Back to List

Where is Cancer?

Sep 23, 2021 — 2 min read

Jonathan Yang

  • Breast Cancer Case

  • Of the three small masses, which one is malignant?

  • Please note: All cases underwent a de-identification process when received, thus no history or content can be provided regarding the patient.

. . .

If you selected Case #3, then you are correct!

The other two cases were benign.

Let's see what our Lunit AI found:

. . .


Several recent studies have shown that AI-CAD can increase the diagnostic accuracy of breast cancer without increasing false-positive rate which was a vulnerability of C-CAD.

Since the false-positivity rate can cause significant fatigue for readers as well as delays in reading time, it is important to reduce the false positives as much as to increase the breast cancer detection rate in order to be used efficiently as an aid.

According to a study by the Korean Society for Breast Screening, AI-CAD can contribute to increase the efficiency in image interpretation by significantly decreasing false-positive rate compared with C-CAD.

94% reduction in false-positive rate

Comparison of false-positive rate for C-CAD and AI-CAD

Reference: Kim, E, Lee, S, & Kim, M 2020, ‘Comparison of conventional CAD and AI-CAD applied to digital mammography in respect of false-positive marks’, Journal of the Korean Society for Breast Screening 2020; 17 ( 2 ) : 70–76.

AIArtificial IntelligenceBreast CancerBreast RadiologyBusinessLunitMachine LearningMammographyRadiology

More from Blog