brandonb

39 ms·

841.
by brandonb 10y ago
Here are a few efforts from the Y Combinator and broader tech communities that help with those goals. On veterans, several YC alumni, after working on the healthcare.gov rescue team, went on to help start the US Digital Service ( https:&#x2
842.
by brandonb 10y ago
True -- the available data sets (like MIT-BIH arrhythmia database) are quite small. In our case, we launched a study with UCSF Cardiology which has recruited about 10,000 people so far, and that's how we get a data set large enough for
843.
by brandonb 10y ago
The app is free to you, but if we can detect (say) diabetes early and route you to the right medical care, we save the healthcare system money. We've built a platform called Cardiogram Connect which optionally lets you share aggregated
844.
by brandonb 10y ago
Not for now unfortunately--but maybe once we get a little larger.
845.
by brandonb 10y ago
If you're a strong software engineer but without any particular expertise in deep learning, feel free to email me as well. brandon@cardiogr.am :) We (Cardiogram) in particular are hiring for: mobile engineers (building an Apple Watch a
846.
by brandonb 10y ago
I wrote a bit about the systematic issues preventing promising AI research from moving to production in medicine, including a very brief history of MYCIN: https://blog.cardiogr.am/three-challenges-for-artificial-int... I th
847.
by brandonb 10y ago
This is the second major study applying deep learning to medicine, after Google Brain's paper in JAMA in December, and there are several more in the pipeline. If you've developed expertise in deep learning and want to apply your s
848.
Dermatologist-level classification of skin cancer with deep neural networks(nature.com)
2 points by brandonb 10y ago|0 comments
849.
by brandonb 10y ago
Sure, my email is brandon@cardiogr.am.
850.
by brandonb 10y ago
The NIPS workshop was Machine Learning for Health. The paper isn't out as a PDF since a more mature version of the work will be submitted to a medical journal: http://nipsml4hc.ws
851.
by brandonb 10y ago
Not sure if this is what you're looking for, but Andrew Ng touched on a lot of pragmatic parts of software engineering in machine learning teams at Deep Learning School: https://www.youtube.com/watch?v=F1ka6a13S9I&t
852.
by brandonb 10y ago
The math is more approachable than it may seem. My recommendation is to read textbooks (not online tutorials or crash courses), which will guide you through the mathematical concepts and notation. Each little bit builds on the previous one,
853.
by brandonb 10y ago
Yes! Deep learning is a young field. I have a B.S., for example, and have published work at Interspeech (speech recognition conference) and NIPS workshop (machine learning conference). You can too. My advice: * 80% of machine learning i
854.
Judge Blocks Aetna’s $37B Deal for Humana(nytimes.com)
1 points by brandonb 10y ago|0 comments
855.
by brandonb 10y ago
I can't edit my post anymore, but as of right now, the link works over http rather than https: http://www.nipsml4hc.ws/
856.
by brandonb 10y ago
For those interested in AI in medicine, the most recent NIPS held a workshop on exactly that topic: https://nipsml4hc.ws There's quite a bit of active research applying LSTMs, CNNs, and the like to medical data sets. I&#x27
857.
Predicting Medical AI in 2017(lukeoakdenrayner.wordpress.com)
77 points by brandonb 10y ago|10 comments
858.
by brandonb 10y ago
Just wanted to say: thank you for your service!
859.
You’ll Never Be the Same Again (US Digital Service)(medium.com)
2 points by brandonb 10y ago|0 comments
860.
by brandonb 10y ago
We do plan to expose predictions for you within Cardiogram. The biggest barrier is that any predictions relating to a health condition (like atrial fibrillation) are FDA-regulated, so there's a high level of scientific validation that
861.
by brandonb 10y ago
Startups often have health data and an interest in machine learning. For example, we presented research at a NIPS workshop where we trained an LSTM to predict abnormal heart rhythms from about 793 million heart rate measurements. The heart
862.
Proceedings of the First Machine Learning for Healthcare Conference(jmlr.org)
4 points by brandonb 10y ago|0 comments
863.
Data from wearables can predict disease(theverge.com)
1 points by brandonb 10y ago|0 comments
864.
The President’s Final Ask of Us (Code for America)(medium.com)
2 points by brandonb 10y ago|0 comments
865.
Smartwatches could soon tell you when you’re getting sick(techcrunch.com)
3 points by brandonb 10y ago|0 comments
866.
by brandonb 10y ago
Interesting idea! I think this is analogous to the idea of a de-noising autoencoder in computer vision. Here, instead of introducing Gaussian noise at the pixel level and using a CNN, you're introducing grammatical "noise" at
867.
Alcohol Abuse Increases Risk of Heart Attack, Atrial Fibrillation, and HF(cnn.com)
1 points by brandonb 10y ago|0 comments
868.
by brandonb 10y ago
Well, I guess my sincerity detector is officially broken, huh?
869.
by brandonb 10y ago
This seems like a sincere comment, so in case it's helpful, here's my interpretation of why others may be downvoting it. Deep learning practitioners are aware that they're using non-convex optimization, stochastic gradient de
870.
Apple Watch, New Year’s resolutions, and losing 50 pounds(9to5mac.com)
2 points by brandonb 10y ago|0 comments

More ›