Spektroskopie und Chemometrie Neuigkeiten Wöchentlich #16, 2017

Get the Chemometrics and Spectroscopy News in real time on Twitter @ CalibModel and follow us.

Chemometrics

UV/Vis spectroscopy combined with chemometrics for monitoring solid-state fermentation with … LINK

What is MODEL SELECTION? What does MODEL SELECTION mean? MODEL SELECTION meaning & explanation: LINK

Model selection with multiple regression on distance matrices leads to incorrect inferences LINK



Equipment

Tiny spectrometer turns smartphone into molecular analyst LINK

High-grade, compact spectrometers for Earth observation from SmallSats LINK



Future

Miniature and Micro Spectrometers End-Users Needs, Market & Trends 2015-2021 – Research and Markets – Yahoo Finance LINK



Agriculture

Prediction of milk fatty acid content with mid-infrared spectroscopy in Canadian dairy cattle using differently… LINK!



Other

Smallest seismic sensor uses vibration spectral analysis LINK




Spektroskopie und Chemometrie Neuigkeiten Wöchentlich #5+6, 2017


Chemometrics

Non-Destructive Sensor-Based Prediction of Maturity and Optimum Harvest Date of Sweet Cherry Fruit | sensors LINK


IDC unveils its Top 10 Predictions for global Robotics Industry Industry40 Robotics LINK


Spectroscopy

Global Molecular Spectroscopy Market is expected to reach USD 6.712 billion till 2024. htt… LINK!


Near Infrared

Assessing pre-harvest sprouting in cereals using near-infrared spectroscopy-based metabolomics LINK


Rapid screening of commercial extra virgin olive oil products for authenticity: Performance of a handheld NIR device LINK


Hyperspectral

Imec () launches TDI, multispectral and hyperspectral sensors | imaging HSI LINK


Near-infrared hyperspectral imaging of lamination and finishing processes in textile technology LINK


Spectral Imaging

Viavi Solutions and ESPROS Photonics Corporation Debut New Miniaturized Spectral Sensor and Multispectral Sensor LINK


Equipment

Meta-lenses bring benchtop performance to small, hand-held spectrometer – Science Daily LINK



Scan anywhere with Neospectra Spectrometer Case powered by XPNDBLS PhotonicsWest … LINK!


Agriculture

World feed production exceeds 1 billion MT LINK


Chemometric soil analysis on the determination of specific bands for the detection of magnesium & potassium by … LINK


Other

This app uses spectral analysis to analyze objects and their makeup HawkSpex LINK


Research details developments in the multivariate analysis software industry | MVA LINK

“The worlds first ever spectroscopy enabled iPhone!” Check out our video to see it in action: LINK


Investments in AI will triple in 2017. ($47 billion by 2020 per ) CIO CMO | LINK


Some aspects of fetal development have long puzzled scientists, but new molecular technologies are shining a light: https:/… LINK!


CalibrationModel.com

Spectroscopy and Chemometrics News Weekly 3+4, 2017 | Spectroscopy NIRS MVDA… LINK


Spektroskopie und Chemometrie Neuigkeiten Wöchentlich 3+4, 2017 | NIRS Spektroskopie Chemometrie Multivariate LINK


Spettroscopia e Chemiometria Weekly News 3+4, 2017 | NIRS Spettroscopia Chemiometria news LINK


WHITE PAPER: A novel knowledge-based Chemometric Software Framework for quantitative NIRS Calibration Modeling LINK



Improve Accuracy of fast non-destructive NIR Measurements by Optimal Calibration | spectroscopy sensor modeling LINK


NIRS as a secondary method requires extensive calibration on an ongoing basis | foodindustry Digitalization IoT LINK


Services for Optimization of Chemometric Application Methods of Near-Infrared Spectroscopy | Quality Control NIRS LINK


► Timesaving NIRS Calibration ► near-infrared spectroscopy | protein fat moisture sensor measurement scanning LINK





Spektroskopie und Chemometrie Neuigkeiten Wöchentlich #3+4, 2017


Chemometrics

Fast sampling, analyses and chemometrics for plantbreeding: Bitter acids, xanthohumol and terpenes in lupulin … LINK


Near Infrared

Measurement of Soy Contents in Ground Beef Using Near-Infrared Spectroscopy LINK


Near infrared spectroscopy for body fat sensing in neonates: quantitative analysis by GAMOS simulations | Bodyfat LINK



Fatty acids and fat-soluble vitamins in ewe’s milk predicted by near infrared reflectance spectroscopy. | NIRS LINK


NeoSpectra Micro, a small, chip-scale, near infra-red (NIR) spectral sensor – LINK


Hyperspectral

Drones on the Farm: Agribotix Partners with senseFly, and Sentera Announces Real-Time NDVI Streaming LINK


Early Detection of Aspergillus parasiticus Infection in Maize Kernels Using NIR Hyperspectral Imaging and MDA LINK



Nondestructive Estimation of Moisture Content, pH and Soluble Solid Contents in Intact Tomatoes Using HSI LINK


Equipment

Spectrometers: Excitation source parameters dictate Raman spectroscopy outcomes LINK


Future

Is molecular scanning the next killer smartphone app? killerapp futuretrends sensorik sensor LINK


Other

It took 50 years for the world to install the first million industrial robots. The next million will take only eight … LINK!


“How Statistics lost their power – and why we should fear what comes next” | BigDataAnalytics bigdata LINK



Spectroscopy and Chemometrics News Weekly #47, 2015

Near Infrared

NIR-Sensor ermittelt Trockensubstanz während der Mischwagenbefüllung | Futterkomponente LINK

Ultra-low maintenance FTNIR analyzer for the refining & petrochemical industries | pauto LINK


Infrared

Seeing Through Crude Oil for Efficient Oil Separations using Short-Wave Infrared (SWIR) Cameras – AZoSensors LINK


Facts

RoboBees Can Fly and Swim. What’s Next? Laser Vision – Smithsonian UAS UAV LINK


Equipment

Scientists create an all-organic UV on-chip spectrometer – The U.S. Department of Energy’s Ames LINK


Agriculture

… detection of contaminants in agro-food products, … melamine levels in milk using vibrational spectroscopy LINK


Laboratory

Examining Pigmented Human Tissue using SWIR Raman Spectroscopy – AZoSensors LINK


Other

SCiO Molecular Scanner UNBOXING – Video LINK



CalibrationModel.com

Dear NIR-Spectrometer vendors, this is about how you can improve customer web-traffic | NIRS Spectrometer LINK

Efficient development of new quantitative prediction equations for multivariate NIR spectra | spectra LINK

How to Develop Chemometric Near-Infrared Spectroscopy Calibrations in the 21st Century? | NIR LINK

How to Develop Near-Infrared Spectroscopy Application Today? | pharma lab analysis chemist TechTrends LINK

Improve chemical analysis accuracy by optimized chemometric models for Near-Infra-Red (NIR) Spectroscopy LINK

Improving Accuracy, Precision and Robustness of NIR-analysis LINK

NewsLetter: Spectroscopy and Chemometrics News Weekly 46, 2015 | Molecular Spectroscopy NIRS Chemometrics Raman LINK

Pro Tip: The NIR calibration is the central key to accurate NIR measurement LINK

Services for professional Development of Near-Infrared Spectroscopy Calibration Methods | NIR Quality Testing LINK



Arbeitsweisen zur NIR Kalibrierung – Erstellung von NIRS-Spektroskopie Kalibrierungskurven

Kennen Sie den Effekt, dass Sie bevorzugt ihre Lieblings-Datenvorbehandlungen in Kombination durchprobieren und oft die gleichen Wellenlängen-Selektionen anhand der visualisierten Spektren ausprobieren?

Man probiert z.B. sechs bis zehn Kombinationen aus, bis man davon sein favorisiertes Kalibrationsmodell auswählt, um es dann weiter zu optimieren. Da fallen dann plötzlich Ausreisser (Outlier) auf, weil man in die Tiefe geht, also mit den Daten vertraut ist, man kennt mittlerweile die Spektren-Nummern der Ausreisser und ist mit den Extremwerten vertraut.

Jetzt fokussiert man sich auf die Hauptkomponenten (Principal Components, Latent Variables, Faktoren) und achtet darauf nicht zu über-fitten und nicht zu unter-fitten. Das ganze dauert ein paar Stunden und schliesslich begnügt man sich mit dem gefundenen Modell.

Was wäre nun, wenn man in all den zu Beginn ausprobierten Varianten, die gefundenen Ausreisser entfernt und nochmals berechnet und vergleicht? Wären die Ergebnisse besser als die von der bisherigen Modell Wahl? Man probiert es nicht aus? Weil es mühsam ist und wieder Stunden dauert?

Wir haben eine Software entwickelt die dies so vereinfacht, dass auch die Anzahl der Modell Variationen beliebig erhöht werden kann. Die Varianten Erzeugung läuft automatisiert mit einem intelligenten Regelsystem, so auch die Optimierung und das Vergleichen der Modelle und schliesslich die finale Auswahl des Besten Kalibrations Modell.

Unsere Software beinhaltet alle üblichen bekannten Datenvorbehandlungs Methoden (Preteatments) und kann diese sinnvoll kombinieren. Da viele Preteatments direkt abhängig sind von der Wellenlängen Selektion, so z.B. die Normalisierungen die innerhalb eines Wellenlängen-Bereiches die Skalierungsfaktoren ermittelt, um die Spektren damit zu normieren, werden die Pretreatments mit dem Wellenlängen-Bereichen kombiniert. So kommt eine Vielzahl von sinnvollen Modell Einstellungen zusammen die alle berechnet und optimiert werden.

Für die automatische Auswahl der relevanten Wellenlängen Bereiche kommen verschiedene Methoden zum Einsatz, die sich an den Spektren Intensitäten orientieren. So werden z.B. Bereiche mit Totalabsorption nicht verwendet, oftmals störende Wasserbanden entfernt oder beibehalten.

Über all die berechneten Modell Variationen können so zusammenfassende Outlier Analysen gemacht werden. Werden durch die gefahrenen Versuche neue Outlier (Hidden Outlier) entdeckt, können alle bisherigen Modelle automatisch ohne diese Ausreisser nachberechnet, optimiert und verglichen werden.

Aus dieser Vielzahl berechneter Modelle mit deren Statistischen Güte Bewertungen (Prediction Performance) kann nun die optimale Kalibration ausgewählt werden. Dazu wird nicht einfach nach dem Vorhersage Fehler (Prediction Error, SEP, RMSEP) oder nach dem Bestimmtheitsmaß (Coefficient of Determination r2) sortiert, sondern mehrere Statistik- und Testwerte gemeinsam zur umfänglichen Beurteilung der optimalen Kalibration herangezogen.

Somit haben wir eine Plattform geschaffen, die es ermöglicht hochgradig automatisiert das zu tun, was ein Mensch niemals mit einer handelsüblichen Software tun kann.

Wir bieten damit die grösste Anzahl auf Ihr Applikations-Problem angepasste Modellierungs-Berechnungen und wählen die beste Kalibration für Sie aus!

Das heisst, unsere Ergebnisse sind schneller, genauer, robuster und objektiv ausgewählt (Personen unabhängig) und für Sie ganz einfach anzuwenden.

Die Kontrolle über die von uns gelieferten Modelle haben Sie vollumfänglich, denn wir liefern einen klar strukturierten und detaillierten Bauplan der  kompletten Kalibration, mit allen Einstellungen und Parametern, mit allen notwendigen Statistischen Kenngrössen und Grafiken.

Anhand dieses Bauplans können Sie das quantitative Kalibrations Modell selbst in der von Ihnen verwendeten Software nachstellen, nachvollziehen und vergleichen. Sie haben so alles im Griff, für die Modell-Validierung und die Modellpflege.

Der Datenschutz ist uns sehr wichtig. Die NIR Daten, die Sie uns für die Kalibrations-Erstellung kurzzeitig zu Verfügung stellen bleiben selbstverständlich Ihr Eigentum. Ihre NIR Daten werden nach Abschluss des Auftrags bei uns gelöscht.

Interessiert, dann zögern Sie nicht uns zu kontaktieren.

NIRS Calibration Model Equation – Optimal Predictive Model Selection

To give you an insight what we do to find the optimal model, imagine a NIR data set, where a NIR specialist works hard for 4 hours in his chemometric software to try what he can with his chemometric-, NIR spectroscopic- and his product-knowledge to get a good model. During the 4 hours he finds 3 final candidate models for his application. With the RMSEP of 0.49 , 0.51 and 0.6. Now he has to choose one or to test all his three models on new measured NIR spectra.

That is common practice. But is this good practice?

And nobody asks, how long, how hard have you tried, how many trial have you done, if this really the best model that is possible from the data?
And imagine the cost of the data collection including the lab analytics!
And behind this costs, have you really tried hard enough to get the best out of your data? Was the calibration done quick and dirty on a Friday afternoon? Yes, time is limited and manually clicking around and wait in such kind of software is not really fun, so what are the consequences?

Now I come to the most important core point ever, if you own expensive NIR spectrometer system, or even many of them, and your company has collected a lot of NIR spectra and expensive Lab-reference data over years, do you spend just a few hours to develop and build that model, that will define the whole system’s measurement performance for the future? And ask yourself again (and your boss will ask you later), have you really tried hard enough, to get the best out of your data? really?

What else is possible? What does your competition do?

There is no measure (yet) what can be reached with a specific NIR data set.
And this is very interesting, because there are different beliefs if a secondary method like NIR or Raman can be more precise and accurate, as the primary method.

What we do different is, that our highly specialized software is capable of creating large amounts of useful calibrations to investigate this limits – what is possible. It’s done by permutation and combination of spectra-selection, wave-selection, pre-processing sequences and PC selections. If you are common with this, then you know that the possibilities are huge.

For a pre-screening, we create e.g. 42’000 useful calibrations for the mentioned data set. With useful we mean that the model is usable, e.g. R² is higher than 0.8, which shows a good correlation between the spectra and the constituent and it is well fitted (neither over-fitted nor under-fitted) because the PC selection for the calibration-set is estimated by the validation-set and the predictive performance of the test-set is used for model comparisons.

Here the sorted RMSEP values of the Test Set is shown for 42’000 calibrations.
You can immediately see that the manually found performance of 0.49 is just in the starting phase of our optimization. Interesting is the steep fall from 1.0 to 0.5 where manually optimization found it’s solutions. A range where ca. 2500 different useful calibrations exist. The following less steep fall from 0.5 to 0.2 contains a lot more useful models and between 0.2 to 0.08 the obvious high accurate models are around 2500 different ones. So the golden needle is not in the first 2500 models, it must be somewhere in the last 2500 models in the haystack.

Sorted RMSEP plot of 42'000 NIR Calibration Model Candidates

That allows us to pick the best calibration out of 42’000 models, depending on multiple statistical evaluation criteria, that is not just the R² or RPD, SEC, SEP or RMSEP, (or Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), Multivariate AIC (MAIC) etc.) we do the model selection based on multiple statistical parameters.

Dengrogram plot of similar  NIR Calibration Models

To compare the calibration models by similarity it is best viewed with dendrogram plots like this (zoomed in), where the settings are shown versus the models overall performance similarity. In the settings you can see a lot of different permutations of pre-processings combined with different wave-selections.

NIR Spectroscopy Calibration Report for quantitative predictive models

When you send your quantitative NIR spectra data to our NIR Calibration Model Service, you get a detailed calibration report (calibration protocol) of the found optimal calibration settings, so you are able to see all insights and easily re-build the model in your NIR/Chemometric software.

Here is a part of our calibration report, that exactly describes the data used in the calibration set (CSet), the validation set (VSet) and the test set (TSet). The numbers are the number ids of the spectra in your delivered NIR data file.


The calibration method settings and parameters are
Waveselection : the variable selection or wavenumber selection or wavelength selection
Pretreatments : the spectral data pre-processing
PCs : the number of Principal Components (PC) or Latent Variables (LV)
Method : the modeling method algorithm used, e.g. PLS

Then the statistical analysis of the PLS model by the different sets (CSet, VSet, Tset).

Calibration Report

Statistical analysis of calibration, validation and test results : 1 Name, 2 Unit, 3 N : number of spectra, 4 N : number of samples, 5 Average spectra count per sample, 6 Reference values, 7 Min, 8 Mean, 9 Median, 10 Max, 11 Standard deviation, 12 Skewness : left (-) or right (+) lack of symmetry, 13 Kurtosis : flat (-) or peaked (+) shape, 14 Model statistics, 15 RPD, 16 R², 17 RMSEC, RMSEP, RMSET : root mean square of prediction errors, 18 SEC, SEP, SET : standard error (bias corrected), 19 Bias, 20 Skewness of prediction errors, 21 Kurtosis of prediction errors, 22 Intercept, 23 Slope, 24 Intercept (reverse), 25 Slope (reverse), 26 Sample Prediction Repeatability Error, 27 Sample Prediction Repeatability Error (of Missing data MSet)

This shows how we deliver the optimal settings. With the statistical values, the NIR model predicted values of all spectra and additional plots you are able to compare with your re-built model to verify that the models perform nearly equally.

Einzigartigkeit


Es ist ganz einfach
1. Sie senden NIR Daten (wir sammeln, verteilen oder verkaufen diese nicht)
2. Sie erhalten den optimalen Modell Bauplan
3. Sie bauen das Modell nach, validieren es, verwenden es

Wir haben Chemometrie Software nicht um Chemometrie zu machen,
wir haben die Lösung um das optimale Modell für Ihre NIR Daten zu erstellen
damit Sie bessere NIR Messergebnisse erhalten.

Darum nennen wir es nicht “Chemometrie Software”.
Es ist ein Service, benannt nach dem was er liefert, ein Calibration Model.
Es gibt Ihnen ein optimales chemometrisches Modell für Ihre NIR Daten.
Das ist es was Sie erreichen wollen.

Also mühen Sie sich nicht ab mit Chemometrie und verbringen Sie nicht Ihre Zeit mit den endlosen Möglichkeiten die chemometrische Software bietet mit klicken und warten wenn Sie ein optimales Modell für Ihre Daten via einen Service erhalten können!

Es gibt keine Sperren oder Abhängigkeiten,
da keine Software installiert werden muss.

Sie erhalten keine Black-Box.
Da das Modell wird als detaillierte und vollständige Blaupause in menschenlesbarer Form geliefert.

Sie bleiben unabhängig.
Sie haben immer die Wahl:
- Sie können es immer noch tun, wie Sie es bisher getan haben.
- Sie werden erleben, dass Sie mit dem Service die besseren Modelle schneller und preiswerter erstellen.

Lassen Sie uns einen Versuch
Bitte kontaktieren Sie uns, so können wir Ihnen helfen!

Wie werden Nahinfrarotspektroskopie Kalibrierungen im 21. Jahrhundert entwickelt?


The Problem

Calibration modeling is a complex and very important part of NIR spectroscopy, especially for quantitative analysis. If the model is badly designed the best instrument precision and highest data quality does not help getting good and robust measurement results. And NIR Spectroscopy requires periodically recalibration and validation.


How are NIR models built today?

In a typical usage in industry, a single person is responsible to develop the models (see survey). He or she uses a Chemometric software that has a click-and-wait working process to adjust all the possible settings for the used algorithms in dialogs and wait for calculations and graphics and then to think about the next modeling steps and the time is limited to do so. Do we expect to find the best use-able or optimal model that way? How to develop near-infrared spectroscopy calibrations in the 21st Century?


Our Solution

Why not put all the knowledge a good model builder is using into software and let the machines do the possibilities of calculations and presenting the result? Designing the software that way, that the domain knowledge is built-in, not just only the algorithms for machine learning and make it possible to scale the calculations to multi-core computers and up to cloud servers. Extend the Chemometric Software with the Domain Knowledge and make as much computer power available as needed.

