Medical Reports Summarization Using Text-To-Text Transformer
Abdulkader Helwan, Danielle Azar, Dilber Uzun Ozsahin
Abstract
Summarization of medical reports in order to make them accessible to the large public is an important task that can highly benefit from the recent emergence of the deep learning and large language models (LLM). In this work, we propose a fine-tuned Text-to-Text Transformer (T5) to summarize such reports. We train and test our model on the publicly available Indiana Dataset. We evaluate it using the ROUGE set of metrics. The obtained results are promising.
Topics & Concepts
Automatic summarizationComputer scienceTransformerNatural language processingArtificial intelligenceTest setDeep learningInformation retrievalLanguage modelTraining setEngineeringElectrical engineeringVoltageTopic ModelingBiomedical Text Mining and OntologiesNatural Language Processing Techniques