Article
Biochemistry & Molecular Biology
Johanne Heitmann Solheim, Boris Zimmermann, Valeria Tafintseva, Simona Dzurendova, Volha Shapaval, Achim Kohler
Summary: This paper discusses the application of extended multiplicative signal correction (EMSC) in infrared spectroscopy and its associated techniques. By adding constituent spectra to model known analytes or interferents, EMSC can extract additional information and be used in both regression and segmentation tasks.
Article
Chemistry, Analytical
Xiaoshan Li, Xiaojun Tang, Bin Wang, Youshui Lu, Houqing Chen
Summary: Baseline drift is an important issue in spectral analysis, and most existing methods for baseline correction are not effective in high noise, complex baselines, and overlapping peaks situations. To address these challenges, an adaptive extended Gaussian peak derivative reweighted penalised least squares (agdPLS) method was proposed, which achieved more accurate baseline estimation. Experimental results showed that agdPLS outperformed other methods and was computationally efficient, making it effective for baseline correction in spectra with high noise, complex baselines, and overlapping peaks.
ANALYTICAL METHODS
(2023)
Article
Environmental Sciences
Zijiang Yang, Hisayuki Arakawa
Summary: In this study, a double sliding-window (DSW) method was proposed to estimate the baseline and standard deviation of noise in order to measure microplastics in environmental samples. Compared with two popular methods, DSW method showed better performance in handling spectra of low signal-to-noise ratio and elevated baselines.
MARINE POLLUTION BULLETIN
(2023)
Article
Chemistry, Analytical
Hyeong Geun Yu, Dong Jo Park, Dong Eui Chang, Hyunwoo Nam
Summary: Raman spectroscopy plays a key role in non-contact chemical agent detection, but the performance can be degraded by fluorescence-induced baseline. This study proposes a baseline correction algorithm that effectively removes the baseline and improves detection accuracy, surpassing traditional methods.
Article
Biochemistry & Molecular Biology
Harpreet Kaur, Rainer Kunnemeyer, Andrew McGlone
Summary: In this study, the changes in the water structure of kiwifruit juice with temperature variations were examined using the framework of aquaphotomics. The study focused on the first and second overtone regions of the OH stretch of water. The results showed that the use of external parameter orthogonalisation (EPO) and extended multiple scatter correction (EMSC) pre-processing methods improved the accuracy of predicting the soluble solids concentration (SSC) of kiwifruit juice.
Article
Instruments & Instrumentation
Yunnan Xu, Pang Du, Ryan Senger, John Robertson, James L. Pirkle
Summary: In Raman spectroscopy, baseline correction is a critical step that has been recently improved with procedures relying on asymmetric loss functions. A novel baseline correction procedure called ISREA has been developed, utilizing smoothing splines to estimate the baseline, mimicking asymmetric square root loss, and avoiding direct optimization of a non-convex loss function by iteratively updating prediction errors and refitting baselines. Through extensive numerical experiments, ISREA has been shown to be simple, fast, and capable of yielding consistent and accurate baselines that preserve meaningful Raman peaks.
APPLIED SPECTROSCOPY
(2021)
Article
Chemistry, Analytical
Wenqian Yan, Jiayi Yao, Zilin Yue, Hong Lin, Lei Wang, Kaiqiang Wang, Jinjie Li
Summary: This study investigated the use of Raman spectroscopy combined with multivariate analysis to monitor the freshness of sea bass fillets. PLSR models outperformed PCR and SVMR models in monitoring the TVB-N values of the fillets. The combination of Raman spectroscopy and chemometrics provided a new tool for the non-destructive evaluation of fish quality.
MICROCHEMICAL JOURNAL
(2023)
Article
Environmental Sciences
Hongming Zhang, Xiang Zhou, Zui Tao, Tingting Lv, Jin Wang
Summary: This paper proposes a deep learning method using a one-dimensional U-shape neural network (1D U-Net) to compensate for turbidity interference and obtain water parameters. The experimental results demonstrate that the method achieves good turbidity compensation in sampling real river water.
FRONTIERS IN ENVIRONMENTAL SCIENCE
(2022)
Article
Automation & Control Systems
Jiajin Wei, Chen Zhu, Zhi-Min Zhang, Ping He
Summary: This paper reviews several iteratively reweighted baseline correction methods and proposes a new method called two-stage iteratively reweighted smoothing splines (RWSS) to address the estimation of baselines in complex-structured signals. Simulation studies and real data experiments demonstrate the performance and reliability of the proposed method.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2022)
Article
Physics, Multidisciplinary
Xiang Chen, Mai Hu, Hao Liu, Lu Yao, Zhenyu Xu, Ruifeng Kan
Summary: This paper demonstrates a convenient method for light intensity correction in quartz-enhanced photoacoustic spectroscopy (QEPAS) using photothermal baseline. By minimizing the optical path length for photothermal spectroscopy, the QEPAS system achieves accurate gas concentration measurement and immunity to light intensity variation.
FRONTIERS IN PHYSICS
(2022)
Article
Spectroscopy
Liu Long, Fan Xian-guang, Kang Zhe-ming, Wu Yi, Wang Xin
Summary: Raman spectroscopy is a valuable technology for qualitative and quantitative analysis in various fields. However, baseline drift caused by background fluorescence affects the accuracy of results. Different methods, such as experimental improvements and numerical processing, are used to address this issue.
SPECTROSCOPY AND SPECTRAL ANALYSIS
(2021)
Article
Optics
David R. Smith, Jeffrey J. Field, David G. Winters, Scott R. Domingue, Frauke Rininsland, Daniel J. Kane, Jesse W. Wilson, Randy A. Bartels
Summary: The study presents a method called radio frequency (RF) Doppler Raman spectroscopy for optical amplification of coherent Raman spectroscopy signal. By converting the Doppler frequency shift of a laser probe pulse into periodic timing jitter, amplification of Raman signal outside of focused interaction is enabled. Measurement of timing jitter allows access to lower noise floors than other coherent Raman techniques, opening up the potential to detect very weak Raman signals.
Article
Biochemical Research Methods
Johanne Heitmann Solheim, Ferenc Borondics, Boris Zimmermann, Christophe Sandt, Florian Muthreich, Achim Kohler
Summary: This article introduces an open-source algorithm for fringe correction in infrared spectroscopy and proposes improvements to the Fringe EMSC model. The algorithm achieves more precise fringe frequency estimation through mean centering and window function. It has been validated on two experimental datasets with successful results.
JOURNAL OF BIOPHOTONICS
(2021)
Article
Engineering, Electrical & Electronic
Yuqiang Li, Xinjie Wang, Huijing Yu, Wenli Du
Summary: In this article, a pattern-coupled learning framework considering the local coupling property is proposed for pure spectrum fitting and baseline correction. By characterizing the local sparse structure corresponding to characteristic peaks with wavelength coupling, the proposed method achieves accurate baseline estimation without prior knowledge of spectral structure. The proposed method shows improved performance compared to state-of-the-art methods in three measured NIR datasets, with reduced RMSE and increased R-2 values.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Chemistry, Analytical
Tanveer Ahmed Shaik, Joao L. Lagarto, Enrico Baria, Melis Goktas, Patrick Igoche Onoja, Kerstin G. Blank, Francesco S. Pavone, Juergen Popp, Christoph Krafft, Riccardo Cicchi
Summary: The study demonstrates the potential of nondestructive optical imaging techniques such as SHG, RS, and FLIM in monitoring collagen degradation and assessing mechanical properties in tissues. These techniques can provide label-free monitoring of collagen degradation and rough evaluation of mechanical properties in a nondestructive manner.
ANALYTICAL CHEMISTRY
(2021)
Article
Biochemical Research Methods
Miroslav Kuchta, Sileshi Gizachew Wubshet, Nils Kristian Afseth, Kent-Andre Mardal, Kristian Hovde Liland
Summary: This study investigates the potential of deep neural networks in predicting the future state of enzymatic hydrolysis, described by Fourier-transform infrared spectra of the hydrolysates. Combined with predictions of average molecular weight, this provides a flexible and transparent tool for process monitoring and control.
JOURNAL OF BIOPHOTONICS
(2022)
Article
Chemistry, Analytical
Puneet Mishra, Junli Xu, Kristian Hovde Liland, Thanh Tran
Summary: This paper introduces a novel approach called META-PLS for automatic NIR spectral modeling. The approach utilizes the stepwise nature of PLS algorithm and a weighted randomization test to avoid exhaustive search for preprocessing selection and the optimal number of model components.
ANALYTICA CHIMICA ACTA
(2022)
Article
Automation & Control Systems
Kristian Hovde Liland, Ulf Geir Indahl, Joakim Skogholt, Puneet Mishra
Summary: The article discusses using multilinear partial least squares for analyzing multiway datasets, which allows for building parsimonious models handling various continuous and categorical responses. An advantage in computational speed is achieved by deflating responses and orthogonalising scores.
JOURNAL OF CHEMOMETRICS
(2022)
Article
Materials Science, Ceramics
Kristian Hovde Liland, Roman Svoboda, Giorgio Luciano, Nikita Muravyev
Summary: This study evaluated the performance of several neural network architectures for analyzing complex processes with overlapping reaction mechanisms. The results showed that the CDD and MLP architectures provided accurate estimates of the kinetic parameters in partially overlapping processes, performing on par with traditional methods. However, the accuracy of the neural networks decreased in fully overlapping kinetic processes, but still outperformed traditional approaches.
JOURNAL OF NON-CRYSTALLINE SOLIDS
(2022)
Article
Automation & Control Systems
Puneet Mishra, Kristian Hovde Liland, Ulf Geir Indahl
Summary: A novel unified covariates selection algorithm called SKCovSel is introduced, which is suitable for various data scenarios including single block, multiblock, multiway, and multiple response. The algorithm can be scale and data block order-independent in multiblock scenarios, and multiway mode order-independent in multiway scenarios, depending on user preference. It generalises speed improvements and reformulates the multiway case for proper deflation and feature selection. The method incorporates all popular covariates selection algorithms in the chemometric literature.
JOURNAL OF CHEMOMETRICS
(2022)
Article
Automation & Control Systems
Puneet Mishra, Kristian Hovde Liland
Summary: A new method using iterative re-weighted partial least squares and covariates selection is presented for feature selective modelling in the presence of outliers. The method iteratively down-weights the outlying samples to minimize their influence on the squared covariance estimation for selecting robust features. It is shown that models based on such features outperform those using equal sample weights in terms of prediction accuracy. The method is tested in different scenarios and its performance is demonstrated on a real spectral data set.
JOURNAL OF CHEMOMETRICS
(2023)
Article
Environmental Sciences
Agnieszka Kuras, Maximilian Brell, Kristian Hovde Liland, Ingunn Burud
Summary: Technological innovations and advanced multidisciplinary research have increased the demand for multisensor data fusion in Earth observations, particularly in remote sensing. In this study, we propose a novel approach to multitemporal analysis of urban land cover, inspired by the capabilities of hyperspectral and LiDAR data fusion at the feature level. Our framework utilizes bitemporal datasets and extracts representative endmembers, retrieves abundance maps for segmentation algorithms, and detects urban land cover classes using 2D ResU-Net. We compared segmentation optimization models with and without data augmentation, achieving up to 11% better accuracy after optimization.
Article
Chemistry, Multidisciplinary
Eirik Almklov Magnussen, Boris Zimmermann, Uladzislau Blazhko, Simona Dzurendova, Benjamin Dupuy-Galet, Dana Byrtusova, Florian Muthreich, Valeria Tafintseva, Kristian Hovde Liland, Kristin Tondel, Volha Shapaval, Achim Kohler
Summary: Infrared spectroscopy provides valuable information about the chemical composition, structural properties, and optical properties of intact samples. This study introduces a deep convolutional neural network that utilizes this information to solve full-wave inverse scattering problems and obtain the 3D optical, structural, and chemical properties from infrared spectroscopic measurements of micro-samples.
