Learning Approach Research Articles

Determining the Effect of Fear of Illness and Virus Evaluation and Quality of Life on Diagnosis of Social Phobia in Patients With Chronic Disease: Using Machine Learning Approaches.

While the SARS-CoV-2 pandemic and other epidemics continue, individuals with chronic diseases and those over the age of 60 are most affected by the psychological effects. This research is the first and most crucial study comparing the quality of life, physical activities, fear of disease and virus evaluation, and social phobia in chronic patients and healthy individuals, and modeling the classification of social phobia using the machine learning approach. The quantitative study used STROBE guidelines for the correlational and cross-sectional design. The research questionnaire was designed in four parts: a personal information form, the Liebowitz Social Phobia Scale, the Fear of Illness and Virus Evaluation Scale, and the Quality of Life Scale (EUROHIS-WHOQOL-8). Different algorithms were examined using the machine learning approach to classify social phobia. More participants were reached than the calculated sample size (n = 1068) using simple random sampling, and the final sample size was 1235. Patients with chronic diseases had lower physical activity levels and quality of life scores. Patients with chronic diseases (n=728) had higher Fear of Illness and Virus Evaluation Scale-35 scores and Liebowitz Social Phobia Scale-24 scores compared to healthy participants (n=507) and lower physical activity levels (3.901 ± 3.035) and quality of life scores (29.016 ± 4.782). Two algorithms (K-nearest neighbors and support vector machine algorithm) provided the best performance. In support vector machine algorithm, Fear of Illness and Virus Evaluation Scale-35 was the most critical feature in classifying social phobia. Physical activity level and Liebowitz Social Phobia Scale seem to be positively related in k-nearest neighbors. The model is essential for identifying and understanding social phobia factors in patients with chronic diseases. Support vector machine algorithm is an algorithm that is preferred for identifying patients at risk of fear and will facilitate follow-up when integrated into smartphone applications.

Comparing the effectiveness of multiple text reading and rereading on knowledge retention and metacognitive accuracy

AbstractAlthough learning approaches are designed to enhance individuals' ability to store and retrieve information, not all of them are considered effective. The goal of the present study was to experimentally compare the test performance as well as the accuracy of metacognitive judgements of a multiple text reading group, rereading group, and single reading group in a one-day vs. one-week delayed test (3 × 2 between subject design). A total of 186 psychology students (Mage = 20.76) participated in the experiment focused on reading comprehension, accuracy of metacognitive judgments and knowledge retention. Results of the knowledge test indicate that in the one-day delayed test, multiple text reading and rereading yield similar results: both deliver slightly better results than single reading. In the one-week delayed test, though, multiple text reading yields better results than rereading, and both these reading approaches outperform single reading. Moreover, multiple text reading results in fairly robust knowledge retention with only a slight decrease in scores between the one-day delayed and one-week delayed test. Regarding metacognitive monitoring, judgements of learning in the multiple text reading group remained relatively stable after each reading and participants were underconfident about their knowledge. In the rereading group, judgements of learning increased after each reading and participants were overconfident about their knowledge, especially on the one-day delayed test. These findings have implications for educational practices aimed at enhancing learning outcomes and promoting effective learning.

HRCUNet: Hierarchical Region Contrastive Learning for Segmentation of Breast Tumors in DCE‐MRI

ABSTRACTSegmenting breast tumors from dynamic contrast‐enhanced magnetic resonance images is a critical step in the early detection and diagnosis of breast cancer. However, this task becomes significantly more challenging due to the diverse shapes and sizes of tumors, which make it difficult to establish a unified perception field for modeling them. Moreover, tumor regions are often subtle or imperceptible during early detection, exacerbating the issue of extreme class imbalance. This imbalance can lead to biased training and challenge accurately segmenting tumor regions from the predominant normal tissues. To address these issues, we propose a hierarchical region contrastive learning approach for breast tumor segmentation. Our approach introduces a novel hierarchical region contrastive learning loss function that addresses the class imbalance problem. This loss function encourages the model to create a clear separation between feature embeddings by maximizing the inter‐class margin and minimizing the intra‐class distance across different levels of the feature space. In addition, we design a novel Attention‐based 3D Multi‐scale Feature Fusion Residual Module to explore more granular multi‐scale representations to improve the feature learning ability of tumors. Extensive experiments on two breast DCE‐MRI datasets demonstrate that the proposed algorithm is more competitive against several state‐of‐the‐art approaches under different segmentation metrics.

