Update README.md

README.md

@@ -1,10 +1,17 @@
 ## RadBERT-RoBERTa-4m
 
-This is one variant of our RadBERT models trained with 4 million VA reports, which achieves stronger radiology language understanding performance than previous medical domain models such BioBERT, Clinical-BERT, BLUE-BERT and BioMed-RoBERTa.
+This is one variant of our RadBERT models trained with 4 million deidentified medical reports from US VA hospital, which achieves stronger medical language understanding performance than previous medical domain models such as BioBERT, Clinical-BERT, BLUE-BERT and BioMed-RoBERTa. 
+
+Performances are evaluated on three tasks: 
+(a) abnormal sentence classification: sentence classification in radiology reports as reporting abnormal or normal findings; 
+(b) report coding: Assign a diagnostic code to a given radiology report for five different coding systems; 
+(c) report summarization: given the findings section of a radiology report, extractively select key sentences that summarized the findings.
 
 For details, check out the paper here: 
 [RadBERT: Adapting transformer-based language models to radiology](https://pubs.rsna.org/doi/abs/10.1148/ryai.210258)
 
+Code repeated to the paper is released at [this GitHub repo](https://github.com/zzxslp/RadBERT).
+
 ### How to use
 
 Here is an example of how to use this model to extract the features of a given text in PyTorch: