Engineering and Information Technology Collected Works - Research Publications
Permanent URI for this collection
Search Results
1 - 10 of 536
-
ItemEnhancing Predictive Modeling in Emergency Departments(SCITEPRESS - Science and Technology Publications, 2024)Increasing global Emergency Department (ED) visits, exacerbated by COVID-19, has presented multiple challenges in recent years. Electronic Health Records (EHRs) as comprehensive digital repositories of patient health information offer a pathway to construct prediction systems to address these issues. However, the heterogeneity of EHRs complicates accurate predictions. A notable challenge is the prevalence of high-cardinality nominal features (NFs) in EHRs. Due to their numerous distinct values, these features are often excluded from the analysis, risking information loss, reduced accuracy, and interpretability. This study proposes a framework, integrating a preprocessing technique with target encoding (TE-PrepNet) into machine learning (ML) models to address challenges of NFs from MIMIC-IV-ED. We evaluate performance of TE-PrepNet in two specific ED-based prediction tasks: triage-based hospital admissions and ED reattendance within 72 hours at discharge time. Incorporating three NFs, our approach demonstrates improvements compared to the baseline and outperforms previous research that overlooked NFs. Random forest model with TE-PrepNet in the prediction of hospitalisation achieved an AUROC of 0.8458, compared to the baseline AUROC of 0.7520. For the prediction of ED reattendance within 72 hours, the utilisation of XGBoost yielded an improvement, attaining an AUROC of 0.6975, outperforming the baseline AUROC of 0.6166.
-
ItemDesigning zero-emissions containerized last-mile delivery systems: A case study for Melbourne(Elsevier BV, 2024-02)This research investigates the benefits of using swap containers in Two-Echelon (2E) urban delivery systems, which extend beyond the reduced handling costs and processing time in van-bike delivery systems. By drawing on the success of standardized freight containers in the international shipping industry, swap containers can be used to substitute low-capacity vehicles in the first delivery echelon with large-capacity vehicles such as buses or trams. Standardization of swap containers can also encourage collaboration and bring economies of scale. The study proposes a 2E Capacitated Vehicle Routing Problem (2E-CVRP) and a modified multi-start heuristic solution algorithm to analyze the impact of (1) container standardization, (2) large-scale shipping of containers overnight with on-street and high-capacity public vehicles, and (3) decentralized deployment of satellites in Melbourne. Results indicate that standardization can stimulate collaboration and reduce the required bike fleet by 8 %. Shipping containers by overnight tram services can reduce total delivery costs by up to 25 % and eliminate 190 km of daily van travel distances. Using car parking spaces as storage satellites can decrease operational costs by 8 % and travel distances by 27.4 %.
-
ItemProbabilistic hesitant fuzzy multiple criteria decision-making with triangular norm based similarity and entropy measures(Elsevier, 2024-06-01)Existing probabilistic hesitant fuzzy set (PHFS) measures are constructed using two information measures: hesitancy and unwrapped probabilities. We argue that unifying these semantic terms in PHFS information theory is not logical. We introduce a new class of information measures for PHFSs, which address the logical wrapping of hesitant fuzzy sets (HFS) and probability. We propose several similarity measures for these sets that use the Triangular norm operator. We consider the relationship between measures of entropy and similarity and represent the axiomatic definition of PHFS entropy measures. Finally, we use case studies to demonstrate applications of these information measures. We describe two multiple-criteria decision-making algorithms. The last step is devoted to PHFS ranking procedures: one based on the score function of alternatives and the other based on the relative closeness of alternatives. This contribution describes new information measures and uses case studies to illustrate how they can be applied to decision-making processes.
-
ItemNo Preview AvailableEditorial: Special issue on operations research and machine learning(Taylor and Francis Group, 2024)Many machine learning techniques work through optimizing specific objective functions. Supervised learning techniques are to minimize the prediction error such as mean square error (MSE) and misclassification rate, or maximize the conditional likelihood, posterior probability, etc. Unsupervised learning techniques usually group instances into clusters in a way that instances within each group are optimally similar while they are distant from instances in other groups. In reinforcement learning, the goal of an agent is to maximize its cumulative reward. However, there is still room to exploit optimization and operations research (OR) in machine learning, and vice versa.
-
ItemReducing hydraulic erosion of surficial sand layer by inoculation of cyanobacteria(ICE PUBLISHING, 2022-08)Biological approaches have captured the attention of researchers regarding the beneficial effects of cyanobacteria inoculation in improving surficial soil stability. However, a gap exists in the literature regarding the impact of inoculation by individual cyanobacteria on stability of sand under intense surface-water erosion. This study assesses the improvements achieved in erosion resistance for biological soil crust (BC) formed on medium–coarse silica sand. Specimen groups were inoculated with Nostoc sp. and Calothrix sp., incubated for 32- or 48 day periods and then tested using an erosion function apparatus (EFA), investigating a wide range of flow velocities (hydraulic shear stresses). The significance of BC attachment to (or detachment from) the specimen container sidewall was also investigated in the EFA testing. Compared with untreated sand, inoculated specimens had a significantly greater erosion resistance that increased with the incubation period, with Nostoc inoculum producing greater reductions in erodibility coefficients (45–75%) compared with Calothrix (16–67%). Contrasting bond structures introduced by Nostoc and Calothrix are highlighted by scanning electron microscopy images that showed long Nostoc filaments were entangled more strongly in sand pore voids compared with short Calothrix filaments. In conclusion, this study supports the idea of using cyanobacteria inoculation as an eco-friendly, cost-benefit and effective technique for mitigating land degradation.
-
ItemReducing the Erodibility of Sandy Soils Engineered by Cyanobacteria Inoculation: A Laboratory Investigation(MDPI AG, 2023-02)Windblown and water-induced erosion cause substantial soil losses worldwide, especially for drylands. Any sustainable management program that increases soil organic matter and improves the stability of the crustal layer could considerably enhance soil productivity and the preservation of erosion-prone land. This paper presents a laboratory investigation of cyanobacteria-inoculated medium sand and fine sand soils studied for severe runoff conditions that were simulated using an erosion function apparatus (EFA). Loosely deposited sand specimens prepared by air-pluviation were inoculated with a single native filamentous-cyanobacterium strain (investigating both Nostoc sp. and Calothrix sp.) and then incubated under high exposure to white light for 32- or 48-day periods. Well-developed bio-crusts were produced on the specimens’ top surface that achieved substantial improvements in erosion resistance, as was demonstrated for a wide range of hydraulic shear stress investigated using EFA experiments. Relative improvements in hydraulic erosion resistance were explained in terms of the nature of the cyanobacteria-developed microstructures (cyanobacteria filament infiltration of pore-void spaces and exopolysaccharide excretion), as were observed by scanning electron microscope examinations. The developed microstructure depended on the cyanobacterium strain employed and the nominal pore-void sizes that are related to the sand gradation and density state. The encouraging findings of this experimental investigation suggest a tailored approach (i.e., employing a suitable native cyanobacterium strain chosen for its compatibility with the soil’s physical properties) could lay the basis for developing a novel technology for soil protection.
-
ItemPerformance of a right-triangle stilling basin: a laboratory investigation(IWA Publishing, 2023-09)One of the most used hydraulic structures for energy dissipation of supercritical flow is the hydraulic jump stilling basin. From dimensional analysis, the sequent flow depth ratio of a hydraulic jump over the right-triangle basin is derived as a function of the inflow Froude number and relative length of the basin front. The proposed structure stabilized the hydraulic jump at the toe of the chute spillway and hydraulic jump characteristics were investigated for the Froude number ranging from 4.4 < F1 < 7. The results obtained from both numerical and experimental simulations yielded increased efficiency in the energy dissipation performance of this novel design. The modeling showed the formation of two large recirculation regions at the jump roller and jump bed at the beginning of the downstream channel, which resulted in intense energy dissipation in the right-triangle basin. The relative energy loss is approximately 37% higher for relative basin front lengths larger than three compared to the classic jump. Practitioners and academia on the usefulness of a right-triangle basin for hydraulic purposes and further experimental tests are needed to estimate the scalability and cost–benefit of this modified system for implementation.
