The brain produces what looks like a messy jumble of waves (EEG) that are hard for anyone without special training to make sense of. Researchers at Stanford are overcoming that by turning all those brain waves into sound waves, and in the process allowing clinicians who are not EEG specialists to easily detect silent seizures.
Silent seizures are hard to spot because they lack symptoms, but EEG allows a specialist with a keen eye and formal training to detect abnormalities in the charts. The “brain stethsocope” from Stanford takes advantage of our auditory system’s ability to detect fluctuations in the frequency of sound. When the EEG waves are converted to sound, using special algorithms of course, seizures show up as voice-like bursts that stand out from a white noise-like hum of when a brain is not seizing. In the researchers’ study, published in journal Epilepsia, even lowly medical students were able to spot the presence of seizures nearly 100% of the time.
Some details about the technology from the open access study:
We used a novel sonification algorithm to translate the low‐frequency EEG signals into the audible range by using them as modulators of a voicelike synthesized sound. EEG signals acquired from 1 temporal channel (T3‐T5 or T4‐T6) were applied as modulators of the synthesized sound. For each visual EEG record, we produced 2 sonified EEG clips (1 from each hemisphere). Using the sonification method, vocal pitch, loudness, and formant structure were directly varied by the input signal.
In the seizure case, the sonification renders epileptic spike trains into speechlike declamations with a loud, strong rhythmic character, which is easily distinguished by ear from the quieter, slower, and smoother‐sounding normal case.
Here’s a Stanford video report about the brain stethoscope:
Open access study in journal Epilepsia: Detecting silent seizures by their sound…