Another big experiment!

Things have been off to a rocky start for my “big idea” there’s a lot of moving parts, people to coordinate with, and things to manage. It’s no surprise things haven’t gone well since we started trying to get this project off the ground, but we’re still trying and that has to count for something… right? As it turns out, we may have lucked out this next week!

Okay so eight months or so ago I had a big idea, and since then we’ve been trying to make it happen. Hospital-PI is very excited and I’ve promised a lot of things from this project. Because truthfully if we can get the data, a lot of things will just fall right out that will make a big splash for us. It’s going to be awesome. The problem is getting that data and as we saw last time I gave an update, well things don’t always work out (here).

So the last time we attempted it, we had absolutely nothing to show for it. I’m convinced something came loose or a cable got disconnected and thus we recorded absolutely nothing, but beautiful line noise (a 60 Hz background noise from electricity, well 60 Hz here in the US anyway). Now hospital-PI is worried that the signal to noise ratio is so low that our data was completely obliterated by the noise. Signal to noise ratio is literally signal/noise so if the signal value is very low or the noise value is very high, the signal to noise ratio trends to zero, which is not ideal! We want a high number, the higher the better!

I know for certain that our signal to noise ratio should be fine. Hospital-PI is not used to the work I’m proposing so this is outside his comfort zone. Despite my conviction, he wants to see proof of something being recorded, so we were going to try again. Unfortunately, we hit a snag with that and we are hoping to record in the next week, but we couldn’t do anything the day we had hoped. That wasn’t our fault or the people we’re working with (surgeon-PI for example), it was just something outside everyone’s control.

That said, big idea was going to be a hard thing to pull off from the start. We have a very tight selection criteria for people we can recruit for the project and to make matters worse, even if we find people, there’s no promise they will agree to take part. Sometimes people just don’t want to risk it even though there’s no real risk involved. I don’t blame them and we remind them that the standard of care they will get is no way impacted by the choice they make to help us. We have gotten lucky and have a huge success rate, but there’s always the occasional person who just doesn’t feel comfortable helping.

Since our selection criteria is so tight, finding people is our big problem. Luckily we’ve managed to find a few as quickly as we have. In fact, we’ve recently spoke with someone who fits what we’re looking for and so if all goes according to plan we’ll have two chances at getting some data next week! The first will be with the person who we were going to work with earlier and a new person we just recently found. It’s all very exciting since we don’t normally find people that quickly.

In the four months or so since we’ve undertaken this project, we’ve only found … three?? so far that fit what we need and two of them are the attempts we will do next week, so recruitment for the project is soooooo slow! Ideally we’ll get 10 or so people for the project, the more the better frankly. However, at this rate we may have to settle for a lower number. That’s assuming we can even get the data we want. I’m not worried about the equipment I’ve built for the project, but I am nervous that I haven’t been able to pin down exactly what went wrong with the last experiment.

Since that experiment I’ve tested the equipment with a mock setup and it worked just as I expected it to, so I have no clue what went wrong. I can only guess, but at the end of the day, I know it’s not the equipment, so it was probably something disconnected. At least that’s what I’m hoping. Thankfully everything is ready for next week so it should be a quick experiment assuming that when we connect everything we get something on the recordings.

Basically the moral of the story is that science is hard, sometimes things go wrong and we have no idea why. I’m 98% sure we either connected something wrong or it got disconnected, but whatever the case may be, I’m going to remain just a touch nervous until we get a good dataset collected. I know this will work, I just need time if something goes wrong to track down the problem. Unfortunately time isn’t something we have a lot of when it’s time for the experiment, so we just have to hope I’m right and everything gets connected correctly this time.