Computer Aided Fitting (CAF?) Food for thought

Just thinking, could the way an optometrist identifies and corrects vision problems also be used to correct hearing problems? Generally speaking, an Optometrist does the following (forget the medical checking, that is another issue):

1st: Vision test to essentially get a base line.
2nd: Has patient look thru lenses and changing them from one lens to another asks the patient is this better or worse?
3rd: Depending on the answer, another lens is used and question repeats itself and continues in this loop until the optometrist knows how much correction is needed to give the patient better vision. Of course depending on the patients vision impairments, the end results will not always be perfect (20/20).

Generally speaking the optometrist is nothing but a living robot. A robot could do the same thing as the optometrist. The rub is that using a robot to flip the lenses would require that the robot be integrated with the lenses selector and the patient’s voice for the yes/no answer. Such a computer aided machine for vision correction would probably be too costly because of the mechanical actions it would be required to perform.

Let us look at a similar model used to correct hearing impairment. It might look like:
1st: The computer obtains the audio hearing data from the patient.
2nd: The computer using the audio data just collected would do a general fitting for the HAs.
3rd: The computer computer would then play audio for the patient asking if their hearing is better or worse.
4th: The answer and type of audio the patient was listening to would be used to determine what if any adjustment should be made to the HAs. The computer would continue in this loop using all the different audio recordings until the patient’s aids were adjusted to give the patient the best hearing they can have given their hearing disability and the limitations of the HAs.

This would pretty much take the audi out of the picture and establish a consistent level for fittings.

Eye glasses and hearing aids have nothing in common. It doesn’t matter if everyone had the same audiogram, their hearing needs will not be the same. They all have different environments to deal with, they all have different brains, they all my have gone different lengths of time with out getting the hearing aids they need. We have sacrificed way too much control of to computers already. We call a company we get so,e stupid computer speaking to us with a stupid script. We need to get back to person to person connect and conversations.
The audiogram is only a starting point then the real work to make hearing aids work begins. My audiogram and my final fit doesn’t even look close to each other. But I can finally hear and understand speech, songs, musical instruments, and even determine where sounds are coming from now. With my aids set to my audiogram that was impossible.
If what you bring up here came to past, the computers would to become way more smarter. And developers aren’t that smart either.

1 Like

What you’re proposing sounds a bit like this: http://www.two-pi.com/`. I agree with you on the potential benefits of computer aided fitting.

I don’t think it would be any less costly. You would need a rather complex surround sound system to simulate all the situations. Noisy restaurant, automobile, concert, etc. Probably have to include a t-coil setup, etc. Supply several brands of HA for the testing and setup.
I agree with cvkemp and I have been in the computer business for over 60 years. I still do programming and set up surveillance camera systems.

You would make adjustments on your phone while you’re in those environments. That’s the point.

d_Wooluf thanks for the link. That is pretty much in line with what I was thinking. I did not know about 2pi. The praises on that website for Voice Choice are right in line with my thinking. Hopefully something will be available in the not to distant future for HAs that are not OTC. I think in the general case, a head set could be used as the speakers. An appropriate sound track would provide the different listening environments. However, as you pointed out the phone could be used in the real hearing environment to make adjustment. I hadn’t given that any thought. I was thinking more along the lines of typical fitting scenario such as Phonak’s Target. Instead of the Target application that provides the fitting interface to the audi an application using software algorithms to determine the adjustments and an interface to the patient to collect general information. The software drivers currently used to interface with Noah Wireless and the HAs are already in place so the application would just need algorithms to essential use the same reasoning as that used by the audi’s to make the adjustments.

I think that the problem with computer aided fitting is that it takes time for the brain to adjust to hearing aids and to readjust to new hearing aids that do something different than the old ones. The standard 4 week trial probably isn’t even long enough, which is why Costco is so good with a 6 month trial. In contrast, results with vision lenses are instantaneous.

If computers start doing the fittings, I sure hope they don’t throw out REM. I think that the hype from hearing aid manufacturers to audiologists about new fitting software is partially responsible for some of them leaving REM out of their fitting routine.

2 Likes

A problem you have identified would be covered perfectly by a CAF. You can run the fitting as often as you wish from your phone or computer. If fitting changes are need, the software will make the changes.

As a retired it professional, and a hardware and software troubleshooter, don’t trust software to be that great. I have done enough quality control on software to know just how many bugs are never found, and as the say for ever bug found there are still at least 100 that hasn’t been found. This world is handing ver too much control to computers and it is going to come back to bite our butts.

You are correct about bugs in typical IT software. All to frequent typical software is developed using the “all speed and no control” design mythology and the bugs are numerous and software updates are frequent. However, not all software is developed that way. When fail safe software is developed, a major requirement is the code be bug free (rarely achieved, but the number of bugs are greatly reduced to an acceptable number). For instance, when your customer set is people riding in an elevator for a high rise building, the software written to control the elevator would have to have been rigorously tested and designed to proper standards (ISO Certification is a good start). Ooppps is not an option. So it cannot be thrown together like most software developed for typical customers in an IT environment. Other examples of fail safe software would be the micro code to control satellites, aircraft,automobiles, etc. So, I submit that FAC software could be designed so it was user friendly and had an acceptable failure rate. Would it be designed that way? Your guess is as good as mine.

I still am not sure my Audiologist has taught me a lot about the Gene2 software, but I still don’t know enough about my type of hearing loss to make the judgment to what is best and what could hurt. I look at my Audiogram then I see what my aids are set to and I shack my head, the two don’t even look close to the same. My Audiologist just says that my hearing requirements go well beyond what the audiogram dictates. You see over the last almost 3 years. dealing with 3 sets of hearing aids, and about 25 adjustments my aids are finally set up correctly for my hearing needs. but the software keeps throwing errors every time that I am in to get my aids checked. Say that my audiogram and my settings do not match.