search

stanfordmlgroup / chexpert-labeler

CheXpert NLP tool to extract observations from radiology reports.
MIT License
350 stars 79 forks source link

Results #29

Open luantunez opened 3 years ago

luantunez commented 3 years ago

Hello and thank you very much for sharing your great work! The prediction ran smoothly, but I am not sure on how to interpret the resulting csv file. There are some categories present, some with a "0", others with a "-1" and other void. What does that mean? Here is an example:

Reports,No Finding,Enlarged Cardiomediastinum,Cardiomegaly,Lung Lesion,Lung Opacity,Edema,Consolidation,Pneumonia,Atelectasis,Pneumothorax,Pleural Effusion,Pleural Other,Fracture,Support Devices
"PA and moderate loss of the chest demonstrate stable moderate cardiomediastinal silhouette with atherosclerotic calcifications of the aortic XXXX and mild aortic ectasia. Emphysematous changes with flattening of the hemidiaphragms. Blunting of the costophrenic XXXX, and XXXX secondary to scarring/emphysematous changes. No evidence of focal airspace consolidation large pleural effusion or pneumothorax. Visualized osseous structures appear intact",,-1.0,,,-1.0,,0.0,,,-1.0,-1.0,,,

Thank you very much in advance!

jirvin16 commented 3 years ago

Hi! 0 means the labeler found that the mention of the category was negated, -1 means the labeler found that the mention of the category was uncertain, and void means the labeler did not find a mention of the category. We treated 0 and void as negative, 1 as positive, and experimented with different methods of handling -1 (uncertain).

luantunez commented 3 years ago

Thank you very much!

al-nguyen commented 3 years ago

Hi @luantunez ! So sorry if my question is not relating to your question. But I wish to ask whether you were faced the issue of step 6 in the prerequisites? Or how can you overcome it because I cannot import bllipparser lib?

luantunez commented 3 years ago

Hi! I followed this issue https://github.com/stanfordmlgroup/chexpert-labeler/issues/3