{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

BME 314 Lecture 10 Mia Markey 2010

# BME 314 Lecture 10 Mia Markey 2010 - Informatics 1...

This preview shows pages 1–14. Sign up to view the full content.

UT Biomedical Informatics Lab Informatics 1

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
UT Biomedical Informatics Lab Biomedical Informatics biomedical informatics = bioinformatics + medical  informatics bio informatics : study of the optimal storage and use  of  biological data  in biomedical research medical  informatics : study of the optimal storage  and use of  medical data  for clinical decision making  and related tasks. Now more commonly called  “clinical informatics”. 2
UT Biomedical Informatics Lab Confusion Matrix 303 17 973 965 + + - - Truth Prediction 3

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
UT Biomedical Informatics Lab Accuracy Number correctly classified samples divided  by the total number of samples Accuracy =  (965 + 303) / (965 + 17 + 973 + 303) = 56% 303 17 973 965 + + - - Truth Prediction 4
UT Biomedical Informatics Lab Two Kinds of Mistakes Accuracy just counts up the number of times the  right answer was obtained But, there are two  kinds of mistakes… Could classify as positive when actually negative Or, could classify as negative when actually positive So, we need two measures to separate these  errors 303 17 973 965 + + - - Truth Prediction 5

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
UT Biomedical Informatics Lab Accuracy Depends on Prevalence Another disadvantage of accuracy is that it depends upon the prevalence of the population, i.e., the frequency with which the two classes occur 6
UT Biomedical Informatics Lab Prevalence Number positive (disease) samples divided  by the total number of samples 303 17 973 965 + + - - Truth Prediction Prevalence =  (965 + 17) / (965 + 17 + 973 + 303) = 43% 7

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
UT Biomedical Informatics Lab Prevalence & Accuracy Accuracy =  (0 + 99) / (99 + 0 + 1 + 0) = 99% Prevalence =  1 / (99 + 0 + 1 + 0) = 1% 99 1 0 0 + + - - Truth Prediction 8
UT Biomedical Informatics Lab Sensitivity, Specificity Introduce two measures called  sensitivity  and  specificity  that separate the two kinds of errors and  do not depend on prevalence 9

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
UT Biomedical Informatics Lab Sensitivity or True Positive Fraction Number samples classified as positive that  were actually positive divided by the number  of samples that were actually positive Sensitivity = 965 / (965 + 17) = 98% 303 17 973 965 + + - - Truth Prediction 10
UT Biomedical Informatics Lab Specificity or True Negative Fraction Number samples classified as negative that  were actually negative divided by the number  of samples that were actually negative Specificity = 303 / (303 + 973) = 24% 303 17 973 965 + + - - Truth Prediction 11

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
UT Biomedical Informatics Lab Prevalence & Accuracy, Sensitivity & Specificity Accuracy =  (0 + 99) / (99 + 0 + 1 + 0) = 99% Prevalence =  1 / (99 + 0 + 1 + 0) = 1% 99 1 0 0 + + - - Truth Prediction Sensitivity =  (0) / (0 + 1) = 0% Specificity =  (99) / (0 + 99) = 100% 12
UT Biomedical Informatics Lab Dependence of Accuracy on Prevalence Accuracy =

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern