Interpreting Waveforms


Written by Travis M. Moore
Last edited 22-Jun-2020


Art or Science?

You might be thinking, "Wait, didn't we just talk about this? Isn't quantifying waveforms the same thing?" This is a valid question. The fact of the matter, however, is that in order to quantify a wave, you first have to find the wave. Electrophysiological tests do not produce textbook results that are just waiting to be labelled. Actually being able to tell whether a peak or trough is present or absent can be extremely tricky and takes time to master. And the only way to practice is to get exposure to real waveforms and start labelling.

You might be wondering how hard this could possibly be. You see a peak or trough at an expected latency and you label it. Done. However, in real life you might find that peaks are wider than you expect and have multiple peaks and valleys on the slopes. You could also find your peak is really a stretch of flat line with no clear point. Especially when measuring thresholds, you will likely have trouble deciding if a peak is present or absent because of the surrounding noise (remember the noise looks exactly like the peaks and troughs we're after). These are just a few of the many obstacles in interpreting waveforms properly.

When clinicians label waves just by looking at the tracing (as opposed to an algorithm objectively doing it), the process is referred to as "peak-picking." It would be best if a completely objective computer algorithm could identify peaks with 100% accuracy, but it turns out it's extremely difficult to describe to a computer how to identify jittery waves in a mess of background noise. The human brain does a much better job than a computer at identifying waves. The human brain's ability to recognize patterns is unmatched by even the best computers at present. So the subjective "art" of peak-picking is the standard in the clinic.

The good news is that seasoned audiologists are very consistent in what they label. A 2016 study showed that individual audiologists are extremely reliable when they label waves of the ABR.1 Importantly, this study also showed that there was excellent consistency across different audiologists in labelling the same waveforms. In other words, audiologists very often picked the same peaks and valleys when looking at ABR waveforms. The highest consistency came from those with the most years working with ABRs. This study suggests that with time and experience, audiologists can indeed use electrophysiological tests reliably, despite the subjective nature of "peak-picking."

More good news is that there are several techniques for further cleaning up the waveforms so that things look a little less messy. Those techniques are discussed below.

Cleaning the Data

Content not yet available


Test Your Understanding

Answer One
Answer Two
Answer Three


REFERENCES

1Zaitoun, M., Cumming, S., Purcell, A., & O’brien, K. (2016). Inter and intra-reader variability in the threshold estimation of auditory brainstem response (ABR) results. Hearing, Balance and Communication, 14(1), 59-63.
Leave a Comment: