Gervase Tuxworth
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Gervase Tuxworth has a total of 48 co-authors. Here are the top co-authors by number of shared publications:
Adrian Meedeniya
Eskitis Institute for Drug Discovery, Griffith University, Brisbane, QLD, Australia
3 shared publications- Alan Mackay-Sim
Griffith Institute for Drug Discovery, Griffith University, Brisbane, Qld 4111, Australia
3 shared publications - Brenton Cavanagh
Cellular and Molecular Imaging Core, RCSI University of Medicine and Health Sciences, Dublin, Ireland
2 shared publications Michael Blumenstein
Griffith University
2 shared publications- René Hexel
School of Information and Communication Technology (ICT), Griffith University, Brisbane, Australia
2 shared publications - Zhongyi Zhang
School of Information and Communication Technology, Griffith University, Nathan, QLD 4111, Australia
2 shared publications
Latest Publications
- Open Access
- 31 August 2026
- Published by Elsevier in SSRN Electronic Journal
A Review of Portability and Software Reuse in Safety-Critical Systems
In safety-critical embedded systems, migrating software to a new processor, real-time operating system, or hardware board costs far more than porting source code. A major part of the work is rebuilding the evidence that justified the system's safety: worst-case execution time (WCET) analyses, schedulability arguments, diagnostic-coverage claims, and safety cases. Practitioners face this cost repeatedly, yet the literature we identified contains no prior review examining what determines whether such evidence carries over after a platform change. We define assurance portability through three categories: evidence that remains valid without modification, that can be regenerated at bounded cost, or that must be produced anew. Evaluating six families of academic approaches alongside industrial ecosystems, we find that most of those targeting code movement achieve high functional portability but leave assurance portability low; no surveyed approach achieves both high integrity and broad portability. We trace this gap to portability boundaries that standardise function signatures but leave timing bounds, blocking guarantees, interference characteristics, and fault-reporting semantics unspecified. Six cross-cutting patterns, organised around this underspecification, show how the gap propagates across portability strategies. We argue that portability boundaries must be specified as behavioural contracts, making these obligations explicit before migration.
- 31 August 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessA Review of Portability and Software Reuse in Safety-Critical Systems
In safety-critical embedded systems, migrating software to a new processor, real-time operating system, or hardware board costs far more than porting source code. A major part of the work is rebuilding the evidence that justified the system's safety: worst-case execution time (WCET) analyses, schedulability arguments, diagnostic-coverage claims, and safety cases. Practitioners face this cost repeatedly, yet the literature we identified contains no prior review examining what determines whether such evidence carries over after a platform change. We define assurance portability through three categories: evidence that remains valid without modification, that can be regenerated at bounded cost, or that must be produced anew. Evaluating six families of academic approaches alongside industrial ecosystems, we find that most of those targeting code movement achieve high functional portability but leave assurance portability low; no surveyed approach achieves both high integrity and broad portability. We trace this gap to portability boundaries that standardise function signatures but leave timing bounds, blocking guarantees, interference characteristics, and fault-reporting semantics unspecified. Six cross-cutting patterns, organised around this underspecification, show how the gap propagates across portability strategies. We argue that portability boundaries must be specified as behavioural contracts, making these obligations explicit before migration.
- Open Access
- 16 October 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-14
Using digital archaeology and machine learning to determine sex in finger flutings
One of the earliest and most enigmatic forms of rock art are finger flutings and previous methods of studying them relied on biometric finger ratios from modern populations to make assumptions about the people who left the flutings, which is theoretically and methodologically problematic. This work is a proof-of-concept for a paradigm shift away from error-prone human measurements and controversial theories to computational digital archaeology methods for an innovative experimental design using a tactile, virtual, and machine learning approach. We propose a digital archaeology experiment using a tactile and virtual approach based on multiple samples from 96 participants. We trained a machine learning model on the known data to determine the sex of the person who made the fluting. While the virtual dataset did not provide sufficiently distinct features for reliable sex classification, the tactile experiment results showed potential for the identification of the sex of fluting artists, but more samples are needed to make any generalization. The significant contribution of this study is the development of a foundational set of methods and materials. We provide a novel digital archaeology approach for data creation, data collection, and analysis that makes the experiment replicable, scalable, and quantifiable.
- 16 October 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-14
Open AccessUsing digital archaeology and machine learning to determine sex in finger flutings
One of the earliest and most enigmatic forms of rock art are finger flutings and previous methods of studying them relied on biometric finger ratios from modern populations to make assumptions about the people who left the flutings, which is theoretically and methodologically problematic. This work is a proof-of-concept for a paradigm shift away from error-prone human measurements and controversial theories to computational digital archaeology methods for an innovative experimental design using a tactile, virtual, and machine learning approach. We propose a digital archaeology experiment using a tactile and virtual approach based on multiple samples from 96 participants. We trained a machine learning model on the known data to determine the sex of the person who made the fluting. While the virtual dataset did not provide sufficiently distinct features for reliable sex classification, the tactile experiment results showed potential for the identification of the sex of fluting artists, but more samples are needed to make any generalization. The significant contribution of this study is the development of a foundational set of methods and materials. We provide a novel digital archaeology approach for data creation, data collection, and analysis that makes the experiment replicable, scalable, and quantifiable.
- 1 September 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (5)
Multicenter Validation of Automated Segmentation and Composition Analysis of Lumbar Paraspinal Muscles Using Multisequence MRI
A deep learning model was developed to segment the lumbar paraspinal muscles and quantify muscle volumes across multisequence MR images and measure fat infiltration ratios on Dixon images, demonstrating comparable performance to expert manual measurements.
- 1 September 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (5)
Multicenter Validation of Automated Segmentation and Composition Analysis of Lumbar Paraspinal Muscles Using Multisequence MRI
A deep learning model was developed to segment the lumbar paraspinal muscles and quantify muscle volumes across multisequence MR images and measure fat infiltration ratios on Dixon images, demonstrating comparable performance to expert manual measurements.
- Open Access
- 10 December 2024
- Published by Association for Computing Machinery (ACM) in Journal on Computing and Cultural Heritage
- Vol. 17 (4) , 1-19
Digitising the Deep Past: Machine Learning for Rock Art Motif Classification in an Educational Citizen Science Application
Digitising the Deep Past (DDP) is an interdisciplinary project based at Griffith University, Australia, that innovates in three areas: Indigenous cultural heritage, Indigenous education, and Machine Learning (ML) and Artificial Intelligence (AI). The project investigates the use of a purpose-built citizen science application that engages Indigenous youth in educational exercises grounded in local cultural heritage, specifically rock art, making learning more engaging and exposing them to digital technologies. Furthermore, ML models trained with the data gathered through these educational activities can then assist with classifying new rock art images and assisting rangers and archaeologists with site archiving and conservation efforts. This article discusses the project’s significance in enhancing Indigenous science and technology education and outlines its results in using ML for rock art classification. Adopting deep learning in rock art classification offers a compelling avenue for the automated analysis and interpretation of heritage objects and places. However, training deep neural networks from scratch often requires enormous datasets and computational resources, posing challenges for domain-specific applications with smaller datasets. With a dataset comprising approximately 3,100 labeled rock art images, we evaluated various tools within the transfer learning toolbox using three prominent pre-trained architectures: VGG19, ResNet50 and EfficientNet V2 S. Through the collaborative efforts of Indigenous students and ML, we demonstrate that even with limited training resources, using transfer learning to re-purpose an existing model can achieve motif classification Top 1 accuracy of 79.76% and Top 5 of 94.56%. The project ran from 2021 to 2023, including 3 week-long sessions with students of Laura State School to trial the citizen science app and the evaluation, development and refinement of the ML models. The DDP project not only serves as a beacon for community-centric research but also forges a new frontier in integrating Indigenous cultural heritage with modern technology. The impact reaches beyond academia, directly enriching the educational experience for Indigenous students in Laura and equipping local rangers and archaeologists with advanced tools for rock art conservation.
- 10 December 2024
- Published by Association for Computing Machinery (ACM) in Journal on Computing and Cultural Heritage
- Vol. 17 (4) , 1-19
Open AccessDigitising the Deep Past: Machine Learning for Rock Art Motif Classification in an Educational Citizen Science Application
Digitising the Deep Past (DDP) is an interdisciplinary project based at Griffith University, Australia, that innovates in three areas: Indigenous cultural heritage, Indigenous education, and Machine Learning (ML) and Artificial Intelligence (AI). The project investigates the use of a purpose-built citizen science application that engages Indigenous youth in educational exercises grounded in local cultural heritage, specifically rock art, making learning more engaging and exposing them to digital technologies. Furthermore, ML models trained with the data gathered through these educational activities can then assist with classifying new rock art images and assisting rangers and archaeologists with site archiving and conservation efforts. This article discusses the project’s significance in enhancing Indigenous science and technology education and outlines its results in using ML for rock art classification. Adopting deep learning in rock art classification offers a compelling avenue for the automated analysis and interpretation of heritage objects and places. However, training deep neural networks from scratch often requires enormous datasets and computational resources, posing challenges for domain-specific applications with smaller datasets. With a dataset comprising approximately 3,100 labeled rock art images, we evaluated various tools within the transfer learning toolbox using three prominent pre-trained architectures: VGG19, ResNet50 and EfficientNet V2 S. Through the collaborative efforts of Indigenous students and ML, we demonstrate that even with limited training resources, using transfer learning to re-purpose an existing model can achieve motif classification Top 1 accuracy of 79.76% and Top 5 of 94.56%. The project ran from 2021 to 2023, including 3 week-long sessions with students of Laura State School to trial the citizen science app and the evaluation, development and refinement of the ML models. The DDP project not only serves as a beacon for community-centric research but also forges a new frontier in integrating Indigenous cultural heritage with modern technology. The impact reaches beyond academia, directly enriching the educational experience for Indigenous students in Laura and equipping local rangers and archaeologists with advanced tools for rock art conservation.
- Open Access
- 30 August 2024
- Published by Elsevier in Journal of Agriculture and Food Research
- Vol. 18
Early detection of sugarcane smut and mosaic diseases via hyperspectral imaging and spectral-spatial attention deep neural networks
- 30 August 2024
- Published by Elsevier in Journal of Agriculture and Food Research
- Vol. 18
Open AccessEarly detection of sugarcane smut and mosaic diseases via hyperspectral imaging and spectral-spatial attention deep neural networks
- Open Access
- 24 June 2022
- Published by IGI Global Scientific Publishing
- p. 151-161
Socially Responsible, Transformative Leadership for STEM Graduates
As socially responsible organizations, the role of universities must extend beyond knowledge generation to positive transformation in society. There is a growing recognition of the importance of nurturing socially responsible leadership in STEM graduates but an economic, knowledge-based perspective on leadership of STEM graduates persists. The purpose of the chapter is to direct attention to the necessity of socially responsible leadership development as a focal point in STEM disciplines and to fuel the discussion about how this might be achieved. Beyond providing conditions for learning of discipline knowledge and skills, educators should be influential leaders, guiding and mentoring their students holistically. The chapter serves to prompt reflection on the necessity of socially responsible leadership development as a focal point in STEM disciplines and to fuel discussion about conditions conducive to its development.
- 24 June 2022
- Published by IGI Global Scientific Publishing
- p. 151-161
Open AccessSocially Responsible, Transformative Leadership for STEM Graduates
As socially responsible organizations, the role of universities must extend beyond knowledge generation to positive transformation in society. There is a growing recognition of the importance of nurturing socially responsible leadership in STEM graduates but an economic, knowledge-based perspective on leadership of STEM graduates persists. The purpose of the chapter is to direct attention to the necessity of socially responsible leadership development as a focal point in STEM disciplines and to fuel the discussion about how this might be achieved. Beyond providing conditions for learning of discipline knowledge and skills, educators should be influential leaders, guiding and mentoring their students holistically. The chapter serves to prompt reflection on the necessity of socially responsible leadership development as a focal point in STEM disciplines and to fuel discussion about conditions conducive to its development.
- 17 February 2022
- Published by Elsevier in Journal of Space Safety Engineering
- Vol. 9 (2) , 257-262
Dynamic verification of satellite systems using Ilities
- 17 February 2022
- Published by Elsevier in Journal of Space Safety Engineering
- Vol. 9 (2) , 257-262
Dynamic verification of satellite systems using Ilities
- 1 September 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3437-3441
Insect Classification Using Squeeze-and-Excitation and Attention Modules - a Benchmark Study
Insect recognition at the species level is an active research field with a variety of applications. With the advancement of convolutional neural networks an automatic fine-grained image classifier has displayed encouraging performance. Despite these recent advances, differentiating images at the species level is still a challenge. To address the problems arising from insect-specific peculiarities, this paper presents a novel network that consists of squeeze-and-excitation modules and attention modules, enabling the network to focus on more informative and differentiating features with a limited number of training iterations and a small dataset. The proposed model is trained on an insect dataset collected from Atlas of Living Australia. The results reveal that the integrated model achieves higher accuracy than several alternative methods on the introduced insect dataset.
- 1 September 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3437-3441
Insect Classification Using Squeeze-and-Excitation and Attention Modules - a Benchmark Study
Insect recognition at the species level is an active research field with a variety of applications. With the advancement of convolutional neural networks an automatic fine-grained image classifier has displayed encouraging performance. Despite these recent advances, differentiating images at the species level is still a challenge. To address the problems arising from insect-specific peculiarities, this paper presents a novel network that consists of squeeze-and-excitation modules and attention modules, enabling the network to focus on more informative and differentiating features with a limited number of training iterations and a small dataset. The proposed model is trained on an insect dataset collected from Atlas of Living Australia. The results reveal that the integrated model achieves higher accuracy than several alternative methods on the introduced insect dataset.
- 30 August 2017
- Published by World Scientific Pub Co Pte Ltd
- p. 97-105
MULTISENSORY GUIDANCE OF GOAL-ORIENTED BEHAVIOUR OF LEGGED ROBOTS
- 30 August 2017
- Published by World Scientific Pub Co Pte Ltd
- p. 97-105
MULTISENSORY GUIDANCE OF GOAL-ORIENTED BEHAVIOUR OF LEGGED ROBOTS