Introductory Statistics for Laboratorians dealing with High Throughput Data sets Centers for Disease Control.

Slides:



Advertisements
Similar presentations
MANOVA (and DISCRIMINANT ANALYSIS) Alan Garnham, Spring 2005
Advertisements

Chapter 13 and 14. Error  Type 1: False positive. The effect was not really present, but our statistics lead us to believe that it is.  Type 2: False.
Multiple Analysis of Variance – MANOVA
Introductory Statistics for Laboratorians dealing with High Throughput Data sets Centers for Disease Control.
MANAGEMENT OF THE ABNORMAL PAP SMEAR
Tissue Fluorescence Spectroscopy Lecture 16. Outline Steady-state fluorescence –Instrumentation and Data Analysis Methods Statistical methods: Principal.
Introduction to Factorial ANOVA Designs
Evolution of Neoplasia The Uterine Cervix As a Model Raj C. Dash, MD Duke University Medical Center Durham, North Carolina.
Comparison of HPV Testing and Spectroscopy Combined with Cytology for the Detection of High- grade Cervical Neoplasia C Werner, W Griffith III, R Ashfaq,D.
More on ANOVA. Overview ANOVA as Regression Comparison Methods.
Statistics for Managers Using Microsoft® Excel 5th Edition
Lecture 22 Multiple Regression (Sections )
Cervical Cancer Cervical dysplasia Cervical cancer Causes Risk factors
What Is Multivariate Analysis of Variance (MANOVA)?
ENHANCING DETECTION.... YOUR PRACTICE.
1 1 Slide © 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole.
Basic Business Statistics, 11e © 2009 Prentice-Hall, Inc. Chap 15-1 Chapter 15 Multiple Regression Model Building Basic Business Statistics 11 th Edition.
Analysis of Variance & Multivariate Analysis of Variance
Stat 112: Lecture 9 Notes Homework 3: Due next Thursday
CERVICAL CANCER IN BOTSWANA By Monkgogi Khana Khomela and Wedu King.
Multiple Regression Farrokh Alemi, Ph.D. Kashif Haqqi M.D.
 The term post menopause is applied to women who have not experienced a menstrual bleed for a minimum of 12 months, assuming that they do still have.
ANCOVA Lecture 9 Andrew Ainsworth. What is ANCOVA?
Screening for Cervical Cancer by Visual Inspection Techniques Dr Aruna Batra VMMC & SJH.
CANCER CERVIX A PREVENTABLE CANCER Dr NEETA DHABHAI Sr Consultant. – Gynaecologist Member Expert - Indian Cancer Winners’ Association
Early methods of diagnostic in oral medicine
Cervical Cancer in California Janet Bates, MD MPH Research Program Director Research and Surveillance Program California Cancer Registry.
Articulating Confocal Endoscope for Imaging Cancers in vivo This work is supported in part by the Center for Subsurface Sensing and Imaging Systems, under.
SoftPAP® A Novel Collection Device for Cervical Cytology.
Introductory Statistics for Laboratorians dealing with High Throughput Data sets Centers for Disease Control.
Screening for cervical cancer. Screening for cervical lesions Common disease Cancer is preventable Screening is easy MUST BE PERFORMED.
1 1 Slide © 2008 Thomson South-Western. All Rights Reserved Chapter 15 Multiple Regression n Multiple Regression Model n Least Squares Method n Multiple.
International Scientific Congress for Students, Young Physicians and Pharmacists Marisiensis 2014.
Introductory Statistics for Laboratorians dealing with High Throughput Data sets Centers for Disease Control.
Multiple regression - Inference for multiple regression - A case study IPS chapters 11.1 and 11.2 © 2006 W.H. Freeman and Company.
MANOVA Mechanics. MANOVA is a multivariate generalization of ANOVA, so there are analogous parts to the simpler ANOVA equations First lets revisit Anova.
MBP1010H – Lecture 4: March 26, Multiple regression 2.Survival analysis Reading: Introduction to the Practice of Statistics: Chapters 2, 10 and 11.
Chapter 10: Analyzing Experimental Data Inferential statistics are used to determine whether the independent variable had an effect on the dependent variance.
Inference and Inferential Statistics Methods of Educational Research EDU 660.
Chapter 13 Multiple Regression
Chapter 7 Point Estimation of Parameters. Learning Objectives Explain the general concepts of estimating Explain important properties of point estimators.
Section 4.2 Building Linear Models from Data. OBJECTIVE 1.
Irwin/McGraw-Hill © Andrew F. Siegel, 1997 and Methods and Applications CHAPTER 15 ANOVA : Testing for Differences among Many Samples, and Much.
Tuesday, April 8 n Inferential statistics – Part 2 n Hypothesis testing n Statistical significance n continued….
Multivariate Analysis of Variance
Basic Business Statistics, 10e © 2006 Prentice-Hall, Inc. Chap 15-1 Chapter 15 Multiple Regression Model Building Basic Business Statistics 10 th Edition.
Introductory Statistics for Laboratorians dealing with High Throughput Data sets Centers for Disease Control.
 Seeks to determine group membership from predictor variables ◦ Given group membership, how many people can we correctly classify?
Unit 7 Statistics: Multivariate Analysis of Variance (MANOVA) & Discriminant Functional Analysis (DFA) Chat until class starts.
D/RS 1013 Discriminant Analysis. Discriminant Analysis Overview n multivariate extension of the one-way ANOVA n looks at differences between 2 or more.
Canonical Correlation. Canonical correlation analysis (CCA) is a statistical technique that facilitates the study of interrelationships among sets of.
Anthropometry Practical. Write Up You are provided with all the data from this year and previous years You should use the data to produce a short research.
Appendix 2 Comparison of screening from age 20 and age 25 Table of harms and benefits.
Date of download: 6/21/2016 Copyright © 2016 SPIE. All rights reserved. Selection of up to three hot spots per biopsy specimen (40×). Criteria for admissible.
Inferential Statistics Psych 231: Research Methods in Psychology.
Chapter 4: Basic Estimation Techniques
Differences Among Group Means: Multifactorial Analysis of Variance
Basic Estimation Techniques
Efficacy of random cervical biopsy and routine endocervical curettage in subjects with normal colposcopy.
Comparison of means test
Building Linear Models from Data
Hypothesis testing using contrasts
Basic Estimation Techniques
One-Way Analysis of Variance: Comparing Several Means
Tissa R. Hata, Theresa A. Scholz, Lynn K. Pershing 
Joanna Romaniuk Quanticate, Warsaw, Poland
Psych 231: Research Methods in Psychology
Fixed, Random and Mixed effects
Inferential Statistics
Chapter 10 Introduction to the Analysis of Variance
Presentation transcript:

Introductory Statistics for Laboratorians dealing with High Throughput Data sets Centers for Disease Control

Light Scattering Measurements for Detection of Precancerous Conditions of the Cervix. Objective: To examine the utility of in vivo elastic light scattering measurements to diagnose high grade squamous interepithelial lesions (HSIL) of the cervix. Methods: A newly developed fiber optic probe was used to measure light transport in the cervical epithelium of 36 patients undergoing standard colposcopy. Both unpolarized and polarized light transport were measured in the visible and near-infrared.

Groups: Three groups of women were identified using histopathology of the measured sites: – Group 1: Women with HSIL – Group 2: Women with low-grade lesions and – Group 3: Women with normal tissue Dependent Measures: Three spectroscopic parameters: – SP1 – SP2 – SP3

Analysis with ANOVA ANOVA: – Three groups – 1 outcome/dependent variable We have 3 Dependent Measures Could do 3 ANOVA’s But get 3 times the error rate ANOVA for SP1 Scores Group 1 HSIL Group 2 Low-Grade Lesions Group 3 Normal Tissue SP1 score

ANOVA for SP2 Scores Group 1 HSIL Group 2 Low-Grade Lesions Group 3 Normal Tissue SP2 score ANOVA for SP3 Scores Group 1 HSIL Group 2 Low-Grade Lesions Group 3 Normal Tissue SP3 score

MANOVA Provides one overall test for all the dependent measures combined Creates a linear combination of the dependent measures and tests the significance of the combined variable Treats the combined variable as a single dependent variable and does something similar to ANOVA on it.

MANOVA LAYOUT Group 1 HSIL Group 2 Low-Grade Lesions Group 2 Low-Grade Lesions SP1- Scores SP2- Scores SP3 - Scores SP1- Scores SP2- Scores SP3 - Scores SP1- Scores SP2- Scores SP3 - Scores X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3 X1X1 X2X2 X3X3

Computation of Combined Variable W Group 1 HSIL Group 2 Low-Grade Lesions Group 2 Low-Grade Lesions SP1- Scores SP2- Scores SP3 - Scores W’s Gp 1SP1- Scores SP2- Scores SP3 - Scores W’s Gp 2SP1- Scores SP2- Scores SP3 - Scores W’s Gp 3 X1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 W X1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 W X1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 W X1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 W X1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 W X1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 W X1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 W X1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 W X1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 W X1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 WX1X1 X2X2 X3X3 W

MANOVA The MANOVA tests the significance of the difference between the means of the W’s Computer selects the b’s that make the difference between the means as large as possible If it still isn’t significant you know none of the X’s are doing anything ANOVA for SP1 Scores Group 1 HSIL Group 2 Low-Grade Lesions Group 3 Normal Tissue WWW WWW WWW WWW WWW WWW WWW WWW Mean W for Group 1 Mean W for Group 2 Mean W for Group 3

MANOVA Provides an overall test of the differences among the means of the groups on the combined variable. Provides tests of the significance of each of the b’s. If the b for a particular X variable is not significantly different from zero, that X variable isn’t doing anything Provides the equation for the optimal combination of the variables. May lead to the discovery of emergent variables. Provides post hoc tests comparing the means of the groups on each variable

Results and Conclusions Results: Three spectroscopic parameters are statistically different for HSIL compared with low- grade lesions and normal tissue. Conclusions. : Reflectance measurements of elastically scattered light show promise as a non- invasive, real-time method to discriminate HSIL from other abnormalities and normal tissue. These results compare favorably with those obtained by fluorescence alone and by fluorescence combined with light scattering.