Frank Westad - Academia.edu (original) (raw)

Papers by Frank Westad

NIR news

With the latest release of Unscrambler, Camo Analytics introduced support for Python scripting, g... more With the latest release of Unscrambler, Camo Analytics introduced support for Python scripting, giving users the best of two worlds. This Python extension allows users to tap into the vast ecosystem of Data Science tools that are continually being produced in the Python community, while still leveraging the familiar data handling, validation and visualization features of Unscrambler – all contained within a fully compliant framework. This paper discusses the value propositions that the Python extension can provide to Unscrambler users, and follows this up with some specific examples of common workflows that are enabled by this extension: Data Importing, Spectral Preprocessing and Innovative Modeling methods.

Analytical Methods

Hyperspectral imaging of environmental samples with infrared microscopes is one of the preferred ... more Hyperspectral imaging of environmental samples with infrared microscopes is one of the preferred methods to find and characterize microplastics.

Measurement Science and Technology

Chemie Ingenieur Technik, 2016

ABSTRACT This chapter presents the concept and theory of independent component analysis (ICA). Th... more ABSTRACT This chapter presents the concept and theory of independent component analysis (ICA). The method originated from signal processing research, where unknown signal sources are mixed to a new set of signals. This general objective of separating signals into pure sources is called blind source separation (BSS). ICA has been shown to be useful in solving the BSS problem, and if the pure sources are found, then also the mixing system may be estimated. Similar situations occur in chemometrics where the pure spectra of chemical compounds and their concentrations are observed with different type of instrumentation such as spectroscopy. The necessity of proper validation in ICA is emphasized and put into a chemometric framework. Examples on simulated as well as spectroscopic data are shown to illustrate the potential of ICA in chemometric applications.

Journal of Spectral Imaging

Independent component analysis combined with various strategies for cross-validation, uncertainty... more Independent component analysis combined with various strategies for cross-validation, uncertainty estimates by jack-knifing and critical Hotelling’s T2 limits estimation, proposed in this paper, is used for classification purposes in hyperspectral images. To the best of our knowledge, the combined approach of methods used in this paper has not been previously applied to hyperspectral imaging analysis for interpretation and classification in the literature. The data analysis performed here aims to distinguish between four different types of plastics, some of them containing brominated flame retardants, from their near infrared hyperspectral images. The results showed that the method approach used here can be successfully used for unsupervised classification. A comparison of validation approaches, especially leave-one-out cross-validation and regions of interest scheme validation is also evaluated.

PLOS ONE

A pilot study demonstrating real-time environmental monitoring with automated multivariate analys... more A pilot study demonstrating real-time environmental monitoring with automated multivariate analysis of multi-sensor data submitted online has been performed at the cabled LoVe Ocean Observatory located at 258 m depth 20 km off the coast of Lofoten-Vesterålen, Norway. The major purpose was efficient monitoring of many variables simultaneously and early detection of changes and time-trends in the overall response pattern before changes were evident in individual variables. The pilot study was performed with 12 sensors from May 16 to August 31, 2015. The sensors provided data for chlorophyll, turbidity, conductivity, temperature (three sensors), salinity (calculated from temperature and conductivity), biomass at three different depth intervals (5-50, 50-120, 120-250 m), and current speed measured in two directions (east and north) using two sensors covering different depths with overlap. A total of 88 variables were monitored, 78 from the two current speed sensors. The time-resolution varied, thus the data had to be aligned to a common time resolution. After alignment, the data were interpreted using principal component analysis (PCA). Initially, a calibration model was established using data from May 16 to July 31. The data on current speed from two sensors were subject to two separate PCA models and the score vectors from these two models were combined with the other 10 variables in a multi-block PCA model. The observations from August were projected on the calibration model consecutively one at a time and the result was visualized in a score plot. Automated PCA of multisensor data submitted online is illustrated with an attached time-lapse video covering the relative short time period used in the pilot study. Methods for statistical validation, and warning and alarm limits are described. Redundant sensors enable sensor diagnostics and quality assurance. In a future perspective, the concept may be used in integrated environmental monitoring.