Abstract 4125691: A machine learning approach to examining the associations of minority stressors and physical activity among sexual and gender minority adults

Introduction: Physical activity (PA) is an important modifiable factor for cardiovascular disease (CVD). Sexual and gender minority (SGM; e.g., lesbian/gay, bisexual, transgender) adults face higher risks of inadequate PA and CVD due to increased exposure to minority stressors, such as experienced and anticipated discrimination, which may influence SGM adults’ willingness to engage in PA. To our knowledge, researchers have not investigated the impact of minority stressors on objectively-measured PA among SGM adults. Goal: Using k-means clustering, we sought to identify whether clusters characterized by greater exposure to minority stressors were associated with lower moderate and vigorous PA (MVPA) and higher sedentary behavior among SGM adults. Methods: This daily diary study included an online sample of SGM adults living in the United States. Participants completed daily surveys about personally experienced discrimination and anticipated discrimination (i.e., expectation of experiencing discrimination) and wore wrist accelerometers for 28 days to objectively measure PA. We used k-means clustering to identify clusters based on reports of experienced and anticipated discrimination. We first determined the optimal number of clusters using established partition criteria. Next, we ran linear regression models (adjusted for demographic factors) to examine the associations of minority stress clusters with MVPA and sedentary time per week. Results: Among 42 SGM adults (mean age 27.0±7.7 years) with 1133 person-days of accelerometry data (~3% missing data), we identified four minority stress clusters: low anticipated/low experienced (LALE; reference group); low anticipated/high experienced (LAHE); high anticipated/low experienced (HALE); and high anticipated/high experienced discrimination (HAHE). Participants in the HALE cluster (n=12) engaged in 202 fewer minutes of MVPA than those in the LALE cluster (n=7). Participants in the LAHE cluster (n=10) had 123 fewer minutes of vigorous PA than those in the LALE cluster. No differences were identified for sedentary time. Conclusions: This is the first study to examine the association of minority stressors with objective PA among SGM adults. Participants in the HALE and LAHE clusters engaged in significantly lower PA than those with low levels of both experienced and anticipated discrimination. Findings underscore the importance of assessing minority stressors and the need for interventions to improve PA among SGM adults.

Enhancing Alaskan wildfire prediction and carbon flux estimation: a two-stage deep learning approach within a process-based model

Abstract Wildfires in boreal forests release substantial amounts of carbon into the atmosphere. However, current land-surface models are limited in their representation of fire processes, including their ignition and spread. This study thus developed FireDL, a novel data-driven machine-learning model for the prediction of natural wildfires, and combined it with a land-surface model to better understand the impact of fire on carbon fluxes. FireDL has a two-stage deep learning structure that sequentially combines a long short-term memory (LSTM) algorithm and an artificial neural network (ANN). Preliminary random forest analysis identified fire duration as an important factor in predicting the burned area. Thus, in FireDL, the LSTM algorithm was employed to predict fire occurrence and duration, utilizing lightning, vegetation, and climate datasets. Subsequently, the ANN predicted the total burned area using the LTSM-derived fire duration predictions and climate datasets as input. FireDL produced a robust performance in predicting large fires (>10 000 ha), achieving a correlation coefficient of 0.72. The daily-scaled burned area predictions derived from FireDL were integrated into the Community Land Model version 5—Biogeochemistry (CLM5-BGC) to produce CLM5-BGC-FireDL. This integration considerably improved carbon emission estimations. Notably, the total net ecosystem exchange (NEE) estimated using CLM5-BGC-FireDL in 2019, the year with the highest recorded burned area during our study, was twice that estimated using the standard CLM5-BGC. Discrepancies in the NEE can significantly influence atmospheric CO2 levels, highlighting the importance of our fire prediction model in forecasting the burned area and carbon emissions. The use of FireDL with future climate scenarios is thus anticipated to yield valuable insights into ecosystem management and climate change mitigation strategies.

