Automatic Text Summarization of Biomedical Text Data: A Systematic Review

In recent years, the evolution of technology has led to an increase in text data obtained from many sources. biomedical domain, information also evidenced this accelerated growth, and automatic summarization systems play essential role optimizing physicians’ time resources identifying relevant information. paper, we present a systematic review research for textual data, focusing mainly on methods employed, type input text, areas application, evaluation metrics used assess systems. The survey was limited period between 1st January 2014 15th March 2022. collected WoS, IEEE, ACM digital libraries, while search strategies were developed with help experts NLP techniques previous reviews. four phases by PRISMA methodology conducted, five factors determined studies included: Input, Purpose, Output, Method, Evaluation metric. Results showed that 3.5% 801 met inclusion criteria. Moreover, Single-document, Biomedical Literature, Generic, Extractive proved be most common approaches based Machine Learning performed 16 Rouge (Recall-Oriented Understudy Gisting Evaluation) reported as metric 26 studies. This found more transformer-based methodologies purposes have been implemented compared survey. Additionally, there are still some challenges different domains, especially field terms demand further research.