Extracting Appointment Spans from Medical Conversations

Abstract

Extracting structured information from medical conversations can reduce the documentation burden for doctors and help patients follow through with their care plan. In this paper, we introduce a novel task of extracting appointment spans from medical conversations. We frame this task as a sequence tagging problem and focus on extracting spans for appointment reason and time. However, annotating medical conversations is expensive, time-consuming, and requires considerable domain expertise. Hence, we propose to leverage weak supervision approaches, namely incomplete supervision, inaccurate supervision, and a hybrid supervision approach and evaluate both generic and domain-specific, ELMo, and BERT embeddings using sequence tagging models. The best performing model is the domain-specific BERT variant using weak hybrid supervision and obtains an F1 score of 79.32.

ASR Error Detection via Audio-Transcript entailment

“…Beyond dictation solutions, there has also been an uptick in adopting manual scribing solutions to reduce the documentation burden so that physicians can focus more on patient care. More recently, there has been a growing interest in leveraging the ASR transcripts and automating parts of the scribing workflows using spoken conversation understanding technologies such as utterance classification [5], information extraction [6,7,8], and conversation summarization [9]. Since speech recognition is upstream of all these dictation and scribing-centered documentation workflows, words mistranscribed by the ASR system [10] can be problematic, leading to users performing time-consuming and laborious corrections [11].…”