Clinical NLP

PULSAR at MEDIQA-Sum 2023: Large Language Models Augmented by Synthetic Dialogue Convert Patient Dialogues to Medical Records
This paper describes PULSAR, our system submission at the ImageClef 2023 MediQA-Sum task on summarising patient-doctor dialogues into clinical records. We explore the impact of domain-specific pre-training and synthetic data augmentation while fine-tuning LLMs for this task.
Viktor Schlegel, Hao Li, Yuping Wu, Anand Subramanian, Thanh-Tung Nguyen, Abhinav Ramesh Kashyap, Daniel Beck, Xiaojun Zeng, Riza Theresa Batista-Navarro, Stefan Winkler, Goran Nenadic
PDF Cite Code
A robust section identification method for scanned electronic health records
We build a deep-learning based section identification that is robust to the errors introduced by OCR when processing image-based documents.
Anand Subramanian, Praveen Kumar Suresh, Sudarsun Santhiappan
PDF Cite Slides DOI