Shaumik-Ashraf/BART-MIMIC-CXR
Use BART word embedder to do the abstractive summarization NLP task on radiology reports from MIMIC-CXR v2.0.0.
This project helps radiologists and medical researchers automatically create concise summaries from detailed radiology reports. It takes raw, lengthy chest X-ray reports as input and produces shorter, abstractive summaries, making it easier to grasp key findings quickly. Medical professionals working with large volumes of patient imaging data would find this useful.
No commits in the last 6 months.
Use this if you need to generate quick, high-level overviews from extensive radiology reports, especially those from the MIMIC-CXR dataset.
Not ideal if you need summaries from different types of medical reports or require strictly extractive (copy-pasted phrases) rather than abstractive summaries.
Stars
7
Forks
1
Language
Jupyter Notebook
License
MIT
Category
Last pushed
May 22, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/Shaumik-Ashraf/BART-MIMIC-CXR"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
kenlimmj/rouge
A Javascript implementation of the Recall-Oriented Understudy for Gisting Evaluation (ROUGE)...
uoneway/KoBertSum
KoBertSum은 BertSum모델을 한국어 데이터에 적용할 수 있도록 수정한 한국어 요약 모델입니다.
udibr/headlines
Automatically generate headlines to short articles
bheinzerling/pyrouge
A Python wrapper for the ROUGE summarization evaluation package
xiongma/transformer-pointer-generator
A Abstractive Summarization Implementation with Transformer and Pointer-generator