COMMUNICATIONS CHEMISTRY
(2022)
Article
Chemistry, Analytical
Runar Helin, Ulf Indahl, Oliver Tomic, Kristian Hovde Liland
Summary: This paper introduces a residual modeling scheme using modern neural network architectures, which combines linear and neural network models to achieve good model performance with interpretability. Results show that residual modeling can enhance the performance of a linear model and achieve similar performance and valuable interpretations compared to pure neural network models for complex data modeling. This study is significant and novel in advancing explainable AI and making artificial neural network modeling more transparent.
ANALYTICA CHIMICA ACTA
(2023)
Article
Physics, Applied
M. Vukovic, K. H. Liland, U. G. Indahl, M. Jakovljevic, A. S. Flo, E. Olsen, I. Burud
Summary: Photoluminescence imaging is a promising technique for high-throughput on-site inspection of photovoltaic modules. A noninvasive photoluminescence imaging method has been proposed recently, which acquires images during a current-voltage curve sweep to detect continuously changing photoluminescence signals. An alternative algorithm based on the Pearson correlation coefficient is employed to process the images efficiently and unsupervised, enabling real-time surveillance and detection of functional anomalies. This algorithm is robust to varying solar irradiance and can process photoluminescence signals from multiple asynchronized strings.
JOURNAL OF APPLIED PHYSICS
(2023)
Article
Automation & Control Systems
Puneet Mishra, Kristian Hovde Liland
Summary: A new algorithm for robust multiblock modelling is proposed, combining robust modelling technique with partial least squares and component-wise multiblock partial least squares modelling. The algorithm automatically down-weights outlying observations to minimize their contribution during block-wise partial least squares estimation, resulting in robust modelling that is minimally affected by outliers. The algorithm is demonstrated with tests on simulated and real multiblock data sets in the presence of outlying observations.
JOURNAL OF CHEMOMETRICS
(2023)
Article
Instruments & Instrumentation
Elise L. Gjelsvik, Martin Fossen, Anders Brunsvik, Kristian H. Liland, Kristin Tondel
Summary: Crude oils are complex mixtures with various components, and traditional analytical techniques may not be sufficient for detailed characterization. Fourier transform ion cyclotron resonance mass spectrometry and infrared spectroscopy can be used for crude oil analysis. This study compared different regression models (single block PLSR, multiblock PLSR, and SO-PLSR) to predict crude oil density, and found that combining multiple blocks of data improved the prediction performance.
APPLIED SPECTROSCOPY
(2023)
Article
Automation & Control Systems
Joakim Skogholt, Kristian H. Liland, Tormod Naes, Age K. Smilde, Ulf G. Indahl
Summary: In this paper, a sparse analysis method is proposed, which uses a voting approach to select variables, aiming to obtain more concise models while maintaining information and performance.
JOURNAL OF CHEMOMETRICS
(2023)
Article
Psychology, Multidisciplinary
Frode Moen, Svein Arne Pettersen, Kine Gjertsas, Marte Vatn, Martijn Ravenhorst, Atle Kvalsvoll, Kristian Hovde Liland, Ellen F. Mosleth
Summary: The study investigated the physical loads on game days and the impact of BEMER therapy on sleep among elite women's football players in Norway. Results showed that physical performance peaked on game days compared to training days. Sleep quantity and quality were significantly reduced on game nights due to the increased physical load. The use of BEMER therapy significantly improved sleep, with the most positive effects observed in players who used it for more than 440 hours.
FRONTIERS IN PSYCHOLOGY
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Anna Jenul, Bimal Bhattarai, Kristian Hovde Liland, Lei Jiao, Stefan Schrunner, Cecilia Futsaether, Ole-Christoffer Granmo, Oliver Tomic
Summary: Tabular data with few observations and many features are common in healthcare. Tsetlin Machine, with its rule-based approach, can provide interpretability to medical personnel. However, the presence of noise in healthcare data may hinder its performance. This study shows that intelligent pre-filtering of healthcare measurement data using component-based methods can benefit the performance of Tsetlin Machines, particularly for data sets with high features-to-observations ratios.
2022 INTERNATIONAL SYMPOSIUM ON THE TSETLIN MACHINE (ISTM 2022)
(2022)