Aller au menu principal Aller au contenu principal

STAT-IR 6

11 mai au 13 mai 2022 En ligne

Infrared microspectroscopy and imaging give fundamental information on the chemical composition and spatial organization of complex materials and on the conformation of molecules with spatial resolutions in the dozen microns to the submicron range. Data analysis is one of the most crucial parts of the process to unravel the rich information contained in the infrared spectra of complex systems and numerous specific techniques can be applied.

One of the most successful approaches lies on the statistical analysis of large datasets to explore their spectral variability, classify samples on the basis of their infrared spectra, find correlations between data even including data from different techniques, or establish predictive models for quantification. Such methods can only be successful if specific spectral preprocessing methods are applied to remove artefacts and confusing variations. Specific software, preprocessing techniques and strategies must be applied requiring specialized training combining signal processing and multivariate statistical analysis.

The SMIS beamline organizes the sixth edition of its infrared microspectroscopy data analysis course that will be held at SOLEIL from Wednesday, May 11 to Friday, May 13. The language course will be English.

The SMIS beamline is the infrared microspectroscopy beamline of SOLEIL synchrotron facility located in Gif-sur-Yvette near Paris. The SMIS beamline has over 10 year experience in the measurement and analysis of microspectroscopy data. The course will be given by Ferenc Borondics (an expert in physical-chemistry), Marko Toplak (a machine learning specialist), and Christophe Sandt (biospectroscopist).

The formation will be held online and opened to all past and future users of the beamline. The training will be held free of charge.

The course will deal with the main multivariate statistical analysis techniques such as principal component analysis, multivariate regression techniques and cluster analyses. Machine learning methods such as artificial neural networks will also be touched upon. We will devote some time to the analysis of spectral images with the open-source software Quasar. A hand-on session with participant data will be held.

Downloading courses

Dowload the courses PDF icon Quasar training (12.17 MB).

Speakers

The course will be given by

Ferenc Borondics (an expert in physical-chemistry),
Marko Toplak (a machine learning specialist),
and Christophe Sandt (biospectroscopist).

Program

Preliminary programm

 
May 11th
May 12th

Supervised learning

May 13rd

Unsupervised learning

09:00 - 10:30

SESSION 1
Getting started with Quasar (installation, basic Orange and Quasar functionality)
Speaker: C Sandt, M Toplak

SESSION 1
Introduction to supervised learning
Speaker: C Sandt

SESSION 1
Introduction to unsupervised learning
Speaker: C Sandt

10:30 - 10:45 Break Break Break
10:45 - 12:00

SESSION 2
Spectral Preprocessing
Visualization – mapping and imaging

Speaker: F Borondics

SESSION 2
Classification of spectra and hyperspectral datasets using various methods
Speaker: F Borondics

SESSION 2
Clustering of spectra and hyperspectral datasets using various methods
Speaker: M Toplak

12:00 - 13:30 Lunch break Lunch break Lunch break
13:30 - 15:30

SESSION 3
Statistical data exploration
PCA, PCA visualization
PCA imaging

Speaker: C Sandt

SESSION 3
Model inspection and cross-validation
Prediction
Common errors

Speaker: M Toplak

 
15:30 - 15:45 Break Break  
15:45 - 16:00

SESSION 4
Hands-on work with participants' data

SESSION 4
Hands-on work with participants' data

 

 

The formation will be held online and opened to all past and future users of the beamline. The training will be held free of charge.
 

Registration is open: registration.

Deadline

  • Registration : May 10th, 2022

How to attend the training

You will be able to log in to the course using the following link:
Available soon

Download Quasar

The training will be held with the open-source software Quasar.
Please download it before the course here.

For further information please contact: school-STATIR6-2022@synchrotron-soleil.fr