENCE AND ENGINEERING VI SEM CSE Principles of Compiler Design Unit-IV Question and answers UNIT IV CODE GENERATION 9 Issues in the design of code generator The target machine Runtime Storage

More information

Design Analysis of Everyday Thing: Nintendo Wii Remote

Design Analysis of Everyday Thing: Nintendo Wii Remote 1 Philip Stubbs Design Analysis of Everyday Thing: Nintendo Wii Remote I. Introduction: Ever since being released in November 2006, the Nintendo Wii gaming system has revolutionized the gaming experience

More information

REPORT WRITING GUIDE

REPORT WRITING GUIDE Report Writing Guide F2009 1 REPORT WRITING GUIDE Introduction The importance of good report writing and data presentation cannot be overemphasized. No matter how good an experiment, or how brilliant a

More information

Network Machine Learning Research Group. Intended status: Informational October 19, 2015 Expires: April 21, 2016

Network Machine Learning Research Group. Intended status: Informational October 19, 2015 Expires: April 21, 2016 Network Machine Learning Research Group S. Jiang Internet-Draft Huawei Technologies Co., Ltd Intended status: Informational October 19, 2015 Expires: April 21, 2016 Abstract Network Machine Learning draft-jiang-nmlrg-network-machine-learning-00

More information

0.75 75% ! 3 40% 0.65 65% Percent Cards. This problem gives you the chance to: relate fractions, decimals and percents

0.75 75% ! 3 40% 0.65 65% Percent Cards. This problem gives you the chance to: relate fractions, decimals and percents Percent Cards This problem gives you the chance to: relate fractions, decimals and percents Mrs. Lopez makes sets of cards for her math class. All the cards in a set have the same value. Set A 3 4 0.75

More information

5 Homogeneous systems

5 Homogeneous systems 5 Homogeneous systems Definition: A homogeneous (ho-mo-jeen -i-us) system of linear algebraic equations is one in which all the numbers on the right hand side are equal to : a x +... + a n x n =.. a m

More information

Visualization and Feature Extraction, FLOW Spring School 2016 Prof. Dr. Tino Weinkauf. Flow Visualization. Image-Based Methods (integration-based)

Visualization and Feature Extraction, FLOW Spring School 2016 Prof. Dr. Tino Weinkauf. Flow Visualization. Image-Based Methods (integration-based) Visualization and Feature Extraction, FLOW Spring School 2016 Prof. Dr. Tino Weinkauf Flow Visualization Image-Based Methods (integration-based) Spot Noise (Jarke van Wijk, Siggraph 1991) Flow Visualization:

More information

1.1.1 Introduction to Cloud Computing

1.1.1 Introduction to Cloud Computing 1 CHAPTER 1 INTRODUCTION 1.1 CLOUD COMPUTING 1.1.1 Introduction to Cloud Computing Computing as a service has seen a phenomenal growth in recent years. The primary motivation for this growth has been the

More information

Image Segmentation and Registration

Image Segmentation and Registration Image Segmentation and Registration Dr. Christine Tanner (tanner@vision.ee.ethz.ch) Computer Vision Laboratory, ETH Zürich Dr. Verena Kaynig, Machine Learning Laboratory, ETH Zürich Outline Segmentation

More information

Scanners and How to Use Them

Scanners and How to Use Them Written by Jonathan Sachs Copyright 1996-1999 Digital Light & Color Introduction A scanner is a device that converts images to a digital file you can use with your computer. There are many different types

More information

Reconciliation Best Practice

Reconciliation Best Practice INTRODUCTION This paper provides an outline statement of what we consider to be best practice with respect to the use of reconciliation software used by asset managers. It is not a description of any particular

More information

Algorithms and optimization for search engine marketing

Algorithms and optimization for search engine marketing Algorithms and optimization for search engine marketing Using portfolio optimization to achieve optimal performance of a search campaign and better forecast ROI Contents 1: The portfolio approach 3: Why

