|
Activity Number:
|
225
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #301170 |
|
Title:
|
Maximum Likelihood Estimators for Fraction of False Negative Lesions when Radiologist Panel Is Non-Unanimous
|
|
Author(s):
|
Arkendra De*+ and Gene Pennello+
|
|
Companies:
|
U.S. Food and Drug Administration and U.S. Food and Drug Administration
|
|
Address:
|
, , , , , ,
|
|
Keywords:
|
CAD ; radiology ; imaging
|
|
Abstract:
|
In settings where CAD (computer aided detection) devices for images are studied, the fraction of marks made by CAD which are true lesions yet missed by the radiologist reader (i.e. false negative fraction) is one metric which can be used to quantify CAD performance. In order to determine how radiologist readers will interpret the CAD marks, a panel of radiologists (e.g., three) is used in CAD studies. Situations in which the panel unanimously agrees with the interpretation of a CAD mark can easily be utilized in the estimation of false negative fraction. We present maximum likelihood estimators for false negative fraction when a variable number of radiologists agree.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2008 program |
