Dear friends,
It's been 5 years I haven't logged in to this forum. I am now 25, I still have some acne, but I can say it is under control and not impacting my social life anymore. Like many others on this forum I achieved it by experimenting on myself. I have also tried several drug products, but I have found all my solutions in the holistic approach: mostly with food, sleep, sport, psyche.
I believe in the holistic approach, I believe it is underrated in the medical community, It lacks scientific support surely due to lack of studies. Holistic studies are more difficult because they should take into account many variables to be significant.
How many times have I been reading or hearing that food does not affect one's acne, or that drugs are the first way to go, when I know for a fact now that in my case they were wrong. I am sure some of you had, have or will have the same experience.
Since my last login I graduated in chemistry and business studies, and I have been working at a major pharmaceutical company for one year working primarily on statistics.
Today I come back to the forum to write to you because I have an idea and I want to hear what you think about it.
I want to research holistics and acne. I want to give everybody a tool to find out what aspect of their life is positively or negatively their acne, whether it is one of the foods and which one, sleep hours, stress level, sun exposure, sexual activity, sport activity, etc...
My idea is to create an app (Android, iPhone) where participants would fill out a form every day with all the necessary variables.
I will then define and setup all possible statistical experiments to give the result to the user in form of a simple report (and an optional detailed one).
This way, in addition to giving personalized and hopefully significant results to every participant, I will hopefully be able to use the entire dataset to finally deliver a very valuable study (or studies) about acne and all the different aspects from our everyday life.
A fundamental requirement is that I find participants who are willing to invest ca. 5 to 10 minutes per day on this app, for themselves and for this cause.
During university years I have oftentimes done something similar for myself using a piece of paper, writing down what i had eaten and what the status of my acne was everyday to then look for correlations.
With this universal app I would take care of all the statistics, which would deliver more consistent, and valuable results for everyone and for science.
What do you guys think, would you be interested in using this app?
A warm greeting to everyone here
Davide
Since you work for a pharmaceutical company, why don't you ask the researchers that test their drugs on mice how the mice get the disease in the first place since they live in a controlled environment?
I believe we all know that the majority of medications either suppress the immune system or a organs function. That's how they work.
I'm not being a smart ass, just stating the truth. By the way, I think you have a good idea.
Hello, good idea.
I was planning to do something similar but with a website rather than a smartphone application (which would still be accessible to smartphones with a browser)
I was also planning to add a face logger (take a picture of yourself with the same kind of lighting almost everyday). This was to see if there is any link with pimples and the other parameters (food, sleep, sun, stress, ...)
I already made a working prototype using OpenCV and Qt (in C++). But it's not for public use yet (too much hardcoded stuff, not user-friendly UI)
So, for now it detects the eyes and rotate/crop the image so that all your pictures are similar and easy to compare (you can check my log topic if you wanna see some real example)
The next step would be to either detect automatically the pimples severity/count (would be more difficult), or the tool could ask the user to locate himself the pimples and judge their severity.
I put this whole thing on hold because I'm not actually sure there is such a link (some people say pimple can appear overnight because of some food, and other people will say acne can take up to 2-3 months for the pimple to show (the bastard would stay hidden all this time, so deducing anything with the other parameters from the face is mission impossible in that scenario)).
Anyway, tell me if you'd be interested to cooperate on this!
BTW, have you heard about ScanZ ? http://techcrunch.com/2013/10/28/myskin-launches-scanz/
(I'm not affiliated, it's just something kind-of similar-ish and worth checking out before working on something like this imo).
Since you work for a pharmaceutical company, why don't you ask the researchers that test their drugs on mice how the mice get the disease in the first place since they live in a controlled environment?
I believe we all know that the majority of medications either suppress the immune system or a organs function. That's how they work.
I'm not being a smart ass, just stating the truth. By the way, I think you have a good idea.
I don't work at that company anymore, I am going rogue 
Regarding mices, unfortunately I was never in clinical research, so I don't know. I know mices have a similar genetic to ours and that drugs are tested for safety, I am not sure mices get infected first.
Anyway thanks for your feedback.
Hello, good idea.
I was planning to do something similar but with a website rather than a smartphone application (which would still be accessible to smartphones with a browser)
I was also planning to add a face logger (take a picture of yourself with the same kind of lighting almost everyday). This was to see if there is any link with pimples and the other parameters (food, sleep, sun, stress, ...)