More information

RADIOLOGY SERVICES. By Dr Lim Eng Kok 1

RADIOLOGY SERVICES. By Dr Lim Eng Kok 1 INTRODUCTION RADIOLOGY SERVICES By Dr Lim Eng Kok 1 Radiology is the branch of medicine that deals with the use of ionising (e.g. x- rays and radio-isotopes) and non-ionising radiation (e.g. ultrasound

More information

Microsoft Excel 2010 Charts and Graphs

Microsoft Excel 2010 Charts and Graphs Microsoft Excel 2010 Charts and Graphs Email: training@health.ufl.edu Web Page: http://training.health.ufl.edu Microsoft Excel 2010: Charts and Graphs 2.0 hours Topics include data groupings; creating

More information

Scheduling. Getting Started. Scheduling 79

Scheduling. Getting Started. Scheduling 79 Scheduling 9 Scheduling An event planner has to juggle many workers completing different tasks, some of which must be completed before others can begin. For example, the banquet tables would need to be

More information

The Universal Laws of Gravitation. Copyright 2012 Joseph A. Rybczyk

The Universal Laws of Gravitation. Copyright 2012 Joseph A. Rybczyk The Universal Laws of Gravitation Copyright 2012 Joseph A. Rybczyk Abstract Close examination of Newton s universal law of gravitation and Galileo s discovery that all objects fall to Earth at the same

More information

Clustering & Visualization

Clustering & Visualization Chapter 5 Clustering & Visualization Clustering in high-dimensional databases is an important problem and there are a number of different clustering paradigms which are applicable to high-dimensional data.

More information

ARTIFICIAL INTELLIGENCE METHODS IN EARLY MANUFACTURING TIME ESTIMATION

ARTIFICIAL INTELLIGENCE METHODS IN EARLY MANUFACTURING TIME ESTIMATION 1 ARTIFICIAL INTELLIGENCE METHODS IN EARLY MANUFACTURING TIME ESTIMATION B. Mikó PhD, Z-Form Tool Manufacturing and Application Ltd H-1082. Budapest, Asztalos S. u 4. Tel: (1) 477 1016, e-mail: miko@manuf.bme.hu

More information

Medical Image Segmentation of PACS System Image Post-processing *

Medical Image Segmentation of PACS System Image Post-processing * Medical Image Segmentation of PACS System Image Post-processing * Lv Jie, Xiong Chun-rong, and Xie Miao Department of Professional Technical Institute, Yulin Normal University, Yulin Guangxi 537000, China

More information

Edge tracking for motion segmentation and depth ordering

Edge tracking for motion segmentation and depth ordering Edge tracking for motion segmentation and depth ordering P. Smith, T. Drummond and R. Cipolla Department of Engineering University of Cambridge Cambridge CB2 1PZ,UK {pas1001 twd20 cipolla}@eng.cam.ac.uk

More information

Multiscale Object-Based Classification of Satellite Images Merging Multispectral Information with Panchromatic Textural Features

Multiscale Object-Based Classification of Satellite Images Merging Multispectral Information with Panchromatic Textural Features Remote Sensing and Geoinformation Lena Halounová, Editor not only for Scientific Cooperation EARSeL, 2011 Multiscale Object-Based Classification of Satellite Images Merging Multispectral Information with

More information

4 Week Body Contour / Lipo Light Program

4 Week Body Contour / Lipo Light Program Natural Health Solutions 14698 Galaxie Ave. Apple Valley, MN 55124 (952) 891-22225 4 Week Body Contour / Lipo Light Program Welcome and Congratulations! This is an important decision towards improving

More information

How To Fix Out Of Focus And Blur Images With A Dynamic Template Matching Algorithm

How To Fix Out Of Focus And Blur Images With A Dynamic Template Matching Algorithm IJSTE - International Journal of Science Technology & Engineering Volume 1 Issue 10 April 2015 ISSN (online): 2349-784X Image Estimation Algorithm for Out of Focus and Blur Images to Retrieve the Barcode