As it was since the beginning

User → Chemometric Software → one Computer → some results to choose from

==> User’s time needed to click-and-wait for creating results

Our Solution

User → (Domain Knowledge → automatized Chemometric Software) → many Computers → the best models

==> User’s time used to study the best models and reasoning about his product / process

Note that the “Domain Knowledge” here does perfectly support the User’s product and process knowledge to get the things done right and efficient.


Scaling at three layers

  • Knowledge : use the domain knowledge to drive the Chemometric Software
  • Chemometric Software : support many machine learning algorithms and data pre-processings and make it automatic
  • Computer : support multi-core calculations and scale it to the cloud

The hard part in doing this, is of course the aggregation of the needed domain knowledge and transform it into software. The Domain Knowledge for building Chemometric NIR Spectroscopic models is well known and it’s huge and spreads multiple disciplines. Knowledge-driven software for computing helps to find the gold needle in the haystacks. It’s all about scaling that makes it possible. See Proof of Concept.


New possibilities

  • NIR users can get help working more efficient and getting better models.
  • New types of applications for NIR can be discovered.
  • Evaluation of NIR Applications to replace conventional analytical methods.
  • Hopeless calibrations development efforts can be re-started.
  • Higher model accuracy and robustness can be delivered.
  • Automate the experimental data part of your application study.
  • Person independent optimization will show new solutions, because it’s not limited by a single mindset => combining all the aggregated knowledge and its combinations.
  • Software independent optimization will show new solutions, because none of vendor specific limitations and missing algorithms are present => combining all open available algorithms and there permutations.
  • Computing service is included.

Contact us for trial

Your NIR data is modeled by thousands of different useful calibration models and you get the best of them! That was not possible before in such a easy and fast way! See How it works

Proof of Concept

Chemometric software competitions (aka shootouts) are a good way to check algorithms, software and knowledge against all other experts in the field.

Imagine that the prediction results can be produced with any kind of software and newest algorithms.

And we just use PLS right to generate models that can be used on all NIR software systems, because PLS is a quasi standard, supported in all major chemometrics software.

Our software framework reached very good results, got gold (rank #1) and silver (rank #2) during well known international NIR Chemometric software shootouts* so far, the competitions are held bi-annual.

Rank / competitors Competition / Conference Year
#1 / 1 ** Kaji / ANSIG 2014
#1 / 150 Kaji / ANSIG 2012
#2 / ??? IDRC / IDRC 2012
The Kaji Competition

A set of NIR spectral data will be available for downloading from the ANISG website and contestants will be asked to find and explain a “best” chemometric model to robustly predict samples of the same type.
A panel will select the three “best” entries based on the predicted results and spectroscopic explanation of the products and attributes of interest.

http://www.anisg.com.au/the-kaji-competition


The IDRC Competition

The Software Shootout has been a staple of the IDRC. It is a competition amongst participants of the conference that aims at determining the person who developed the best model and obtained the lowest prediction error for a particular problem.
Every IDRC, a new challenge is proposed to participants. The challenge consists of a data set with calibration, test and a validation set.
Participants are given target values for the calibration and test sets but must do their best to develop a model that will predict the validation set as accurately and precisely as possible. Challenges from all sorts of fields of NIRS have been used (agriculture, biomedical, pharmaceutical, soil, …).

IDRC


*) The author was unable to present the results at the conferences, so this ranking was not official but confirmed by the shootout organizers. Thanks go to Benoit Igne, IDRC 2012 shootout organizer and Steve Holroyd, Kaji Competition organizer at ANISG Conference 2012.

Conclusion

Our chemometric software framework can significantly reduce the time spent for NIR method development and fine optimization. The time saving can be achieved through highly automated experiments and the usage of cloud computing. Calibrations are built and evaluated using automated good practices protocols resulting in useful, precise and robust Calibrations. The high number of experiments enables a deep screening of the solution domain to find the optimum calibration settings, something currently unavailable in standard chemometric software.

**) We were the only participator that got the 4 competition tasks (4-times more than usual) completed in that short time and submitted the fully documented results. After the competition, the information was given, that the data was originated from forages and the constituents were dry matter, organic matter digestibility, protein and ash. Thanks go to Daniel Cozzolino, Kaji 2014 Competition organizer.