[ohhereIamagain]

I'm a recent medicine graduate working as a junior Dr.  I'm also very interested in somehow pursuing AI and combining it with medicine. Maybe pursuing a degree then starting a niche company? I'm really not too sure

[keghn]

 Hello @ohhereIamagain.

 Right now i see two paths in medicine right now. The first is using Deep neural network to detect microbes and disease tissue on a microscopic level.
 The third is Big data. That is having AI going through a lot of data on patients and finding a pattern.

[ohhereIamagain]

Hello @ohhereIamagain.

 Right now i see two paths in medicine right now. The first is using Deep neural network to detect microbes and disease tissue on a microscopic level.
 The third is Big data. That is having AI going through a lot of data on patients and finding pattern.

Hey keghn,

Thanks for such a quick (and thoughtful) reply!

I'm very interested in anesthetics and risk stratification, both areas that I think neural networks could have a large impact on.

Do you have any ideas on how I could best pursue AI? I understand that developing a deep and professional level knowledge of if likely takes many years, but I was thinking about taking a 1 year break from work and focusing full time on learning as much as I can. Then on returning to work trying to recruit professionals in the area to begin a company or projects, that way I have an understanding of the medical side, and at least a brief understanding of the AI side 

[kei10]

Hello, ohhereIamagain.

There are tons of information and media on the web that helps you to understand the basis of deep neural network and their application, you could try to start there.

This forum welcomes questions from you to be answered by a few named polished creators for you to further pursue the knowledge of AI, or you could try other places.

That being said, the fundamentally approach is to encourage, develop, and test AI programs by oneself while keeping oneself steady along the path.

[Korrelan]

Hi Ohiaa

I know I’m probably stating the obvious but…

If I were in your situation the first thing would be to learn a little about the many different approaches to AI, heuristics, genetic algorithms, convolution nets, liquid/ reservoir computing, connectionist, etc.

Each has its own advantages depending on the type of problem you are ultimately trying to solve; this would give you a better idea of where to concentrate your limited time.

There a lots of pre-built solutions to data quantifying problems like risk stratification; just depends on at what level you want to enter the arena.

 

[keghn]

 Anesthetics......................
 With this subject being brought up. I am very interested in having brain way being read and have artificial neural network
detect length of brain wave patterns.
 Once a certain brain wave pattern is detected a auto dispenser pumps chemicals into the patient. But I am interested
in improving psychological state of a person in terms of the field of sports, cognitive, psychology, and meditation.
  I am keeping this as a very minor project until the day when cheap light weight brain reader that plug into a cell phone
with the proper software become available.

[8pla.net]

In EKGs, machine diagnosis is called "interpretation".  Some EKG machines come with it, some don't.  The one's that don't are interpreted by human experts instead.

However you can not combine a career in medicine with A.I. unless things have changed (which I doubt they have).  Only a licensed physician can make a diagnosis, (at least as far as I know).  Machines are not allowed to, no matter how excellent they are.   IBM Watson can only suggest treatment options to physicians which they may have missed.

Here, I am simply thinking out loud for the purpose of our discussion.  Perhaps I am incorrect.  For example, regulations may apply differently where you are.  So, please tell me, what do you think?