Enhancing MRI Communication and Healthcare with Speech-to-Text technology

Brisbane Convention & Exhibition Centre 

Communication during MRI procedures is often hindered by the scanner’s loud noise. To address this, NALscribe, a speech-to-text app developed by the National Acoustic Laboratories (NAL) in Australia, provides real-time captions and supports transcription in over 15 languages. Originally designed for audiology clinics, it has been widely adopted in settings such as healthcare appointments, classrooms, and family gatherings.

Clear communication between the radiographer and the patient or research participant inside the MRI scanner is crucial for ensuring compliance with scan requirements (e.g., task instructions), minimizing motion artifacts, and providing reassurance. However, the low audio quality of most MRI intercom systems, combined with earplugs, positioning foam, and the scanner’s inherent noise, makes it difficult for even those with normal hearing to understand speech. For individuals with hearing loss (including D/deaf people, older adults, and those with sensory processing disorders), the intelligibility of auditory communication in the scanner is significantly reduced.

We demonstrate how NALscribe can overcome these barriers by displaying real-time captions on a screen inside the MRI room, viewable by patients through a mirror. This approach enhances procedural efficiency, patient understanding, and the overall MRI experience.

Additionally, NALscribe AR, an augmented reality live captioning app for AR headsets, like the Apple Vision Pro, offers spatially-positioned captions. This technology could allow patients to simultaneously view captions alongside medical documents or scans displayed on the headset, creating a more immersive and informed experience. This has potential in preparatory phases, post-imaging debriefings, and neuroscience research, enhancing communication accessibility across healthcare and clinical practice.