Daniel Patrick Killeen
Research Subjects
Daniel Patrick Killeen has a total of 62 co-authors. Here are the top co-authors by number of shared publications:
- Keith Christopher Gordon
Department of Chemistry University of Otago Dunedin New Zealand
16 shared publications - Mengjie Zhang
Centre for Data Science and Artificial Intelligence & School of Engineering and Computer Science, Victoria University of Wellington, New Zealand
9 shared publications - Bing Xue
Centre for Data Science and Artificial Intelligence & School of Engineering and Computer Science, Victoria University of Wellington, New Zealand
7 shared publications - Nigel Brian Perry
Bioeconomy Science Institute Maiangi Taiao Department of Chemistry University of Otago Dunedin New Zealand
7 shared publications Asli Card
The New Zealand Institute for Plant and Food Research Limited Nelson New Zealand
3 shared publicationsElaine J. Burgess
Department of Chemistry, The New Zealand Institute for Plant and Food Research Limited University of Otago P.O. Box 56, Dunedin 9054 New Zealand
3 shared publications
Latest Publications
- Open Access
- 8 May 2026
- Published by Wiley in Journal of the Royal Society of New Zealand
- Vol. 56 (3) , e70051
Symbolically Regressing Fish Biomass Spectral Data: A Linear Genetic Programming Method With Tunable Primitives
Machine learning techniques play an important role in analyzing spectral data. Fish biomass spectral data is useful in fish production as it carries key chemical properties of fish meat. However, it is challenging for existing machine learning techniques to comprehensively discover hidden patterns from fish biomass spectral data because the spectral data often have considerable noise, while the training data are quite limited. To better analyze fish biomass spectral data, we modeled it as a symbolic regression problem and solved it by a linear genetic programming method with newly proposed tunable primitives. In the symbolic regression problem, linear genetic programming automatically synthesized regression models based on the given primitives and training data. The tunable primitives further improved the approximation ability of the regression models by tuning their inherent coefficients. Our empirical results from over ten fish biomass targets showed that our proposed method using tunable primitives improved the overall performance of fish biomass composition prediction. The synthesized regression models are compact and have good interpretability, allowing us to highlight useful features over the spectrum. Our further investigation verified the good generality of the proposed method across various spectral data treatments and other symbolic regression problems.
- 8 May 2026
- Published by Wiley in Journal of the Royal Society of New Zealand
- Vol. 56 (3) , e70051
Open AccessSymbolically Regressing Fish Biomass Spectral Data: A Linear Genetic Programming Method With Tunable Primitives
Machine learning techniques play an important role in analyzing spectral data. Fish biomass spectral data is useful in fish production as it carries key chemical properties of fish meat. However, it is challenging for existing machine learning techniques to comprehensively discover hidden patterns from fish biomass spectral data because the spectral data often have considerable noise, while the training data are quite limited. To better analyze fish biomass spectral data, we modeled it as a symbolic regression problem and solved it by a linear genetic programming method with newly proposed tunable primitives. In the symbolic regression problem, linear genetic programming automatically synthesized regression models based on the given primitives and training data. The tunable primitives further improved the approximation ability of the regression models by tuning their inherent coefficients. Our empirical results from over ten fish biomass targets showed that our proposed method using tunable primitives improved the overall performance of fish biomass composition prediction. The synthesized regression models are compact and have good interpretability, allowing us to highlight useful features over the spectrum. Our further investigation verified the good generality of the proposed method across various spectral data treatments and other symbolic regression problems.
- 1 April 2026
- Published by Springer Nature
- p. 585-599
From Catch to Product: Machine Learning Driven Spectral Analysis for Fish Processing Line Allocation
- 1 April 2026
- Published by Springer Nature
- p. 585-599
From Catch to Product: Machine Learning Driven Spectral Analysis for Fish Processing Line Allocation
- Open Access
Advancing the Analysis of Fatty Acid Composition in Animal-Based Marine Oils Through the Integration of Raman and IR Spectroscopy with Chemometrics
This study investigated the use of Raman and IR spectroscopy, individually and combined, for quantifying fatty acid methyl ester (FAME) profiles in animal-based marine oils and potential adulterants (palm oil, ω-3 concentrates in ethyl ester, and generic fish oil). FAME profiles are important for assessing oil quality, conventionally determined via gas chromatography. This study aimed to provide a rapid and non-destructive alternative. The study utilized Partial Least Squares Regression (PLSR) alongside Raman (r2 = 0.94; RMSEP = 2.4%) and IR spectroscopy (r2 = 0.95; RMSEP = 2.3%), demonstrating similar ω-3 fatty acid predictions. Fusion of IR and Raman spectroscopic datasets improved ω-3 fatty acid (r2 = 0.96; RMSEP = 1.9%), polyunsaturated fatty acids (PUFA) (r2 = 0.83; RMSEP = 4.0%), and saturated fatty acids (SFA) (r2 = 0.79; RMSEP = 4.1%) quantification. The study highlights that fusion of IR and Raman spectroscopic datasets presents a promising avenue for non-destructive fatty acid composition assessment.
Open AccessAdvancing the Analysis of Fatty Acid Composition in Animal-Based Marine Oils Through the Integration of Raman and IR Spectroscopy with Chemometrics
This study investigated the use of Raman and IR spectroscopy, individually and combined, for quantifying fatty acid methyl ester (FAME) profiles in animal-based marine oils and potential adulterants (palm oil, ω-3 concentrates in ethyl ester, and generic fish oil). FAME profiles are important for assessing oil quality, conventionally determined via gas chromatography. This study aimed to provide a rapid and non-destructive alternative. The study utilized Partial Least Squares Regression (PLSR) alongside Raman (r2 = 0.94; RMSEP = 2.4%) and IR spectroscopy (r2 = 0.95; RMSEP = 2.3%), demonstrating similar ω-3 fatty acid predictions. Fusion of IR and Raman spectroscopic datasets improved ω-3 fatty acid (r2 = 0.96; RMSEP = 1.9%), polyunsaturated fatty acids (PUFA) (r2 = 0.83; RMSEP = 4.0%), and saturated fatty acids (SFA) (r2 = 0.79; RMSEP = 4.1%) quantification. The study highlights that fusion of IR and Raman spectroscopic datasets presents a promising avenue for non-destructive fatty acid composition assessment.
- Open Access
- 23 December 2025
- Published by Elsevier in Applied Soft Computing
- Vol. 189
Evolutionary multitasking for multi-objective feature selection in fish chemical analysis
- 23 December 2025
- Published by Elsevier in Applied Soft Computing
- Vol. 189
Open AccessEvolutionary multitasking for multi-objective feature selection in fish chemical analysis
- Open Access
- 14 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 939-942
Designing New Data Augmentation Functions for Fish Spectral Data by Genetic Programming
- 14 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 939-942
Open AccessDesigning New Data Augmentation Functions for Fish Spectral Data by Genetic Programming
- 24 June 2025
- Published by Springer Nature
- p. 29-43
Machine Learning for Raman Spectroscopy-Based Cyber-Marine Fish Biochemical Composition Analysis
- 24 June 2025
- Published by Springer Nature
- p. 29-43
Machine Learning for Raman Spectroscopy-Based Cyber-Marine Fish Biochemical Composition Analysis
- Open Access
- 12 June 2025
- Published by Springer Nature in Intelligent Marine Technology and Systems
- Vol. 3 (1) , 1-15
Hook, line, and spectra: machine learning for fish species identification and body part classification using rapid evaporative ionization mass spectrometry
Marine biomass composition analysis traditionally requires time-consuming processes and domain expertise. This study demonstrates the effectiveness of rapid evaporative ionization mass spectrometry (REIMS) combined with advanced machine learning (ML) techniques for accurate marine biomass composition determination. Using fish species and body parts as model systems representing diverse biochemical profiles, we investigate various ML methods, including unsupervised pretraining strategies for transformers. The deep learning approaches consistently outperformed traditional machine learning across all tasks. For fish species classification, the pretrained transformer achieved 99.62% accuracy, and for fish body parts classification, the transformer achieved 84.06% accuracy. We further explored the explainability of the best-performing and predominantly black box models using local interpretable model-agnostic explanations and gradient-weighted class activation mapping to identify the important features driving the decisions behind each of the best performing classifiers. REIMS analysis with ML can be an accurate and potentially explainable technique for automated marine biomass composition analysis. Thus, REIMS analysis with ML has potential applications in quality control, product optimization, and food safety monitoring in marine-based industries.
- 12 June 2025
- Published by Springer Nature in Intelligent Marine Technology and Systems
- Vol. 3 (1) , 1-15
Open AccessHook, line, and spectra: machine learning for fish species identification and body part classification using rapid evaporative ionization mass spectrometry
Marine biomass composition analysis traditionally requires time-consuming processes and domain expertise. This study demonstrates the effectiveness of rapid evaporative ionization mass spectrometry (REIMS) combined with advanced machine learning (ML) techniques for accurate marine biomass composition determination. Using fish species and body parts as model systems representing diverse biochemical profiles, we investigate various ML methods, including unsupervised pretraining strategies for transformers. The deep learning approaches consistently outperformed traditional machine learning across all tasks. For fish species classification, the pretrained transformer achieved 99.62% accuracy, and for fish body parts classification, the transformer achieved 84.06% accuracy. We further explored the explainability of the best-performing and predominantly black box models using local interpretable model-agnostic explanations and gradient-weighted class activation mapping to identify the important features driving the decisions behind each of the best performing classifiers. REIMS analysis with ML can be an accurate and potentially explainable technique for automated marine biomass composition analysis. Thus, REIMS analysis with ML has potential applications in quality control, product optimization, and food safety monitoring in marine-based industries.
- Open Access
Symbolically Regressing Fish Biomass Spectral Data: A Linear Genetic Programming Method with Tunable Primitives
Machine learning techniques play an important role in analyzing spectral data. The spectral data of fish biomass is useful in fish production, as it carries many important chemistry properties of fish meat. However, it is challenging for existing machine learning techniques to comprehensively discover hidden patterns from fish biomass spectral data since the spectral data often have a lot of noises while the training data are quite limited. To better analyze fish biomass spectral data, this paper models it as a symbolic regression problem and solves it by a linear genetic programming method with newly proposed tunable primitives. In the symbolic regression problem, linear genetic programming automatically synthesizes regression models based on the given primitives and training data. The tunable primitives further improve the approximation ability of the regression models by tuning their inherent coefficients. Our empirical results over ten fish biomass targets show that the proposed method improves the overall performance of fish biomass composition prediction. The synthesized regression models are compact and have good interpretability, which allow us to highlight useful features over the spectrum. Our further investigation also verifies the good generality of the proposed method across various spectral data treatments and other symbolic regression problems.
Open AccessSymbolically Regressing Fish Biomass Spectral Data: A Linear Genetic Programming Method with Tunable Primitives
Machine learning techniques play an important role in analyzing spectral data. The spectral data of fish biomass is useful in fish production, as it carries many important chemistry properties of fish meat. However, it is challenging for existing machine learning techniques to comprehensively discover hidden patterns from fish biomass spectral data since the spectral data often have a lot of noises while the training data are quite limited. To better analyze fish biomass spectral data, this paper models it as a symbolic regression problem and solves it by a linear genetic programming method with newly proposed tunable primitives. In the symbolic regression problem, linear genetic programming automatically synthesizes regression models based on the given primitives and training data. The tunable primitives further improve the approximation ability of the regression models by tuning their inherent coefficients. Our empirical results over ten fish biomass targets show that the proposed method improves the overall performance of fish biomass composition prediction. The synthesized regression models are compact and have good interpretability, which allow us to highlight useful features over the spectrum. Our further investigation also verifies the good generality of the proposed method across various spectral data treatments and other symbolic regression problems.
- Open Access
Machine Learning for Raman Spectroscopy-based Cyber-Marine Fish Biochemical Composition Analysis
The rapid and accurate detection of biochemical compositions in fish is a crucial real-world task that facilitates optimal utilization and extraction of high-value products in the seafood industry. Raman spectroscopy provides a promising solution for quickly and non-destructively analyzing the biochemical composition of fish by associating Raman spectra with biochemical reference data using machine learning regression models. This paper investigates different regression models to address this task and proposes a new design of Convolutional Neural Networks (CNNs) for jointly predicting water, protein, and lipids yield. To the best of our knowledge, we are the first to conduct a successful study employing CNNs to analyze the biochemical composition of fish based on a very small Raman spectroscopic dataset. Our approach combines a tailored CNN architecture with the comprehensive data preparation procedure, effectively mitigating the challenges posed by extreme data scarcity. The results demonstrate that our CNN can significantly outperform two state-of-the-art CNN models and multiple traditional machine learning models, paving the way for accurate and automated analysis of fish biochemical composition.
Open AccessMachine Learning for Raman Spectroscopy-based Cyber-Marine Fish Biochemical Composition Analysis
The rapid and accurate detection of biochemical compositions in fish is a crucial real-world task that facilitates optimal utilization and extraction of high-value products in the seafood industry. Raman spectroscopy provides a promising solution for quickly and non-destructively analyzing the biochemical composition of fish by associating Raman spectra with biochemical reference data using machine learning regression models. This paper investigates different regression models to address this task and proposes a new design of Convolutional Neural Networks (CNNs) for jointly predicting water, protein, and lipids yield. To the best of our knowledge, we are the first to conduct a successful study employing CNNs to analyze the biochemical composition of fish based on a very small Raman spectroscopic dataset. Our approach combines a tailored CNN architecture with the comprehensive data preparation procedure, effectively mitigating the challenges posed by extreme data scarcity. The results demonstrate that our CNN can significantly outperform two state-of-the-art CNN models and multiple traditional machine learning models, paving the way for accurate and automated analysis of fish biochemical composition.
- Open Access
- 11 October 2023
- Published by Wiley in Journal of the American Oil Chemists' Society
- Vol. 101 (3) , 361-367
Hydrophobic natural deep eutectic solvents for marine lipid extraction
‘Type V' natural deep eutectic solvents (TV NADES) are a novel class of solvent media with many potential ‘green chemistry’ applications. Here, we assess the potential of two TV NADES: menthol:carvacrol (M:C) and menthol:thymol (M:T), and a single terpenoid extraction media carvacrol (C), as media from total lipid extraction of three compositionally diverse, freeze‐dried marine tissues: hoki (Macruronus novaezelandaiae), jack mackerel (Trachurus declivis) and green‐lipped mussel (kuku, Perna canaliculus). Room temperature extraction yields into these media were benchmarked relative to lipid yields from chloroform‐methanol‐water extractions. Relative extraction yields using the M:C and M:T media were (both) 87% for jack mackerel, 85%–87% for hoki, and 43%–47% for green‐lipped mussels. Comprehensive chemical analyses of TV NADES lipid extracts suggested that neutral lipids were extracted in greater yield than phospholipids, and that phosphatidylinositol, lysophospholipids and (in the case of mussels) ceramideaminoethylphosphonates were particularly resistant to extraction into C, M:C and M:T. C—a thymol isomer that, unlike menthol and thymol is a liquid at room temperature—appeared to perform (marginally) better than M:C and M:T systems, implying there may be no advantage to using the TV NADES over a single terpenoid with similar chemistry for total lipid extraction.
- 11 October 2023
- Published by Wiley in Journal of the American Oil Chemists' Society
- Vol. 101 (3) , 361-367
Open AccessHydrophobic natural deep eutectic solvents for marine lipid extraction
‘Type V' natural deep eutectic solvents (TV NADES) are a novel class of solvent media with many potential ‘green chemistry’ applications. Here, we assess the potential of two TV NADES: menthol:carvacrol (M:C) and menthol:thymol (M:T), and a single terpenoid extraction media carvacrol (C), as media from total lipid extraction of three compositionally diverse, freeze‐dried marine tissues: hoki (Macruronus novaezelandaiae), jack mackerel (Trachurus declivis) and green‐lipped mussel (kuku, Perna canaliculus). Room temperature extraction yields into these media were benchmarked relative to lipid yields from chloroform‐methanol‐water extractions. Relative extraction yields using the M:C and M:T media were (both) 87% for jack mackerel, 85%–87% for hoki, and 43%–47% for green‐lipped mussels. Comprehensive chemical analyses of TV NADES lipid extracts suggested that neutral lipids were extracted in greater yield than phospholipids, and that phosphatidylinositol, lysophospholipids and (in the case of mussels) ceramideaminoethylphosphonates were particularly resistant to extraction into C, M:C and M:T. C—a thymol isomer that, unlike menthol and thymol is a liquid at room temperature—appeared to perform (marginally) better than M:C and M:T systems, implying there may be no advantage to using the TV NADES over a single terpenoid with similar chemistry for total lipid extraction.