Prototype: Detecting Stress in Patients

Thread Starter

CompSciSarah

Joined Feb 22, 2020
4
I'm working with one of my professors at Stony Brook University researching biomedical applications.

Our group was interested in developing an application for stress detection in patients in regions where medical infrastructure might not be the best if it exists at all.

We also wanted it to be used for triage situations.

What sensors would be best for this task? I was thinking of a pulse rate sensor for monitoring pulse rate and thermometer.

What other suggestions would you make, and what sensors would be best for accomplishing this goal?
 

djsfantasi

Joined Apr 11, 2010
9,163
No, I haven't, but I will for sure. Any ides for measuring it using sensors though?
If you haven’t searched for the established criteria for stress, how can you possibly understand what sensors you need? A sensor measures something and you don’t know what to measure. Sounds like the start of a fail.
 

Thread Starter

CompSciSarah

Joined Feb 22, 2020
4
If you haven’t searched for the established criteria for stress, how can you possibly understand what sensors you need? A sensor measures something and you don’t know what to measure. Sounds like the start of a fail.
I don't see how, I was just asking if there were established guidelines that were common and what sensors would be used.

Of course I could refine my search and derive the answer by myself without any input whatsoever from anyone else, but that kind of defeats the purpose of asking others who might be doing the same work then doesn't it?

My question essentially boils down to: What would be the best criteria for measuring stress in a way that would be implementable.

I see nothing wrong with that request.
 

jpanhalt

Joined Jan 18, 2008
11,087
Look up "lie detectors" (aka polygraph tests). Stress can be exhibited in a number of ways. Based on what you consider important, you may be able to find a detector. For example, if you consider perspiration (sweat) to be important, there are detectors for that.

Professor Bruce Rabin used other measures too. You need to so some background reading.

Put another way, I can't imagine anyone, much less a professor, starting or allowing someone else to start a project involving humans or any animal without doing background reading first.
 

Thread Starter

CompSciSarah

Joined Feb 22, 2020
4
Look up "lie detectors" (aka polygraph tests). Stress can be exhibited in a number of ways. Based on what you consider important, you may be able to find a detector. For example, if you consider perspiration (sweat) to be important, there are detectors for that.

Professor Bruce Rabin used other measures too. You need to so some background reading.

Put another way, I can't imagine anyone, much less a professor, starting or allowing someone else to start a project involving humans or any animal without doing background reading first.
Unfortunately, my professor is very pressed for time and is working on many different projects simultaneously, so most of the research falls to me.

I guess I'll hit the books then. Thanks for the insight.
Might post back later when I get the background information. The polygraph suggestion was great
 

jpanhalt

Joined Jan 18, 2008
11,087
My memories are faded by time, but my professors either gave you a problem or you proposed your own problem. Many of us proposed out own. We were expected to research it and come up with a concrete proposal, including review of literature, and then defend that proposal. The professors had very little input, except to review the proposal before presenting it for defense. We had to do that formally twice on two different projects and informally for virtually every tangent our thesis project took.

Your project has an added twist. Research ethics around the world require additional review of any research involving humans (and animals). In the United States, that review is done by an Institutional Review Board (IRB). Members on an IRB must include laypersons not affiliated with the institution where the work is being done and typically also include clergy from the community. Some people think that research involving humans means doing experiments on them. That is not the case. Review of a patient's chart that is not for the care of that patient can be research. Certainly, review of several patients' charts requires institutional review.

There are some very limited exceptions to the requirement for review, but from what you have said, your experiments would require review. (There are exceptions for experimentation on oneself, but that can be a tricky road.) Besides the need for you to review the current state, an IRB will require an assessment of risk and benefits to the subjects and to the public as a while (i.e., is the research even worth doing even if it is minimal risk).
 
Top