Integrated Environmental Assessment and Management, 2016

The present paper describes integration of environmental monitoring and discharge data, and inter... more The present paper describes integration of environmental monitoring and discharge data, and interpretation using multivariate statistics, Principal Component Analysis (PCA) and Partial Least Squares (PLS) regression. The monitoring was carried out at the Peregrino oil field off the coast of Brazil. One sensor platform and three sediment traps were placed on the seabed. The sensors measured current speed and direction, turbidity, temperature and conductivity. The sediment trap samples were used to determine suspended particulate matter which was characterized with respect to a number of chemical parameters (26 alkanes, 16 polycyclic aromatic hydrocarbons (PAHs), nitrogen, carbon, calcium carbonate and barium). Data on discharges of drill cuttings and water based drilling fluid were provided on a daily basis. The monitoring was carried out during seven campaigns from June 2010-October 2012. each lasting 2-3 months due to the capacity of the sediment traps. The data from the campaigns were preprocessed, combined and interpreted using multivariate statistics. No systematic difference could be observed between campaigns or traps despite the first campaign was carried out prior to drilling, and one of the three sediment traps was located in an area not expected to be influenced by the discharges. There was a strong co-variation between suspended particulate matter and total nitrogen and organic carbon suggesting that the majority of the sediment samples had a natural and biogenic origin. Furthermore, the multivariate regression showed no correlation between discharges of drill cuttings and sediment trap or turbidity data taking current speed and direction into consideration. Due to this lack of correlation with discharges from the drilling location, a more detailed evaluation of chemical indicators providing information about origin was performed in addition to numerical modeling of dispersion and deposition. The chemical indicators and the modeling of dispersion and deposition support the conclusions from the multivariate statistics. This article is protected by copyright. All rights reserved.

Chemometrics and Intelligent Laboratory Systems, 2015

COMPSTAT 2004 — Proceedings in Computational Statistics, 2004

ABSTRACT

Data Handling in Science and Technology, 2013

ABSTRACT In this chapter, a survey of the theory behind the main chemometric methods used for mul... more ABSTRACT In this chapter, a survey of the theory behind the main chemometric methods used for multivariate calibration is presented. Ordinary least squares, multiple linear regression, principal component regression, partial least squares regression and principal covariate regression are discussed in detail. Tools for model diagnostics and model interpretation are presented, together with strategies for variable selection.

Journal of Near Infrared Spectroscopy, 2004

Journal of Chemometrics, 2006

Page 1. A modification of canonical variates analysis to handle highly collinear multivariate dat... more Page 1. A modification of canonical variates analysis to handle highly collinear multivariate data Lars Nørgaard1*, Rasmus Bro1, Frank Westad2 and Søren Balling Engelsen1 1Department of Food Science, Quality and Technology ...

Energy & Fuels, 2010

ABSTRACT Positive electrospray ionization mass spectrometry (ESI−MS) and multivariate regression ... more ABSTRACT Positive electrospray ionization mass spectrometry (ESI−MS) and multivariate regression (chemometrics) have been used for the identification and quantification of fatty acid methyl ester (FAME) in jet fuel in concentrations from 3 to 35 ppm. The jet fuel samples were injected directly and undiluted into the ion source. Each analysis takes less than 1 min to perform. Calibration series with rapeseed methyl ester (RME) and soybean methyl ester (SME) alone or in combination were used to create regression models with excellent prediction properties. An independent test set with known amounts of RME and SME was made several weeks later, and the regression model was used to predict the concentration of RME and SME with a root-mean-square error of prediction (RMSEP) of 2.6 and 1.2 ppm, respectively.

Background and objectives: Foods high in protein are known to satiate more fully than foods high ... more Background and objectives: Foods high in protein are known to satiate more fully than foods high in other constituents. One challenge with these types of food is the degree of palatability. This study was aimed at developing the frankfurter style of sausages that would regulate food intake as well as being the preferred food choice of the consumer. Design and measures: 16 sausage varieties with commercial (PE% 20) or higher amount of protein (PE% 40), being modified with vegetable fat (3% of rapeseed oil), and smoked or not, underwent a sensory descriptive analysis, in which the information was used to choose a subsample of four sausages for a satiety test. Twentyseven subjects were recruited based on liking and frequency of sausage consumption. The participants ranged in age from 20 to 28, and in body mass index (BMI) between 19.6 and 30.9. The students were served a sausage meal for five consecutive days and then filled out a questionnaire to describe their feelings of hunger, satiety, fullness, desire to eat an their prospective consumption on a visual analogue scale (VAS) starting from right before, right after the meal, every half hour for 4 h until the next meal was served, and right after the second meal. Results and conclusion: The higher protein sausages were less juicy, oily, fatty, adhesive, but harder and more granular than with lower amount of protein. The high-protein sausages were perceived as more satiating the first 90 min after the first meal. Some indication of satiety effect of added oil versus meat fat. No significant differences in liking among the four sausage varieties.

ABSTRACT: The effect of packages with different oxygen transmission rates (OTR), different gas-to... more ABSTRACT: The effect of packages with different oxygen transmission rates (OTR), different gas-to-product-volume (GP) ratios, and various levels of residual oxygen after packaging on the color stability of cooked ham exposed to commercial retail light conditions was studied. Sliced cooked ham was packaged in thermoformed packages with OTR of 0.04 and 0.06 mL O2/pkg×24 h and GP ratios of 2.6 and 4.1. After packaging, the packages were additionally divided into groups with 4 levels of residual oxygen ranging from 0.09% to 0.46%. The packaged ham was stored in darkness at 4 ◦C up to 33 d, and during the storage period samples were withdrawn and exposed to light for 2 d before instrumental and visual color evaluation. In order to maintain an acceptable color of this particular ham product when exposed to typical retail light conditions, the highest acceptable level of oxygen in the headspace of the packages was 0.15% oxygen at the time of illumination. This threshold level was independe...

Analytica chimica acta, Jan 17, 2015

In this tutorial, we focus on validation both from a numerical and conceptual point of view. The ... more In this tutorial, we focus on validation both from a numerical and conceptual point of view. The often applied reported procedure in the literature of (repeatedly) dividing a dataset randomly into a calibration and test set must be applied with care. It can only be justified when there is no systematic stratification of the objects that will affect the validated estimates or figures of merits such as RMSE or R(2). The various levels of validation may, typically, be repeatability, reproducibility, and instrument and raw material variation. Examples of how one data set can be validated across this background information illustrate that it will affect the figures of merits as well as the dimensionality of the models. Even more important is the robustness of the models for predicting future samples. Another aspect that is brought to attention is validation in terms of the overall conclusions when observing a specific system. One example is to apply several methods for finding the signif...

Meat Science, 2009

The study aimed at examining the effects of freezing of raw materials, holding time for fresh raw... more The study aimed at examining the effects of freezing of raw materials, holding time for fresh raw materials post mortem and addition of 0.5-1.0% NaCl on the colour of ground beef under low oxygen (O 2 ) modified atmosphere storage. The samples were exposed to 0.1-3.0% O 2 at 4°C for up to 10 days, and analysed for O 2 concentrations, instrumental and visual colour. Residual O 2 in the headspace of the packages oxidizes myoglobin and discolours the meat. Meat may have the ability to scavenge residual O 2 , and ground beef differs from intact muscles by having a much higher capacity for O 2 consumption. In this experiment, the use of frozen/thawed raw materials and addition of NaCl both decreased the rate of O 2 consumption and increased discolouration. Using raw materials from 2 days rather than 7 days post mortem greatly increased the rate of removal of O 2 and improved redness. In low O 2 packaging, ground beef preferably should be stored for at least 2 days in an atmosphere with less than 0.1% residual O 2 to produce a purple pigment predominantly consisting of deoxymyoglobin.