Extracting Biomedical Entities from Noisy Audio Transcripts
This work addresses the ASR-NLP gap in the biomedical domain, which is crucial for enhancing healthcare documentation practices, though it is incremental as it builds on existing ASR and NLP methods with a new dataset and cleaning approach.
The paper tackled the problem of extracting biomedical entities from noisy audio transcripts, a challenge known as the ASR-NLP gap, by introducing a novel dataset of almost 2,000 clean and noisy recordings and a transcript-cleaning method using GPT4, resulting in improved performance for named entity recognition in clinical applications.
Automatic Speech Recognition (ASR) technology is fundamental in transcribing spoken language into text, with considerable applications in the clinical realm, including streamlining medical transcription and integrating with Electronic Health Record (EHR) systems. Nevertheless, challenges persist, especially when transcriptions contain noise, leading to significant drops in performance when Natural Language Processing (NLP) models are applied. Named Entity Recognition (NER), an essential clinical task, is particularly affected by such noise, often termed the ASR-NLP gap. Prior works have primarily studied ASR's efficiency in clean recordings, leaving a research gap concerning the performance in noisy environments. This paper introduces a novel dataset, BioASR-NER, designed to bridge the ASR-NLP gap in the biomedical domain, focusing on extracting adverse drug reactions and mentions of entities from the Brief Test of Adult Cognition by Telephone (BTACT) exam. Our dataset offers a comprehensive collection of almost 2,000 clean and noisy recordings. In addressing the noise challenge, we present an innovative transcript-cleaning method using GPT4, investigating both zero-shot and few-shot methodologies. Our study further delves into an error analysis, shedding light on the types of errors in transcription software, corrections by GPT4, and the challenges GPT4 faces. This paper aims to foster improved understanding and potential solutions for the ASR-NLP gap, ultimately supporting enhanced healthcare documentation practices.