New paper citing MedCAT

We published a paper that was accepted to COLING, which just ended. If you want to add this to the list of papers that cite MedCAT, here the paper. Here is the repo that uses it.

Here’s the citation:

@inproceedings{landes-etal-2022-new,
    title = "A New Public Corpus for Clinical Section Identification: {M}ed{S}ec{I}d",
    author = "Landes, Paul  and
      Patel, Kunal  and
      Huang, Sean S.  and
      Webb, Adam  and
      Di Eugenio, Barbara  and
      Caragea, Cornelia",
    booktitle = "Proceedings of the 29th International Conference on Computational Linguistics",
    month = oct,
    year = "2022",
    address = "Gyeongju, Republic of Korea",
    publisher = "International Committee on Computational Linguistics",
    url = "https://aclanthology.org/2022.coling-1.326",
    pages = "3709--3721"
}
2 Likes

Very interesting use of cui2vec.

1 Like

Thanks @plandes - nice work and congrats for the paper acceptance! Added to the readme and media.md

1 Like

Yes, incidentally we also used cui2vec as concatenated embeddings to the word embeddings (after the publication of the paper, so this wasn’t mentioned). While performance did increase, it didn’t by much so I haven’t performed any ANOVA etc for statistical significance. This might be because while the cui2vec dimensional is high, the vocabulary is somewhat limited compared to other embeddings, especially relative to the size of UMLS.

Hi @plandes, paper added to the repo. Really interesting work, thanks for letting us know.

1 Like