Forum Replies Created
-
Posts
-
Hi,
First, it should be noted that SCiO is NOT a medical device.
However, planned future versions include the ability to scan different parts of the body and analyze human tissue and fluids such as blood chemistry data.
Our research has found that substances like glucose have concentration levels that are undetectable for SCiO when used non-invasively. Other molecules, such as hemoglobin, may perhaps one day be detectable because they exist in sufficiently large concentrations within the body.
Ayelet,
The ConsumerPhysics team
Hi all,Regarding the displayed graphs from SCiO scans:X is the wavelength represented in nm (nanometers).
Y is dependent on the state of the toggle buttons you select.
When:Processed (only): Assumes Beer-Lambert model is valid, and transforms the measured signal to be linear with concentration by doing a log transform and adjusting the result for noise and deviations from the model.
Normalized (only): Performs normalization of the signal. This is meant to compensate for changing measurement conditions (e.g. varied scanning distances) that typically occur from sample to sample. Y axis still means reflectance but in normalized units instead of raw reflectance.
Both Processed and Normalized: First assumes Beet-Lambert model (Processed) and then normalizes the results to compensate for differences in the optical path between samples. This is useful, for example, when there is variation in the thickness of the samples.
Both (log)R))” and Normalized:Similar to Processed and Normalized, uses a more aggressive form of Processed. Adds more noise, but in some cases may be the only way to create a good model.
Typically, different models and types of samples will require different pre-processing methods. You should both choose the pre-processing method to match your experimental setup and optimize the performance of your model. If you planned and gathered your data correctly, these efforts will coincide.
As for comparing with other techniques, there is always some difference between devices, especially if you use different pre-processing methods.
It should be noted that SCiO is a complete system that includes spectrometer, internal illumination source and machine learning algorithms optimized for SCiO. The best way to compare those technique is to try and create chemoemetric models with each technique and evaluate their performance.
Regarding legal disclaimers, you can visit our Terms & Conditions page at our website here.
Regards,
Ayelet,
The ConsumerPhysics team
We definitely have plans for off-line use of SCiO’s models and apps (for example in the agriculture field), however the exact date is still known. We will allow off-line scanning (populating the database) with SCiOlab app in the near future.
Regards,AyeletConsumerPhysicsHi Khoa,
At the moment moving or copying scans between collections option is not available via SCiOlab.
This could be done by us, as long as both data collections have identical attributes list.
The feature will be available in the future in SCiOlab as well.
As for modifying attributes, you can manually modify them in the merged collection.
Please send us the specific data collections you would like to merge and verify that the attributes list is identical.
Regards,
Ayelet
ConsumerPhysics
Dear Marc,SCiO can identify polymers, but with limitations.
Colors may change the IR spectrum, making it impossible to identify the plastic.For example, many black plastics are not only black in the visible, but also in IR, so practically they have no spectral signature for SCiO to measure.Nylons should be easier to detect, however that depends on the sample thickness.Regards,Ayelet HillmanConsumerPhysicsDear Nicola,
You can use the same data collection in order to build two different models based on two different filtered wavelength range.
However, you cannot filter 2 wavelength ranges in the same model.
In the future we may enable choosing multiple wavelengths regions.
Regards,
Ayelet Hillman
ConsumerPhysics
Dear Chris,
SCiO is a learning device, it needs to be trained in order to give useful information. The training process is done by scanning many samples and filling in their attributes (for example, pill brand and type, apple’s nutritional values etc.)
You can verify your scans are valid in two methods:
Visually inspecting the spectrum:
Scans of similar materials should look similar.
Extreme noise and wavy spectra probably indicate an invalid spectrum. SCiO’s spectral range usually include up to 4 strong absorption bands (in special cases more). If the whole spectrum is wavy, it is probably invaild.
Attached is an example for invalid spectrum.
Using the model performance diagrams:
Once you have enough samples, you can create a model (classification / estimation), verify if any of the scans are not classified/estimated correctly and filter them out.
Attached is an example for invalid scans.
Once you have accomplished the above stages, your models can be used as the basis of a future Application.
In order for your scan to be classified to any group (in case of classification model) or be estimated as containing several component (in case of estimation model), you can use the ‘Test Model’ button in your SCiOlab app.
With your permission, we can copy your data collection to our account and give you some feedback.
