College - Author 1

College of Engineering

Department - Author 1

Computer Science Department

Degree Name - Author 1

BS in Computer Science

College - Author 2

College of Engineering

Department - Author 2

Computer Science Department

Degree - Author 2

BS in Computer Science

Date

12-2023

Primary Advisor

Zoë Wood, College of Engineering, Computer Science and Software Engineering Department

Abstract/Summary

DocAI presents a user-friendly platform for recording, transcribing, summarizing, and classifying doctor-patient consultations. The application utilizes AssemblyAI for conversational transcription, and the user interface allows users to either live-record consultations or upload an existing MP3 file. The classification process, powered by 'ml-classify-text,' organizes the consultation transcription into SOAP (Subjective, Objective, Assessment, and Plan) format – a widely used method of documentation for healthcare providers. The result of this development is a simple yet effective interface that effectively plays the role of a medical scribe. However, the application is still facing challenges of inconsistent summarization from the AssemblyAI backend. Future work involves refining summarization consistency and exploring integration with medical platforms.

Share

COinS