Relationships between utilisation of human resources and academic performance in engineering at Technical and Vocational Institutions in Nakuru County, Kenya

The purpose of this article was to analyse the relationships between the utilisation of human resources and academic performance in engineering at technical and vocational institutions in Nakuru County, Kenya. The problem that the study addressed was the high failure rate of students in engineering. The study used a descriptive survey design utilising closed-ended questionnaires and interview schedules. The sample consisted of 331 students, 11 technicians, and 22 Administrators (11 Heads of Departments, 7 Deputy Principals, and 4 Principals). A proportionate sampling technique was used to obtain student samples per institution. Participants were selected through simple random sampling. The hypothesis was tested and established that there was a positive and statistically significant relationship between utilisation of human resources and academic performance in engineering (p=0.000<p=0.05), concluding that ineffective utilisation of human resources in the TVET institutions contributes to the observed low performance in engineering programs. The study recommended that institutions improve the coordination of practical sessions, enhance student-centred learning approaches, and invest in upgrading physical resources.

Alternative Learning Paradigms for Image Quality Transfer

Image Quality Transfer (IQT) aims to enhance the contrast and resolution of low-quality medical images, e.g. obtained from low-power devices, with rich information learned from higher quality images. In contrast to existing IQT methods in the literature which adopt supervised learning frameworks, in this work, we propose two novel formulations of the IQT problem. The first approach uses an unsupervised learning framework, whereas the second is a combination of both supervised and unsupervised learning. The unsupervised learning approach considers a sparse representation (SRep) and dictionary learning model, which we call IQT-SRep, whereas the combination of supervised and unsupervised learning ap- proach is based on deep dictionary learning (DDL), which we call IQT-DDL. The IQT-SRep approach trains two dictionaries using a sparse representation model using pairs of low- and high-quality volumes. Subsequently, the sparse representation of a low-quality block, in terms of the low-quality dictionary, can be directly used to recover the corresponding high-quality block using the high-quality dictionary. On the other hand, the IQT-DDL ap- proach explicitly learns a high-resolution dictionary to upscale the input volume, while the entire network, including high dictionary generator, is simultaneously optimised to take full advantage of deep learning methods. The two models are evaluated using a low-field mag- netic resonance imaging (MRI) application aiming to recover high-quality images akin to those obtained from high-field scanners. Experiments comparing the proposed approaches against state-of-the-art supervised deep learning IQT method (IQT-DL) identify that the two novel formulations of the IQT problem can avoid bias associated with supervised meth- ods when tested using out-of-distribution data that differs from the distribution of the data the model was trained on. This highlights the potential benefit of these novel paradigms for IQT.

High-Throughput Approaches to Engineer Fluorescent Nanosensors.

Optical sensors are powerful tools to identify and image (biological) molecules. Because of their optoelectronic properties, nanomaterials are often used as building blocks. To transduce the chemical interaction with the analyte into an optical signal, the interplay between surface chemistry and nanomaterial photophysics has to be optimized. Understanding these aspects promises major opportunities for tailored sensors with optimal performance. However, this requires methods to create and explore the many chemical permutations. Indeed, many current approaches are limited in throughput. This affects the chemical design space that can be studied, the application of machine learning approaches as well as fundamental mechanistic understanding. Here, an overview of selection-limited and synthesis-limited approaches is provided to create and identify molecular nanosensors. Bottlenecks are discussed and opportunities of non-classical recognition strategies are highlighted such as corona phase molecular recognition as well as the requirements for high throughput and scalability. Fluorescent carbon nanotubes are powerful building blocks for sensors and their huge chemical design space makes them an ideal platform for high throughput approaches. Therefore, they are the focus of this article, but the insights are transferable to any nanosensor system. Overall, this perspective aims to provide a fresh perspective to overcome current challenges in the nanosensor field.

