thtang / CheXNet-with-localization

Weakly Supervised Learning for Findings Detection in Medical Images
https://www.csie.ntu.edu.tw/~yvchen/f106-adl/doc/HTCMedical.pdf
GNU General Public License v3.0
314 stars 108 forks source link

Localization Eval Metric Definition&&Code Wanted. #11

Open magemoumou opened 4 years ago

magemoumou commented 4 years ago

Recently I am working on reconstruct your work's localization part. I fill up the Eval Part with Precision and Recall as eval metric. But the I just can not get your report's result even with the boundingbox.txt you provided. So I wonder if you could provide me with the Eval related code. I promise for scientific research only.

Aliktk commented 3 years ago

@magemoumou hello sir, how did you find the precision and recall in this study, can you provide an example that can save or give me these values. Furthermore, I get the BB for predicted and actual image. But it still confusing why they need an actual bounding box. as in real-world a radiologist gave only an image and the model perform and give the prediction in terms of disease name with bounding box 1,2 or 3. Give me some clarity on this confusion. @thtang if you can help me I will be thankfull thanks