ChangeTracker evaluation

From NAMIC Wiki
Jump to: navigation, search
Home < ChangeTracker evaluation

Definitions

  • Sensitivity = TP / (TP + FN)
  • Specificity = TN / (FP + TN)
  • False Discovery Rate = FP / (FP + TP)

Syntehtic data

Growth experiment

  • deformation of the follow-up image
Case02
Case02 growth SENS.png
Case02 growth SPEC.png
Case02 growth FDR.png
Case03
Case03 growth SENS.png
Case03 growth SPEC.png
Case03 growth FDR.png

Shrinkage experiment

Case02
Case02 shrink SENS.png
Case02 shrink SPEC.png
Case02 shrink FDR.png
Case03
Case03 shrink SENS.png
Case03 shrink SPEC.png
Case03 shrink FDR.png

Growth + Noise

Case02
Case02 noisy SENS.png
Case02 noisy SPEC.png
Case02 noisy FDR.png
Case03
Case03 noisy SENS.png
Case03 noisy SPEC.png
Case03 noisy FDR.png


Growth + resampling

  • 2x
Case02
Case02 2x SENS.png
Case02 2x SPEC.png
Case02 2x FDR.png
Case03
Case03 2x SENS.png
Case03 2x SPEC.png
Case03 2x FDR.png


  • 3x
Case02
File:Case02 3x SENS.png
File:Case02 3x SPEC.png
File:Case02 3x FDR.png
Case03
File:Case03 3x SENS.png
File:Case03 3x SPEC.png
File:Case03 3x FDR.png


Clinical data

Clinical SENS.png

Clinical SPEC.png

Clinical FDR.png

References

Wikipedia: Sensitivity and Specificity

Retrieved from "https://www.na-mic.org/w/index.php?title=ChangeTracker_evaluation&oldid=52928"