ML I assumed was machine learning.
Yes I know they used a ball lens, I still think "garbage in garbage out" applies. They said hemotology not cell counts, so I assume you need better data, but I could be wrong.
On Aug 4, 2014 11:21 AM, "John Griessen" <john@industromatic.com> wrote:
-- On 08/04/2014 11:46 AM, Nathan McCorkle wrote:
The images in the first link you posted are quite crappy IMO.
Yes, but...
They say, "The attachment magnifies/focuses on the sample by means of a 1mm ball lens."
That's sounding like Van Leeuwenhoek's early microscope lenses, and yes the images look
very so so, but maybe the imaging software can do an edge detect and lose all the out-of-focus-blur
and analyze what's left... Instead of a perfect image, it may be more important to
scan along the whole area under consideration, then change depth of focus, rescan,
repeat-while-necessary to get a 3D map of images, then recombine some of them if needed
for a positive ID of what is floating at that spot. If you're talking about doing White cell count
and red cell count, identifying parasites is not so important. You already know you started
with someone's blood, so all you need to tell is is it a WB cell or a red corpuscle.
Getting a volume correct might be more valuable and maybe a standard count by humans
uses such a small volume squashed under a cover slip that a lancet drop
would give better counts than a big old vial.
Garbage in, garbage out.
On Mon, Aug 4, 2014 at 8:06 AM, Will Sutton <wsutton17@gmail.com> wrote:
Athelas, an iphone app to do hematology (blood imaging) won first place in a
Y-Combinator hackathon this weekend. It strikes me as very DIY-esque, doing
something complicated and expensive on a minimum budget.
That's just tech innovation is all.
Over in the discussion on H/N, I'm seeing a lot of poo-poo-ing of the
feasibility by actual technicians.
What do you guys think: Can good ML on cheap devices increasingly replace
lab tests?
What do you mean by ML?
Machine Lab?
Yes, it going to be the answer, but not on cheap generic hardware, on semi secret sauce
MEMS/CMOS fabbed silicon hardware by theranos.com See this article: http://fortune.com/2014/06/12/theranos-blood-holmes/
Also, is this a harbinger of more buy-in for DIYbio concepts from
the mainstream venture community?
Like Bryan said, no, the VCs want leverageable ways to offer more for less in high volume
to the masses and delivered by teams of twenty-somethings that will work for stock options
and make them, (the VCs), rich by 100X ROI.
There's nothing DIY about more for less.
More bang for less bucks the driver behind business in general.
DIY is often less for less. Less as in -- with the materials at hand, crude and unpolished.
--
-- You received this message because you are subscribed to the Google Groups DIYbio group. To post to this group, send email to diybio@googlegroups.com. To unsubscribe from this group, send email to diybio+unsubscribe@googlegroups.com. For more options, visit this group at https://groups.google.com/d/forum/diybio?hl=en
Learn more at www.diybio.org
--- You received this message because you are subscribed to the Google Groups "DIYbio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diybio+unsubscribe@googlegroups.com.
To post to this group, send email to diybio@googlegroups.com.
Visit this group at http://groups.google.com/group/diybio.
To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/53DFCF0D.9090902%40industromatic.com.
For more options, visit https://groups.google.com/d/optout.
-- You received this message because you are subscribed to the Google Groups DIYbio group. To post to this group, send email to diybio@googlegroups.com. To unsubscribe from this group, send email to diybio+unsubscribe@googlegroups.com. For more options, visit this group at https://groups.google.com/d/forum/diybio?hl=en
Learn more at www.diybio.org
---
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diybio+unsubscribe@googlegroups.com.
To post to this group, send email to diybio@googlegroups.com.
Visit this group at http://groups.google.com/group/diybio.
To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/CA%2B82U9%2B%2BLzv81y2H5PErdg0TP1h6qqLqBZqTt9zfqBDmaQn6Lg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
0 comments:
Post a Comment