Meena Mani

@meena_uvaca

medical imaging | start-ups | data science | machine learning | deep learning

meenamani.org
Vrijeme pridruživanja: prosinac 2013.
121 fotografija ili videozapis Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @meena_uvaca

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @meena_uvaca

  1. Prikvačeni tweet
    12. ruj 2015.

    My new article on the tumor resection brain surgery I observed:

    1 reply 2 proslijeđena tweeta 7 korisnika označava da im se sviđa
    Poništi
  2. 13. sij

    A great story of reverse engineering the Silicon Valley interview process -- a process that I might add is random, sucky, unfair and full of humiliation.

    On my new year flight from Lima to SF, I wrote a blog post on the path from my previous job to the current one. TL;DR => it was an ocean of pain and experience obtained at was very useful 😀 If you like it => 50 claps. If not => 49 😀
    1 korisnik označava da mu se sviđa
    Poništi
  3. 25. pro 2019.

    This is another* nice review of the latest trends at . (*For those who have not already read 's excellent review, here's the link: )

    With in the rearview mirror, what were the main trends worth noting from this year’s show? Sectra’s shares his thoughts:
    1 proslijeđeni tweet 3 korisnika označavaju da im se sviđa
    Poništi
  4. 16. pro 2019.

    Presentation slides from the Dec 14 Med-Neurips 2019 event.

    Paper presentations and keynotes now available on-demand: Session1 : Session2 : Session3 : Session4 :
    1 proslijeđeni tweet 2 korisnika označavaju da im se sviđa
    Poništi
  5. 14. stu 2019.

    The RSNA ICH competition top solutions (1,2,3, 6,16 etc.) were interesting because (i) they incorporated radiologist-style windowing as a preprocess step: and (ii) sequence models or some arrangement that stacked consecutive 3D slides

    Congrats to SeuTao team winning (preliminary) Intracranial Hemorrhage Detection competition on This computer vision competition was won partially with a sequence model! Looking forward to see this innovative and possibly reusable solution:
    4 korisnika označavaju da im se sviđa
    Poništi
  6. 8. stu 2019.

    Helpful guide for Deep Learning on high resolution histopathology images including choice of network, ways to aggregate patches, dealing with white spaces etc. Classification of Histopathology Images with Deep Learning: A Practical Guide by Jason Wei

    9 proslijeđenih tweetova 31 korisnik označava da mu se sviđa
    Poništi
  7. 7. stu 2019.

    Machine Learning in Healthcare (MLHC 2019) talks on YouTube: Found Dr Rajan Dewar's talk on Cost Effective Healthcare Solutions in Low Income Countries interesting.

    3 korisnika označavaju da im se sviđa
    Poništi
  8. 2. stu 2019.

    A review of UNet++ — A Nested U-Net Architecture by Sik-Ho Tsang. The skip pathways in the UNet++ have convolutional layers and DenseNet-style connections.

    2 proslijeđena tweeta 4 korisnika označavaju da im se sviđa
    Poništi
  9. 2. stu 2019.

    There is more than one target market for radiology AI products. For products designed for low resource communities, for non-profits, to increase accessibility, different engagement/sales models are needed along with metrics to measure success.

    Odgovor korisnicima @curtlanglotz @mattlungrenMD
    I think more companies are learning that AI doesn’t necessarily translate to $. You need much more than just strong predictions, like innovations in the engagement/business/sales model.
    Poništi
  10. proslijedio/la je Tweet
    1. stu 2019.

    Most of what we do has no proven downstream outcome effects. IMHO the main reasons the radiology AI companies have not met investor expectations are: 1. high quality clinical training data is a lot harder to come by than radiology AI startups anticipated.

    Odgovor korisniku/ci @curtlanglotz
    The bubble will pop - and the market will consolidate. There's a lot of talent out there, and the will to drive AI in radiology forward, but the question of payment still remains in the face of a lack of solid prospective evidence for downstream clinical outcome improvement.
    11 replies 80 proslijeđenih tweetova 187 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    29. lis 2019.

    I've had more MRIs that I can remember, mostly because of a bad hip joint. That's one of the reasons I'm so interested in this fascinating project from and -- their using AI to try to speed these scans up. Here's my inside look:

    3 proslijeđena tweeta 2 korisnika označavaju da im se sviđa
    Poništi
  12. proslijedio/la je Tweet
    30. lis 2019.

    This is where machine vision during colonoscopy is headed: not just helping to pick up flat, sessile, missed lesions, but also to real-time classify regarding via H/T

    37 replies 500 proslijeđenih tweetova 1.310 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  13. proslijedio/la je Tweet
    29. lis 2019.

    Exciting news! We received FDA 510(k) Clearance for our One Click™ Cardiac MRI Package, the First AI-assisted Cardiac MRI Scan Solution! To find out what this means for cardiologists, patients & the wider healthcare industry, see the full release here:

    7 proslijeđenih tweetova 9 korisnika označava da im se sviđa
    Poništi
  14. proslijedio/la je Tweet
    3. lis 2019.

    Clinical has always moved towards higher magnetic fields, but compelling work showing that low fields may have role as well. I'm partial to the lung application, and also can't stop watching the real-time speech imaging

    1 reply 9 proslijeđenih tweetova 17 korisnika označava da im se sviđa
    Poništi
  15. 28. lis 2019.

    NVIDIA and King’s College London Debut First Privacy-Preserving Federated Learning System for Medical Imaging

    1 proslijeđeni tweet 2 korisnika označavaju da im se sviđa
    Poništi
  16. 28. lis 2019.

    An new Indian brain atlas IBA100 based on 100 human brains. Other population specific Asian atlases are the Chinese2020 and the smaller Korean and Japanese atlases.

    Check out the work we did when I was a master's student at . We constructed first ever structural brain atlas for young Indian population. Related paper:
    Poništi
  17. 26. lis 2019.

    Fun thread

    The pointless point of care MRI
    Prikaži ovu nit
    Poništi
  18. 26. lis 2019.

    The first low cost, low power portable MRI. There can be so many use cases, some outlined in the article below, once such a device is made available. I see this as having huge potential in a country like India. Here's a YouTube video:

    that can come to patients anywhere, anytime, Hyperfine Point-of-Care MRI
    2 proslijeđena tweeta 4 korisnika označavaju da im se sviđa
    Poništi
  19. proslijedio/la je Tweet
    24. lis 2019.

    Come see more eye candy at booth #251 at ACEP 2019 Denver, October 27-29! Hyperfine has one foot in the past and one in the future. These images were taken from a scanner the size of a mini-fridge. It plugs into a standard wall outlet and moves to the bedside.

    1 reply 11 proslijeđenih tweetova 39 korisnika označava da im se sviđa
    Poništi
  20. 24. lis 2019.

    The Data Science Bowl competitions have focused on biomedical + imaging in the past and we've also had 2 recent competitions (APTOS + RSNA challenge) to work with. Still was hoping for another big competition.

    Our 5th annual Data Science Bowl competition is now live! 
Help educators around the globe better understand how kids learn. 
Read the details and get started today:
    1 proslijeđeni tweet
    Poništi
  21. proslijedio/la je Tweet
    23. lis 2019.

    Our preprint is out. With test time augmentation and deep learning we could further improve single cell segmentation. Reached an ~2% further gain on the kaggle Data Science Bowl.

    1 reply 10 proslijeđenih tweetova 23 korisnika označavaju da im se sviđa
    Poništi

@meena_uvaca još nije objavio nijedan Tweet.

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·