Artificial Intelligence Can Change the future of Medical Diagnosis | Shinjini Kundu | TEDxPittsburgh

Share it with your friends Like

Thanks! Share it with your friends!


Medical diagnosis often is based on information visible from medical images. But what if machine learning and artificial intelligence could help our doctors gain more knowledge from medical images and therefore predict diseases earlier? New research indicates this is indeed possible, and the implications for preventive medicine and several disease types are large. Learn about it in this talk.

Shinjini graduated from high school at age 16, earned a PhD at 25, and was named one of Pittsburgh’s 40 under 40 in 2016. In her PhD work at Stanford, Shinjini employed machine learning that trained computers to detect patterns in medical images. That technology enables doctors to detect osteoarthritis three years before symptoms are evident. Shinjini is currently an MD-PhD scientist in the Medical Scientist Training Program (MSTP), a collaboration between the University of Pittsburgh and Carnegie Mellon University. Her research focuses on medical diagnosis in the pre-disease stage when a patient is still asymptomatic. She is also an advocate for women and minorities in STEM professions.

This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at


A O says:

She sounds nervous. Good talk however!

Comic Warriors says:

Awesome speech . Awesome news for the near future

Albert Barrera says:

I want to use the osteoarthritis project in my thesis but cannot find any papers online for me to cite. Can someone direct me?

Tom Ridey says:

Brilliant speech!

omar 1234 says:

This presentation is so biased . using extrems for general datas

Bianca Bucurenci says:

hello! I am a 3rd year student at medical engineering and I've got a question regarding machine learning ang image proccesing. Where sould I start studying from? Please suggest me some sfecific softwares for these.

R Kenne says:

absolutely brilliant, thx for the great work & inspiration

Hunter B says:

this is years of research dont know wy you dont have many likes people just dont know

Eddison Lewis says:


She want the M.D. says:

Christ she sucks at public speaking


I'm generally pretty concerned about the dangers of AI. However, this is the application of AI that I'm looking forward to so much! 🙂

Steven Smith says:

Would be beneficial to just get healthy people. 3 years of normal patient data to get a more precise reading on changes. where would that data come from, well I'm not sure. Just as important as finding out who is sick it's important to find out who is healthy.

Edit: wow I just realized you might be able to determine the rate of lets say tumor growth not just if they sick or healthy.

Write a comment