Using cTAKES to Build a Simple Speech Transcriber Plugin for an EMR

• Main Authors: Wenceslao, Stephen John Matthew, Estuar, Ma. Regina Justina E
• Format: text
• Published: Archīum Ateneo 2019
• Subjects: Applied computing; Life and medical sciences; Consumer health; Health care information systems; Computing methodologies; Artificial intelligence; Natural language processing; Speech recognition; Computer Sciences; Health Information Technology
• Online Access: https://archium.ateneo.edu/discs-faculty-pubs/186; https://dl.acm.org/doi/abs/10.1145/3340037.3340044
• Institution: Ateneo De Manila University

Electronic medical records (EMR) in general provide significant benefits to healthcare organizations and clinicians. However, a major challenge of clinicians who use EMRs is the lowered perceived quality of patient-doctor communication and interaction as a result of doctors being distracted with EMR use during consultations. A unique approach to this problem is through applications that automatically document clinical encounters in real-time. This study aims to develop a speech transcriber plugin for a web-based EMR for real-time clinical encounter documentation. We make use of available speech-to-text services on the web as well as cTAKES for clinical annotation. A draft summary of the clinical encounter is presented to the user in editable SOAP format. Blockchain technology for the speech recording is also explored to secure access to the recording. Internal testings showed that the prototype is able to capture audio conversations into text and parse the transcription for medical concepts. However, after a single formal usability evaluation we found that there is much to be done in terms of the usability of the summarization component.