Forum Replies Created
-
AuthorPosts
-
AyeletKeymaster
Hi Leigh,
The liquid accessory is indeed not available yet, but will be in the near future.
As for self fabrication of a liquid accessory: SCiO needs to be in contact with the optical window so you should make sure there’s no direct illumination from the illumination source to the spectrometer. No focusing is needed. You will also need some kind of a reflector to make sure the light comes back to the sensor.
Ayelet
May 29, 2016 at 8:08 am in reply to: Classification model not working despite high correlation #3426AyeletKeymasterHi,
Since you only used 16 samples, most likely the problem is indeed the database size.
I would recommend, as I mentioned before, adding more samples to the the data collection.
Keep us updated!
Ayelet
AyeletKeymasterHi Eugene,
Currently there are no planned meet-ups at all.
We will keep you posted!
Ayelet
AyeletKeymasterYes, Kickstarter backers (non-dev) will get v1.1 hardware.
Ayelet
AyeletKeymasterHi Ruby,
This is a correct workflow.
SCiO Lab allows you to insert the values either before the scanning process or retroactively. The final results are identical.
Ayelet
AyeletKeymasterIt is now updated:
https://dev.consumerphysics.
com/mobile-api/android-api/ creating-your-scio-app/scan-a- sample/scan-a-sample/?preview= true&preview_id=1626&preview_ nonce=6d405a98b6 AyeletKeymasterHi Arndt,
Thanks for reporting the issue.
We have removed the current Android article and will upload soon the correct version.
Ayelet
AyeletKeymasterHi,
First, we believe that SCiO can be very relevant to the healthcare field and we are currently studying several directions, mostly lifestyle centered, such as body fat and cosmetics personalization.
Other applications may also be feasible as long as they fit SCiO’s sensitivity requirements.
Your suggestion to scan the body and determine if it is functioning properly by its molecular composition or to create a database of human molecular states in various emotional situations – will not be feasible with SCiO.
As for scanning a body part before and after a treatment to determine how it effected the molecular composition of the body – this may be feasible in terms of moisture and fat estimation, however, further research is necessary.
Ayelet
AyeletKeymasterHi Rojer,
We will contact and update you personally via e-mail as soon as possible.
Ayelet
AyeletKeymasterAyeletKeymasterHi Nga,
I have responded you via e-mail.
Ayelet
May 15, 2016 at 11:12 am in reply to: Classification model not working despite high correlation #3306AyeletKeymasterHi Guoweiz,
The problem may be derived from a number of reasons.
We would like to ask you for some more information:
1. Have you scanned new samples in different temperatures? (specifically different temperatures than measured when collecting the database)
2. Have you used actual different samples or scanned multiple scans for identical samples and referred them as new samples?
3. Did you use the solid sample holder?
A high correlation on a low number of samples can originate from over-fitting as well (in over-fitting, a statistical model describes random error or noise instead of the underlying relationship).
Therefore, even if the classification model seems successful, it is not robust.
May I ask which samples did you scan? Can you please add a photo of the samples in the data collection?
Ayelet
AyeletKeymasterYes, access to chemical characteristics and other analysis methods results are required. It might be expensive but since SCiO is a learning device and based on machine learning algorithms – it must be populated by known values.
Ayelet
AyeletKeymasterThe most successful detection threshold is: 0.1%.
AyeletKeymaster -
AuthorPosts