GersteinLab at MEDIQA-Chat 2023: Clinical Note Summarization from Doctor-Patient Conversations through Fine-tuning and In-context Learning

TL;DR

This paper explores clinical note summarization from doctor-patient conversations using fine-tuning and in-context learning with GPT models, achieving high performance in the MEDIQA-2023 shared task.

Contribution

It introduces two pipelines—fine-tuning and in-context learning—for dialogue summarization in clinical settings, demonstrating state-of-the-art results and effective section header prediction.

Findings

Both methods achieved high ROUGE-1, BERTScore, and BLEURT scores.

ICL GPT-4 notes outperformed baseline models based on expert annotations.

Our team ranked fourth in the shared task.

Abstract

This paper presents our contribution to the MEDIQA-2023 Dialogue2Note shared task, encompassing both subtask A and subtask B. We approach the task as a dialogue summarization problem and implement two distinct pipelines: (a) a fine-tuning of a pre-trained dialogue summarization model and GPT-3, and (b) few-shot in-context learning (ICL) using a large language model, GPT-4. Both methods achieve excellent results in terms of ROUGE-1 F1, BERTScore F1 (deberta-xlarge-mnli), and BLEURT, with scores of 0.4011, 0.7058, and 0.5421, respectively. Additionally, we predict the associated section headers using RoBERTa and SciBERT based classification models. Our team ranked fourth among all teams, while each team is allowed to submit three runs as part of their submission. We also utilize expert annotations to demonstrate that the notes generated through the ICL GPT-4 are better than all other baselines. The code for our submission is available.