Examining The Variables That Impact Academic Performance In Higher Education

This research paper examines the multifaceted variables that impact academic performance in higher education, with a particular focus on time management skills. Academic performance is influenced by a complex interplay of economic, social, psychological, environmental, and personal factors, each of which plays a role differently in a student's success. The objective of this study is to explore how these factors, especially time management, influence academic outcomes, and to provide a comprehensive review of the literature surrounding this topic. Academic performance in higher education is a multifaceted phenomenon influenced by an intricate mix of economic, social, psychological, environmental, and personal factors. This review critically examines key variables affecting student performance, focusing on underexplored areas such as time management skills, psychological well-being, and the changing landscape of digital learning environments. Drawing from extensive literature, this paper identifies gaps in current research and evaluates how higher education institutions can address these issues through targeted interventions. The results highlight the need for greater emphasis on mental health support, personalized learning approaches, and strategies to enhance students' time management and adaptability in evolving academic contexts. This paper highlights the significance of addressing both external and internal factors that affect academic performance. By fostering a supportive environment, providing necessary resources, and offering time management interventions, institutions can better support students in achieving academic success.

BLENDED LEARNING IN TEACHING ENGLISH LITERATURE DURING THE COVID-19 PANDEMIC: TEACHERS’ PERCEPTIONS

This study investigates teachers' perceptions of blended learning in the context of English Literature during the Covid-19 pandemic. The research aims to understand how English teachers adapted to the sudden shift from traditional to blended learning environments. The research includes exploring the strategies employed by teachers when integrating literature lessons into a blended learning approach, impacts of blended learning during the pandemic to teachers’ teaching professionalism and problems faced by teachers when designing Literature lessons during the pandemic. Four English Literature teachers from SMK Datuk Peter Mojuntin, Penampang participated in semi-structured interviews, providing different perceptions into their experiences with blended learning. The findings reveal that technology played an essential role in facilitating teaching and learning during the pandemic. The combination of traditional and digital methods allowed for a balanced approach to teaching English Literature, fostering interactive and collaborative learning experiences. Despite these benefits, the transition to blended learning presented significant challenges in terms of adaptation, professional growth and connectivity issues. The study helps to understand the implications of this research especially for researchers, teachers and students. The data gained from this study offer guidance for improving blended learning practices and addressing the needs of students and teachers in a rapidly evolving educational setting.

Abstract 4138376: A Machine Learning Approach to Predict Percutaneous Coronary Intervention in Patients with Critical Illness and Signs of Myocardial Injury

Background: Critically ill patients frequently experience multi-organ dysfunction and unstable vital signs, increasing their risk of myocardial injury and elevated serum troponin levels. Myocardial injury from oxygen supply-demand mismatch, whether due to systemic disease or obstructive coronary artery disease (CAD), often leads to patients undergoing an invasive coronary angiogram (ICA) and potentially percutaneous coronary intervention (PCI) if obstructive CAD is identified. However, this procedure carries significant risks for critically ill patients. To address this challenge, we aimed to develop a machine learning (ML) model to predict the need for PCI in critically ill patients with myocardial injury. Methods: The study cohort included critically ill patients with elevated serum troponin levels who underwent ICA during their hospitalization at a single institution from 2012 to 2022. We defined critical illness as a life-threatening condition requiring intensive support for vital sign instability. Patients with atherothrombosis identified during ICA were excluded from the study, focusing only on patients with type 2 MI. We utilized Matplotlib to analyze patient data, including demographics, lab results, vital signs, and comorbidities. We then used XGBoost, a supervised ML algorithm, for binary classification tasks. An ensemble of decision trees was employed, with gradient boosting to correct errors from previous trees. XGBoost iteratively adjusted predictions, creating a model to predict which patients required PCI. Results: Out of 765 patients, 173 (22.6%) underwent PCI. The cohort had a mean age of 60.5 ± 13.5 years, with 41.7% being female. The ML model accurately identified non-PCI candidates at 75% and PCI candidates at 64%, achieving an overall accuracy of 73%. The three most important features identified by the model were a history of diabetes, diastolic heart failure, and whether the patient received a blood transfusion during hospitalization. Conclusion: A ML approach using XGBoost demonstrated moderate accuracy in predicting the need for PCI in critically ill patients with signs of myocardial injury. This prediction holds clinical significance due to the heightened risks associated with ICA in these patients. Given that PCI occurred in only 22.6% of the cohort, future models should include a higher proportion of patients who underwent PCI to improve the robustness and true positive accuracy of the model.