I already made a working prototype using OpenCV and Qt (in C++). But it's not for public use yet (too much hardcoded stuff, not user-friendly UI)
So, for now it detects the eyes and rotate/crop the image so that all your pictures are similar and easy to compare (you can check my log topic if you wanna see some real example)
The next step would be to either detect automatically the pimples severity/count (would be more difficult), or the tool could ask the user to locate himself the pimples and judge their severity.
I put this whole thing on hold because I'm not actually sure there is such a link (some people say pimple can appear overnight because of some food, and other people will say acne can take up to 2-3 months for the pimple to show (the bastard would stay hidden all this time, so deducing anything with the other parameters from the face is mission impossible in that scenario)).
Anyway, tell me if you'd be interested to cooperate on this!
BTW, have you heard about ScanZ ? http://techcrunch.com/2013/10/28/myskin-launches-scanz/
(I'm not affiliated, it's just something kind-of similar-ish and worth checking out before working on something like this imo).
Hi Anonyshamed,
thank you very much for the precious feedback, I checked out SCANZ, I can clearly see their approach, which is interesting but also different to mine. Mine is a preventive one, just to get to know yourself. In my opinion everyone who believes in holism should back their experiences with statistics.
I also would not go with any image recognition. I think for starters it is good enough to count pustules, papules, nodules, cysts. This we would be our "output variables", and since they require a classification I can take them into consideration to simply define a improvement or worsening.
Anyway I was able to control my acne with holistic, this is why I want to give everyone a tool to do it themselves, I don't know if it will work for others, what I know is that this way we will for sure learn much more about acne and different foods, acne and sleep, acne and sports and much more. So, like you I am also not a 100% sure this can work for everyone, this is why I do it, to test it.
I would go for an app, not only desktop, since it requires daily dedication it should be as practical and easy as possible for everyone participating.
I also would not go with any image recognition. I think for starters it is good enough to count pustules, papules, nodules, cysts. This we would be our "output variables", and since they require a classification I can take them into consideration to simply define a improvement or worsening.
I think you'd miss on a lot of variables not using image recognition.
- Position. I think most people would agree position has some meaning (hormonal is more around jawline and lips). [well I guess for this one you could ask for the count for each region]
- Timing. I noticed using my program that I can more easily see when pimples started (which is an interesting piece of info imo) whereas if I was only asked a number, I wouldn't have counted it because at the time it was too small or hardly visible.
- Duration. I don't know if it's specific to me or not, but I noticed some last way longer than others (not sure why).
I'm not saying you need it for the first version, but as someone personally suffering from acne, I would prefer using an app that would try to use the maximum amount of data possible.
I also would not go with any image recognition. I think for starters it is good enough to count pustules, papules, nodules, cysts. This we would be our "output variables", and since they require a classification I can take them into consideration to simply define a improvement or worsening.
I think you'd miss on a lot of variables not using image recognition.
- Position. I think most people would agree position has some meaning (hormonal is more around jawline and lips). [well I guess for this one you could ask for the count for each region]
- Timing. I noticed using my program that I can more easily see when pimples started (which is an interesting piece of info imo) whereas if I was only asked a number, I wouldn't have counted it because at the time it was too small or hardly visible.
- Duration. I don't know if it's specific to me or not, but I noticed some last way longer than others (not sure why).
I'm not saying you need it for the first version, but as someone personally suffering from acne, I would prefer using an app that would try to use the maximum amount of data possible.
I agree with you if it is a properly made image recognition it would be useful indeed, but to begin with a form should do it. I personally agree position is important (especially the more hormonal acne) but perhaps the position can be taken into account also in a form with the count. Also duration and timing are valuable (I also don't know yet all the variables i want to use in a first version, this comes later).
As you say, if I can get the message across that more variables equal a clearer result than I think it could work, me too, like you, I would not care about if I have to spend 5 to 10 minutes a day to really find out what is causing my acne.
The goal now is to find out if other people would be interested in using such an app.
It's a good idea, but aren't there too many variables to keep track of?
There are many variables yes. If I can get a large number of users, than the variables can be distributed among users. I think I should leave this (which variables to keep track of) decision up to the user.
There already is an app for looking up infrmation. Combining with a food journal would be good.
Pimple counts will cause a lot of anxiety for the many people who can't seem to learn to eat right without developing an eating disorder r stressing over every little thing.
Hey Alternativista, I do not understand your last statement. I understand pimple count can cause anxiety, so I think I will let the user decide what he wants to input, Already having subjective information if there was an improvement or worsening with respect to the previous day (instead of the count) is valuable for my statistics.