Novotel Rotorua, 9th and 10th April 2018
Half-day and full-day short courses will be offered, to precede the 18th ANISG/NZNIRSS Conference.
- Half day short courses on NIR Basics (Jim de Heseth)
- Half day short courses on Introduction to Chemometrics (Brad Swarbrick)
- Full day courses on Sampling for NIR (Anders Larsen) and Hyperspectral Imaging (Jose Amigo Rubio)
These short courses are not to be missed.
Anders Larsen, Interline
Anders Larsen has 20+ years of experience with industrial implementation of PAT and spectroscopy in a wide range of industries and applications. He’s background is in opto electronics and physics. His focus has always been on the practical aspects of NIR projects including a passion for sampling in a wide sense. He has been an external lecturer at the University of Copenhagen for 10+ years, is the co-writter of the mythbuster column in NIRnews and a frequent speaker at short courses involving sampling and industrial spectroscopy.
The course will focus on a variety of sampling errors, visualized and understood with practical examples as well as appropriate levels of in depth theory.
The course addresses four major topics relevant for both at-line and on-line use.
- How to take a representative sample
- Time dependent sampling from the process
- Presenting the sample to the analyzer
- Analyzer settings “optical sampling”
Brad Swarbrick, Quality by Design Consultancy.
Brad Swarbrick is the owner of Quality by Design Consultancy, an Australian company that services a number of industries in Oceania, SE Asia and Japan, particularly supporting pharmaceutical applications of Chemometrics, PAT and QbD when applied to spectroscopic and process data.
Brad has over 20 years of experience in the application of Chemometrics and Design of Experiments (DoE) methodology in a wide range of industries, including agriculture, petrochemical and pharmaceuticals. Until recently, Brad was the Chief Operating Officer of CAMO Software, a world leading provider of the Chemometrics software package The Unscrambler where he successfully brought a number of new software developments and large projects to commercial reality. Prior to CAMO Software, Brad worked in the pioneering Pfizer Process Analytical Technology (PAT) group, developed the entire NIR program for Sigma (now Aspen) pharmaceuticals and worked as a GMP consultant for one of SE Asia’s largest pharmaceutical consultancy groups.
This half day introduction provides an overview of the methods used to visualise and analyse data generated by NIR spectrometers and is applicable to any field of application. The course is presented as both theory and demonstrations and will allow new comers to have an appreciation of what is involved in the analysis of spectroscopic data and will provide some tips and tricks for experienced campaigners.
The course outline is provided as follows.
- Visualisation and inspection of NIR spectra before analysis.
- General overview of preprocessing methods common to NIR spectroscopy.
- Exploratory data analysis using Principal Component Analysis (PCA).
- Introduction to Multivariate Classification
- Defining calibration and validation sets.
- Developing robust models using Partial Least Squares Regression (PLSR).
- Interpreting and implementing regression models.
- Course summary and discussion.
No software downloads are required for the course, however, if you want to follow along, then the data sets will be provided along with a demo version of software for the analysis, or just use your own favourite package to analyse the data sets.
Dr. José Manuel Amigo, Department of Food Science of the University of Copenhagen
Born in 1978. He obtained his PhD (Cum Laude) in Chemistry from the Autonomous University of Barcelona, Spain. Since 2007 he has been employed at the Department of Food Science of the University of Copenhagen, Denmark, as Associate Professor. During 2017 he was Guest Professor of the Federal University of Pernambuco, Brazil. Current research interests include hyperspectral and digital image analysis, Food Sciences, environmental modelling, curve resolution and teaching chemometrics. He has authored more than 140 publications (95+ peer-reviewed papers, books, book chapters, proceedings, etc.) and given more than 60 conferences and courses at international meetings. Jose has supervised or is currently supervising several Masters, Post Docs and PhD students and he is an editorial board member of four scientific journals in the fields of chemometrics, pharmaceutical sciences and analytical chemistry. He received the “2014 Chemometrics and Intelligent Laboratory Systems Award” for his achievements in the field of Chemometrics.
This mini-course will explore basic concepts of hyperspectral analysis, plus it will make emphasis in the different analytical situations that can be found. Near infrared hyperspectral imaging (NIR-HSI) has gained a tremendous importance in the laboratories of very diverse fields (Forensics, food production, pharmaceutical production, etc.). At the same time, it has generated the need of applying different strategies of analysis and algorithms in order to extract the most relevant information from the hyperspectral images in the different analytical situations. The course topics will be:
- Hyperspectral image acquisition, pre-processing and data selection
- Basic Chemometrics. Exploration. Regression. Segmentation
- Different types of hyperspectral images require different types of data analysis methods
- Integration of NIR-HSI in different analytical frameworks
The participants will be given different sets of hyperspectral images to be analysed by using HYPER-Tools (www.hypertools.org). Therefore, it is preferable that the students bring their own laptop with Matlab installed (version 2014b or further) and HYPER-Tools already installed.
Jim de Haseth, Professor Emeritus, University of Georgia (USA), and Technical Officer, LLS Instruments, Inc.
Jim de Haseth has more than 40 years experience with vibrational spectroscopy and vibrational spectrometer instrumentation. His interests have involved studies in the mid-infrared, Raman, and near-infrared, development of analytical methodologies, hyphenated methods, data processing, protein conformation, and fundamental studies of mid-infrared, near-infrared, and Raman spectra.
This course will emphasize the assignment of spectral bands in the NIR region. This will start with a discussion of the fundamental vibrations in the mid-infrared and explain how overtone and combination bands can appear in the NIR. Some direct guidelines will be presented to show which transitions are allowed. This will not be a theoretical discussion that involves mathematics or quantum mechanics, but rather an understanding of the physical phenomena that led to the development of the mathematical approaches. Further topics will be instrumental issues such as resolution, data intervals, instrumental effects on spectral lineshape, and, effects of mathematical manipulation of spectra.