How AI Changed MRI Forever

Artificial intelligence, MRI machines, and radiation are vital topics that I would love to talk to you about today! We at Skane University Hospital use these topics daily to fight cancer. And how artificial intelligence revolutionized radiotherapy. So I hope you enjoy this video!
Special thanks to:
- The team at Spectronic Medical AB for their work on integrating AI with MRI to fight cancer, and for being so kind to show us around and share their technology and vision with us.
- To my colleague, friend, and the expert on combining AI with MRI to create the MRI-only workflow, Christian Jamtheim Gustafsson. Thank you for joining us to make this video.
If you would like to know more about the research behind the MRI-only topic, here are some links to scientific publications and other useful resources:
- [www.researchga...](www.researchga...)
- [medical.spectr...](medical.spectr...)
- [portal.researc...](portal.researc...)
Thank you for watching and supporting my work!
#MRI #AI #Radiotherapy #Cancer

[@luciususiholo6956]

Wonderful to see developments in medical technology

[@MustiKadhim]

Thanks for watching 🙏

[@totallysawthat]

This goes over the fact that CT exams can be 5-10 minutes, MRI can be 2+ hours.

[@MustiKadhim]

Hi! That's not necessary true anymore, thanks to advanced AI reconstruction methods, we are able to cut these exams down to less then 5 min 😊 Thanks for watching 🙏

[@Unsensitive]

On average an MRI is 20 to 30 minutes. Two plus hours is generally only when multiple body regions are being imaged and usually not required. Scan times can be significantly reduced with AI. The AI is currently trained to identify noise and remove it from the image. It can also be trained to identify patterns of waves in the raw k-space data, and then extrapolate that data to acquire high resolution images with lower resolution data, both speeding up acquisition. Even early into AI, we're seeing upwards of 75%+ scan time reductions for many applications. It will only get better This can be combined with other acceleration techniques and a 20-30 minute MRI from 5 years ago taking 5-minutes is not unheard of.

[@christianjamtheimgustafsso4625]

Great work Musti!!

[@MustiKadhim]

Thank you for being so kind to join us on this video Christian! 🎉

[@actualsnow345]

Will MRI techs still be needed in the future?

[@MustiKadhim]

Hi! Absolutely! It has superior image contrast and detail quality to anything we currently have as imaging tools 😊

[@nouredinebraik3826]

Excellent content 👏

[@MustiKadhim]

Thanks for watching 🙏

[@michaelbacchiocchi8111]

I’m curious as to what data sets can be shared given privacy laws within the health industry. I assume this also varies between countries. Can you elaborate?

[@MustiKadhim]

Hi Michael! Thanks for this interesting question. In Sweden it is extremely hard to share patient data between hospitals, let alone with the industry. However this company AI tool was developed during a major collaboration project where the data could be used to train AI models to reach optimal performance and then deploy the models in healthcare. Hope this answers your question!

[@SRose-vp6ew]

All identifying information is removed from the MRI images before you’re able to use them for research. It depends on what the IRB says.

[@SRose-vp6ew]

*The Institutional review board. There are many variables, not just by county, but by the type of study in the county.

[@comandante9312]

great work. is that company has shares in stock markets?

[@MustiKadhim]

Thanks a lot! no not yet, but maybe in the future they will! Thanks for watching 🙏

[@comandante9312]

@@MustiKadhim thank you for your information.

[@baris8415]

Siemens Healthineers does the same thing

[@viktor4872]

🙏