e: oae report writing
Posted by: Thierry MORLET (155.58.220.177)
Date: May 9, 2005 06:50PM

The diagnosis of OAEs is based on several criteria. These criteria might already be set on your machine (you’ll usually get a Pass or Refer on your screen), or you have to apply them to decide if OAEs are present or absent. In case your equipment comes with set criteria, you need to know what they are and if you want/need to change them.
The first criterion is the signal-to-noise ratio (for both DPs and TEOAEs): OAEs are said to be present when the signal-to-noise ratio (SNR) reaches a certain value. Teams around the world are not necessarily using the same value: if people agree that the SNR needs to be of at least 3 dB, some may want to use 3 dB, others 5 or 6 dB.
For DPOAEs, the SNR is computed for each recorded DP. For TEOAEs, you can use the SNR obtained for the whole response (usually from 0.5-6 kHz) or for several frequency bands (1, 2, 3, 4 and 5 kHz for example).
The second criterion used when recording TEOAEs is the reproducibility factor: once again, there is no “ideal” value: you can accept a value of 70% to consider the presence of TEOAEs but can also decide of a higher value.
Now, to report the result, you also need to “pick” criteria: let’s say, you’re recording 6 DPs in each ear: some might consider that the DPs are present if a response (based on the SNR) is obtained for at least 4 out of the 6 DPs. You may decide to set the limit to 5 DPs out of 6. For TEOAEs, you can consider only the SNR and reproducibility value for the whole spectrum, or decide that you need to present OAEs for at least 4 out of 5 (or 6) frequency bands to declare that OAEs are present in this particular ear.
These choices are usually made according to your recording environment (sound booth, nursery) and the characteristics of your patients (newborns, children….).
I think that the more precision we give in the report, the better it is for the patient: when I see a report stating (only) that OAEs are present or absent in both ears, I always question the validity of the diagnosis: how many DPs/frequency bands were considered, which SNR was used, etc….
In summary, it seems to me that if a simple response (OAEs present or absent) is useful for the person not aware of the OAE technology it should ALWAYS be followed by a brief description of the criteria used to make this decision (value of SNR) and a short summary of the results by frequency (4 DPs present at low and middle frequencies and 2 DPs absent at 5 and 6 kHz for example).

Re: OAE REport Writing
Posted by: michael gorga (IP Logged)
Date: June 01, 2005 12:32AM

If an outsider may put his 2-cents in, I have a couple of comments related to the response from Thierry Morlet. First, it may be important to note that SNR describes the reliability of the measurement (that is, how much bigger the signal is relative to the noise). But, two variables make up this number, signal and noise. The signal strength is presumably related to cochlear (outer hair cell) status, while the noise is not. Thus, one can have a 6 dB SNR with a signal level of 10 dB SPL (noise = 4 dB SPL), but can also achieve a 6 dB SNR with a signal level of -15 dB SPL (noise = -21 dB SPL). For high-frequency DPOAEs (f2 = 3 kHz and above),where achieving such low noise levels is not difficult, we would conclude that the results in the first case are more consistent with normal hearing but we would conclude that, in the second case, the DPOAE results are more consistent with hearing loss. It is for these reasons that we first evaluate SNR to make sure we have a reliable measurement and then use DPOAE level to reach a conclusion about cochlear status. If I may so immodest, please see our paper on interpreting DPOAEs in the clinic, which appeared in the December 1997 issue of Ear and Hearing. It provides an approach that forms the basis of our DPOAE reports, and could easily be used by others if they see value in it. It also explicitly recogizes that there is overlap in the response distribtuions from normal and impaired ears.

The second point I would like to make is to remind people that SNR and percent reliability are really two sides of the same coin. Clinicians should be careful not to assume that these are independent measures.

The final point that I would lilke to make, which I think is implicit in Theirry's last set of comments, relates to the possible need to alter criteria related to the prevalence of hearing loss in the target population. What may be appropriate in the well-baby nursery (where it is expected that 1 or 2 per 1000 babies will have heairng loss), may be less appropriate in the NICU (where the incidence might increase by a factor of 10), and may be even less appropriate in a speech and hearing or otolaryngology clinic. The hard part then becomes selecting the appropriate criterion for populations where the incidence of hearing loss is not uniform.

Thanks for allowing me to toss in some thoughts.

Michael Gorga

Re: OAE REport Writing
Posted by: LUCIANO MOUTINHO (IP Logged)
Date: October 26, 2005 09:02AM


Could you please give us one example of your reports.

Lucino Moutinho - Brasil

Re: OAE REport Writing
Posted by: Michael Gorga (IP Logged)
Date: October 26, 2005 10:08PM

Some time ago, we posted a series of lectures at the OAE Portal. In that series of lectures, there are several with the word "establishing" in the title. In one of them is the form that we use. In the Portal these files are presented as executable files, which I cannot open at my end, so I can't tell which of these lectures has the form in it. Perhaps you can opent them and look for it. If not,then please go to the December 1997 issue of Ear and Hearing. In it is a paper of ours entitled "From laboratory to clinic: A large scale study....." That paper has the form plus the coordinates if you wanted to generate your own form from those data.

I hope this is helpful.

Michael

Re: OAE REport Writing
Posted by: Stavros Hatzopoulos (IP Logged)
Date: October 26, 2005 11:02PM

The report that Michael talks about can be found as slide 12 in the lecture "Mike_DP Criteria II - A clinical form.exe". Download the file and run it. It will Generate a folder with an index.html file. Use any browser to view the slides.

Let me know if you have any problems ( email me at info@otoemissions.org)


Stavros