Profile cover photo
Profile photo
monkut h
'We are all here on earth to help others; what on earth the others are here for, I don’t know' -- John Foster Hall
'We are all here on earth to help others; what on earth the others are here for, I don’t know' -- John Foster Hall
About
monkut's interests
View all
monkut's posts

Post has attachment

Post has attachment

Post has shared content
Public
I'm very excited about the work our group (g.co/brain) is doing in various areas of medical imaging. Today, we published a preprint of a paper titled "Detecting Cancer Metastases on Gigapixel Pathology Images" by Yun Liu, Krishna Gadepalli, Mohammad Norouzi, George E. Dahl, Timo Kohlberger, Aleksey Boyko, Subhashini Venugopalan, Aleksei Timofeev, Philip Q. Nelson, Greg Corrado, Jason D. Hipp, Lily Peng, and Martin C. Stumpe.

Some key statistics from the paper used to evaluate the effectiveness of this work:

Tumor localization score (FROC): ("find all the tumors")
0.89 (our model)
0.73 (human pathologists with infinite time)

Sensitivity at 8 false positives:
0.92 (our model)
0.73 (human pathologists with infinite time)

(It's also worth point out that Yun Liu is a member of the Google Brain Residency program: g.co/brainresidency)

This work follows on our earlier work on detection of diabetic retinopathy in retinal images.

Pathology blog post: https://research.googleblog.com/2017/03/assisting-pathologists-in-detecting.html
Paper preprint:
https://drive.google.com/file/d/0B1T58bZ5vYa-ZjlSRmIyc3M3MTQ/view

Earlier blog post on diabetic retinopathy work: https://research.googleblog.com/2016/11/deep-learning-for-detection-of-diabetic.html

Edit: update paper preprint link to different URL (but same content)


Post has shared content
Public
Daniel Bogan runs a site called usesthis.com that does interviews with different people from all kinds of professions about their "work setup" (everything from computer scientists to chefs to beekeepers). He asked me if I'd participate, and here's the result:

https://usesthis.com/interviews/jeff.dean/

Browing the other interviews is kind of fun just to hear about what tools are useful for other professions:

https://usesthis.com/interviews/
https://usesthis.com/categories/

Thanks for running this site, Daniel!

Post has shared content
Public
MR STYLY
Preview of upcoming Mixed Reality fashion project by Psychic VR Lab presents their digital clothing designs onto real world mannequins using the Microsoft Hololens platform.
http://prostheticknowledge.tumblr.com/post/157792586526/mr-styly-preview-of-upcoming-mixed-reality-fashion
Animated Photo

Post has shared content
Public
Catch up with our cloud team in the January edition of our monthly newsletter. Find out about #MWC17, get links to tutorials and resources, and check out our upcoming events and webinars!

Post has shared content
Public
Last week, we launched a new feature for Google Maps for Android across 25 US cities that offers predictions about parking difficulty close to your destination so you can plan accordingly. Today, we talk about the machine learning + crowdsourcing techniques that makes this feature possible. Learn more on the Google Research blog, below.

Post has shared content
Public
Raspberry SUPER Pi case #piday #raspberrypi @Raspberry_Pi
https://blog.adafruit.com/2017/02/03/raspberry-super-pi-case-piday-raspberrypi-raspberry_pi/

Via designsome

…I wanted a user to be able to simply plug and play with the case without requiring any desoldering/soldering of components.

Parts were designed and modeled in CAD and FDM 3D printed. The parts required sanding due to the layer thickness of the 3D print.

See more
https://blog.adafruit.com/2017/02/03/raspberry-super-pi-case-piday-raspberrypi-raspberry_pi/
Photo

Post has shared content
Public
PaintsChainer
Online app from @tai2an can colourize line drawings using neural network training. Upload an image and place some points of colour and away your go.
http://prostheticknowledge.tumblr.com/post/156763302411/paintschainer-online-app-from-tai2an-can
Animated Photo

Post has shared content
Public
Text Classification using Neural Networks with python nltk and numpy
Wait while more posts are being loaded