Home▸Bioinformatics▸CheXpert Uncertainty Lab

🩻 CheXpert Uncertainty Lab

Set simulated confidence scores across five chest X-ray findings and switch between CheXpert's U-Ones, U-Zeros and U-Ignore uncertainty-label policies to see how borderline findings flip classification and shift an illustrative AUC metric.

Bioinformatics2DModerate60 FPS

This simulator illustrates how the CheXpert chest X-ray dataset's three uncertainty-label policies, U-Ones, U-Zeros and U-Ignore, change both individual finding labels and an illustrative model-performance score whenever a finding's confidence sits in the ambiguous 0.40–0.60 band.

🔬 What It Demonstrates

🎮 How to Use

Drag each of the five finding sliders to set a simulated raw confidence score, then switch the policy selector between U-Ones, U-Zeros and U-Ignore and watch the bar chart relabel any finding sitting inside the shaded uncertain band, while the illustrative AUC score and its delta from baseline update to show the downstream effect on reported performance.

💡 Did You Know?

Did you know that in the original CheXpert study, no single uncertainty policy won for every finding? U-Ones tended to help some findings and hurt others, which is exactly why the paper reports results under multiple policies side by side rather than picking one "best" approach for the whole dataset.

⚙ Under the hood

CheXpertchest x-ray AIlabel uncertaintyU-OnesU-ZerosU-Ignoreradiology AImulti-label classification

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install