More information

Twelve. Figure 12.1: 3D Curved MPR Viewer Window

Twelve. Figure 12.1: 3D Curved MPR Viewer Window Twelve The 3D Curved MPR Viewer This Chapter describes how to visualize and reformat a 3D dataset in a Curved MPR plane: Curved Planar Reformation (CPR). The 3D Curved MPR Viewer is a window opened from

More information

Predicting the Risk of Heart Attacks using Neural Network and Decision Tree

Predicting the Risk of Heart Attacks using Neural Network and Decision Tree Predicting the Risk of Heart Attacks using Neural Network and Decision Tree S.Florence 1, N.G.Bhuvaneswari Amma 2, G.Annapoorani 3, K.Malathi 4 PG Scholar, Indian Institute of Information Technology, Srirangam,

More information

This software agent helps industry professionals review compliance case investigations, find resolutions, and improve decision making.

This software agent helps industry professionals review compliance case investigations, find resolutions, and improve decision making. Lost in a sea of data? Facing an external audit? Or just wondering how you re going meet the challenges of the next regulatory law? When you need fast, dependable support and company-specific solutions

More information

CHAPTER 1. Introduction to CAD/CAM/CAE Systems

CHAPTER 1. Introduction to CAD/CAM/CAE Systems CHAPTER 1 1.1 OVERVIEW Introduction to CAD/CAM/CAE Systems Today s industries cannot survive worldwide competition unless they introduce new products with better quality (quality, Q), at lower cost (cost,

More information

Analysis of Uncertain Data: Tools for Representation and Processing

Analysis of Uncertain Data: Tools for Representation and Processing Analysis of Uncertain Data: Tools for Representation and Processing Bin Fu binf@cs.cmu.edu Eugene Fink e.fink@cs.cmu.edu Jaime G. Carbonell jgc@cs.cmu.edu Abstract We present initial work on a general-purpose

More information

User research for information architecture projects

User research for information architecture projects Donna Maurer Maadmob Interaction Design http://maadmob.com.au/ Unpublished article User research provides a vital input to information architecture projects. It helps us to understand what information

More information

Quantifying Seasonal Variation in Cloud Cover with Predictive Models

Quantifying Seasonal Variation in Cloud Cover with Predictive Models Quantifying Seasonal Variation in Cloud Cover with Predictive Models Ashok N. Srivastava, Ph.D. ashok@email.arc.nasa.gov Deputy Area Lead, Discovery and Systems Health Group Leader, Intelligent Data Understanding

More information

A MACHINE LEARNING APPROACH TO FILTER UNWANTED MESSAGES FROM ONLINE SOCIAL NETWORKS

A MACHINE LEARNING APPROACH TO FILTER UNWANTED MESSAGES FROM ONLINE SOCIAL NETWORKS A MACHINE LEARNING APPROACH TO FILTER UNWANTED MESSAGES FROM ONLINE SOCIAL NETWORKS Charanma.P 1, P. Ganesh Kumar 2, 1 PG Scholar, 2 Assistant Professor,Department of Information Technology, Anna University

More information

Data, Measurements, Features

Data, Measurements, Features Data, Measurements, Features Middle East Technical University Dep. of Computer Engineering 2009 compiled by V. Atalay What do you think of when someone says Data? We might abstract the idea that data are

More information

Heuristic Evaluation of Three Cellular Phones

Heuristic Evaluation of Three Cellular Phones Heuristic Evaluation of Three Cellular Phones Siemens CF62T Samsung VGA1000 Motorola T720 ISE 217, Human Computer Interaction San Jose State University May 2005 Allison Eckholm Robert Murphy Richard Pack

More information

Unit 3L.4: Body Parts and Functions

Unit 3L.4: Body Parts and Functions Unit 3L.4: Organs in the Human body Keeping Healthy The Skeleton Science skills: Classification Observing Making models Data collection By the end of this unit you should: Compare the structure of humans

More information