Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
11 Cards in this Set
- Front
- Back
|
Evaluating Diagnostic Tests - cause of over estimating accuracy |
due to people not considering bias rate - the likelihood of the disease occuring in the population= pre test probibility = post test probibility |
|
|
risk % of having a pos mamogram |
if a woman has breast cancer she is 90% likely to have pos mamogram woman without mamogram still has 7% liklihood of a pos reading |
|
|
2x2 contigency table - diagram example mamogram |
|
|
|
how to correct percentages to probability and calc sens and secificibility- Mamogram ex |
Pre test probability = .008 (.8%) Sensitivity (%correct positives)= .90 (90%) Specifity (%correct negatives : 100-7)= .93 (93%) |
|
|
Diagnostic Tests and SIgnal Detection Theory |
Diagnostic test can maximise the num of correct hits/diagnosis by adopting a liberal response bias This causes minimisation of false neg by increasing false pos as false neg are regarded as worse
however these false positives may lead to unecessary surgery and treatment
|
|
|
Risk of HIV given a positive screening |
prob of someone with no known risk behaviour is .01%. if a person has HIV 99.98% chance test will be positive. 99.99% chance is someone doesnt have aids that itll come back neg
Prob that person with a positive test has aids ? 50% |
|
|
How to Choose the best pass mark of your diagnostic test |
ROC curve -crucial if you want to pick a pass mark that would lead to best descrimination between people
|
|
|
Concussion- Mild brain injury and ROC curves |
does adding a non-word memory rep task improve detection of current test of cog effects? Ps were patients from hospital with and without concussions
It matters what we choose as a pass mark |
|
|
ROC Curve
|
Reciever Operating Characteristic Is a plot of Correct positive rate (sensitivity) versus false positive rate (1- specificity) where each point on the curve is a different pass mark for the test As true positive rate increases with a change in pass mark,so does false positive rate - sensitivity increases but specificity decreasses the more the line curves away from the diagonal the better the test is at discriminating people with disorder from controls. the area under the curve equated to the overall diagnostic accuracy of the test regardless of pass mark chosen.
|
|
|
where to find the best point for a passmark on ROC curve |
the point on the curve where sensitivity and specifictiy are highest. this point macimises correct hits and misses and minimizers fasse pos and neg |
|
|
still more of lecture you have not done |
yo |