-
ItemTissue-associated and vertically transmitted bacterial symbiont in the coral Pocillopora acuta(OXFORD UNIV PRESS, 2024-01-08)Coral microhabitats are colonized by a myriad of microorganisms, including diverse bacteria which are essential for host functioning and survival. However, the location, transmission, and functions of individual bacterial species living inside the coral tissues remain poorly studied. Here, we show that a previously undescribed bacterial symbiont of the coral Pocillopora acuta forms cell-associated microbial aggregates (CAMAs) within the mesenterial filaments. CAMAs were found in both adults and larval offspring, suggesting vertical transmission. In situ laser capture microdissection of CAMAs followed by 16S rRNA gene amplicon sequencing and shotgun metagenomics produced a near complete metagenome-assembled genome. We subsequently cultured the CAMA bacteria from Pocillopora acuta colonies, and sequenced and assembled their genomes. Phylogenetic analyses showed that the CAMA bacteria belong to an undescribed Endozoicomonadaceae genus and species, which we propose to name Candidatus Sororendozoicomonas aggregata gen. nov sp. nov. Metabolic pathway reconstruction from its genome sequence suggests this species can synthesize most amino acids, several B vitamins, and antioxidants, and participate in carbon cycling and prey digestion, which may be beneficial to its coral hosts. This study provides detailed insights into a new member of the widespread Endozoicomonadaceae family, thereby improving our understanding of coral holobiont functioning. Vertically transmitted, tissue-associated bacteria, such as Sororendozoicomonas aggregata may be key candidates for the development of microbiome manipulation approaches with long-term positive effects on the coral host.
-
ItemInoculation with Roseovarius increases thermal tolerance of the coral photosymbiont, Breviolum minutum(FRONTIERS MEDIA SA, 2023-08-10)Coral reefs are diverse marine ecosystems that have tremendous ecological and cultural value and support more than 25% of eukaryote marine biodiversity. Increased ocean temperatures and light intensity trigger coral bleaching, the breakdown of the relationship between corals and their photosymbionts, dinoflagellates of the family Symbiodiniaceae. This leaves corals without their primary energy source, thereby leading to starvation and, often, death. Coral bleaching is hypothesized to occur due to an overproduction of reactive oxygen species (ROS) by Symbiodiniaceae, which subsequently accumulate in coral tissues. Bacterial probiotics have been proposed as an approach to mitigate coral bleaching, by reducing ROS levels in the coral holobiont through bacterial antioxidant production. Both corals and Symbiodiniaceae are known to associate with bacteria. However, the Symbiodiniaceae-bacteria relationship, and its impact on Symbiodiniaceae thermal tolerance, remains a poorly studied area. In this study, cultured Symbiodiniaceae of the species Breviolum minutum were treated with antibiotics to reduce their bacterial load. The cultures were subsequently inoculated with bacterial isolates from the genus Roseovarius that were isolated from the same B. minutum culture and showed either high or low ROS-scavenging abilities. The B. minutum cultures were then exposed to experimental heat stress for 16 days, and their health was monitored through measurements of cell density and photochemical efficiency of photosystem II. It was found that B. minutum inoculated with Roseovarius with higher ROS-scavenging abilities showed greater cell growth at elevated temperatures, compared to cultures inoculated with a Roseovarius strain with lower ROS-scavenging abilities. This suggests that Roseovarius may play a role in Symbiodiniaceae fitness at elevated temperatures. Analysis of Symbiodiniaceae-associated bacterial communities through 16S rRNA gene metabarcoding revealed that Roseovarius relative abundance increased in B. minutum cultures following inoculation and with elevated temperature exposure, highlighting the contribution they may have in shielding B. minutum from thermal stress, although other bacterial community changes may have also contributed to these observations. This study begins to unpick the relationship between Symbiodiniaceae and their bacteria and opens the door for the use of Symbiodiniaceae-associated bacteria in coral reef conservation approaches.
-
ItemOptimised deep k-nearest neighbour's based diabetic retinopathy diagnosis(ODeep-NN) using retinal images(SPRINGER, 2024-03-09)Diabetes mellitus has been regarded as one of the prime health issues in present days, which can often lead to diabetic retinopathy, a complication of the disease that affects the eyes, causing loss of vision. For precisely detecting the condition's existence, clinicians are required to recognise the presence of lesions in colour fundus images, making it an arduous and time-consuming task. To deal with this problem, a lot of work has been undertaken to develop deep learning-based computer-aided diagnosis systems that assist clinicians in making accurate diagnoses of the diseases in medical images. Contrariwise, the basic operations involved in deep learning models lead to the extraction of a bulky set of features, further taking a long period of training to predict the existence of the disease. For effective execution of these models, feature selection becomes an important task that aids in selecting the most appropriate features, with an aim to increase the classification accuracy. This research presents an optimised deep k-nearest neighbours'-based pipeline model in a bid to amalgamate the feature extraction capability of deep learning models with nature-inspired metaheuristic algorithms, further using k-nearest neighbour algorithm for classification. The proposed model attains an accuracy of 97.67 and 98.05% on two different datasets considered, outperforming Resnet50 and AlexNet deep learning models. Additionally, the experimental results also portray an analysis of five different nature-inspired metaheuristic algorithms, considered for feature selection on the basis of various evaluation parameters.