Siraj Raval
Can AI be used to detect various diseases from a simple body scan? Yes! Normally, doctors train for years to do this and the error rate is still relatively high. From mammograms to cat scans, AI can diagnose a disease better than any human can if given the right training dataset. This will drastically reduce patient death, save medical practices a lot of money, and aid doctors in the patient care process. Everyone will win and its important to remember that AI won’t replace doctors, it will become the most powerful tool they’ve ever used. And once enough AI startups start impacting the field of healthcare, it will become as common a tool as the stethoscope has been.
Code for this video:
https://github.com/llSourcell/AI_in_Medicine_Clinical_Imaging_Classification
Please Subscribe! And like. And comment. That’s what keeps me going.
Want more education? Connect with me here:
Twitter: https://twitter.com/sirajraval
Facebook: https://www.facebook.com/sirajology
instagram: https://www.instagram.com/sirajraval
Curriculum:
https://github.com/llSourcell/AI_For_Business_Curriculum
More learning resources:
https://www.youtube.com/watch?v=3LkbUxqGTfo
https://www.youtube.com/watch?v=S4GvBCMfRew
https://www.youtube.com/watch?v=LxHHsujnF9c
https://www.youtube.com/watch?v=ZPXCF5e1_HI
https://www.youtube.com/watch?v=QfNvhPx5Px8&t=202s
Join us in the Wizards Slack channel:
http://wizards.herokuapp.com/
Sign up for the next course at The School of AI:
https://www.theschool.ai
https://github.com/gregwchase/dsi-capstone
And please support me on Patreon:
https://www.patreon.com/user?u=3191693
Signup for my newsletter for exciting updates in the field of AI:
https://goo.gl/FZzJ5w
Source
How do you verify that your training data is accuratet? How do you know you're not training your AI to misdiagnose?
The course literally started at 5:04 and ended at 10:07 ….Damn!!
You teach more than a semester course in 5 minutes!
now I watch most of educational videos at 2X speed and your videos at 3X. Thanks Siraj
Can u do a video on neural mass modeling
hi, guy. I am from China.I want to know how to begin with AI learning, I want to consult you, could you give me you contact methods,so i can commute to you?
https://youtu.be/byQtPUOXzu4
Siraj , U R the real hero sharing expensive knowledge without expecting anything , Great 🙂
The best tutor and most in-depth videos on YouTube by far. Subscribed!
Thank yo Siraj I have been watching your videos and learning so much. I made my own here using your course and lessons. https://biovision.ai
Guys I need your help. I really want to play with the diabetic retinopathy (DR) dataset but it seems I can't download it on kaggle. Does anyone else have the same problem? Where can I get those datasets?
Only respect for this guy.
you are my hero siraj
I'm a radiologist working on AI/ML projects. I agree that AI/ML has a tremendous potential to HELP doctors including radiologists especially by removing the mundane repetitive work so they can focus more on quality rather than quantity. However, it also has an equal opportunity of being used for cost cutting for huge mega corps and insurance companies.
The current trend is "AI better than Doctor/human X" which I believe is an improper metric designed to commercialize AI/ML. A better more accurate metric would be AI/ML helps doctor X save more lives. "AI/ML supervised by doctors or AI/ML supporting doctors in their practice" is the perfect set up but if we're not careful, it might turn into Machines Replacing Doctors everywhere because they're "cheaper for mass production, faster, don't complain, and are more readily replaced when obsolete." AI/ML is currently learning only with "correlation" and still cannot compete with the human brain's capacity for "cause-and-effect" which is a key differentiating factor between man and machine (but will businesses care about that?).
I hope that more doctors become engaged in the #democratizeAI movement and your #100DaysOfMlCode.
btw, love your videos Siraj. Keep learning a lot from them. I usually only watch and like but had to speak up about this since medicine and Radiology are near and dear to me. I really hope AI/ML is used for the betterment of health care, and not increased economic/production output.
Is computer science a good major for medical informatics?
You're making this world a better place dude.
Sorry but anybody know what is the website metioned 7:43 ? Thank you very much.
Do you have a suggestion on what gpu is best for ML
I wish you a billion dollars, Siraj. So much great content…
Big love from Paris <3
may be its best for radiologist only
You are a good person
I'm a BME Undergrad and I'm gonna be attempting to build a classifier this weekend at Shellhacks in Miami. If anybody's got medical datasets on anything please message me.
you're the Kanye West of eLearning.. fascinating content.. keep up the good work Siraj!
Where is the tutorial boss
Why Floydhub, not another famous cloud?
Mr Siraj Raval Great video. I am from Diabetic Hospital in India n a WHO collaborating Centre in India. Wld like to implement Google Tensor flow for Diabetic Retinopathy.
Can u help us
My email I'd is pk.sundararaman@gmail.com
Thanks n regds
Kumar
Siraj you are simply love. 🙂