Integration of cost-effective datasets to improve predictability of strategic noise mapping in transport corridors in Delhi city, India.

Assessing exposure to environmental noise levels at transport corridors remains complex in conditions where no standardized noise prediction model is available. In planning and policy implementation for noise control, noise mapping is an important step. In the present study, land use regression model has been developed to predict the environmental noise levels in Delhi city, India, using previously developed approaches along with machine learning techniques, however improved using new datasets. Lday, Lnight, LAeq,24h, and Ldn were modeled at daily resolution by utilizing an annual noise levels dataset from 31 locations in Delhi city. The noise-monitored data was integrated with travel time data, nighttime light data along with common parameters including land use, road networks, and meteorological parameters. The developed LUR models showed good fit with R2 of 0.72 for Lday, 0.55 for Lnight, 0.71 for LAeq,24h, and 0.61 for Ldn, which was further improved up to 0.88 for Lday, 0.79 for Lnight, 0.86 for LAeq,24h, and 0.81 for Ldn by integrating machine learning approaches. The developed models were validated through tenfold cross validation and by comparison to a separate noise level dataset. The average travel time variable was observed to be the most influential predictor of LUR models for Lday and LAeq,24h, which signifies the crucial impact of road traffic congestion on environmental noise levels. The study also analyzed the parametric sensitivity of various infrastructural factors reported in the study, which shall be helpful for planning for smart cities.

Abstract Su801: Use of Large Language Models to Optimize Clinical Text Analysis for In-Hospital Cardiac Arrest Identification

Introduction/Background: In-hospital cardiac arrest (IHCA) is experienced by approximately 300,000 patients annually in the US. While individual care teams can readily identify IHCA at the bedside, the reporting of these events after the event often lacks consistency and poses a major challenge for hospitals across the US. Accurate and timely reporting of IHCA events is crucial for facilitating QI initiatives, such as CPR quality review, optimizing team performance, and benchmarking IHCA outcomes. Hypothesis: Large language models (LLMs) can be used to identify IHCA events at the patient encounter-level by analyzing targeted note types from the inpatient record with high performance. Methods: Adult (≥18 years old) inpatient encounters at the Hospital of the University of Pennsylvania from 06/2018 to 03/2022 with a reported clinical emergency were identified from a QI database. Discharge summaries and notes associated with the encounters of interest were extracted from the EHR and deidentified using a natural language processing tool. Notes were reviewed by research staff to ascertain true IHCA events. Positive IHCA was defined as the loss of pulses followed by the delivery of CPR. An LLM, GPT4 v. 32K-Chat, was employed on the Penn Medicine Microsoft Azure Databricks environment to compare zero- and few-shot prompting methods to identify IHCA. Model performance was measured using accuracy, precision, recall and F1 scores. Results: Manual chart review determined 96.8% and 3.2% of cases to be IHCA+ and IHCA-, respectively. A balanced pilot dataset of positive and negative cases was created and the accuracy values across strategically sampled note types was checked. Using zero-shot learning with Chain-of-Thought (COT) prompting, the LLM performed with an accuracy of 86%, precision of 77%, recall of 100%, and F1-score of 87%. Using three-shot learning with COT, the LLM performed with an accuracy of 90%, precision of 83%, recall of 100%, and F1-score of 91%. Conclusions: This study demonstrates the potential efficacy of leveraging LLMs to automatically classify IHCA from strategically sampled note types using both zero and few-shot learning approaches. In the future, we will implement temporal sampling of note types and refine model parameters such as output token size and temperature. These findings suggest that such technologies can be effective in real-world clinical settings, providing a scalable solution to improve patient outcomes and quality improvement efforts.

JUNIOR HIGH SCHOOL STUDENTS' PERCEPTIONS OF PROBLEM BASED LEARNING MODEL IN SCIENCE LEARNING

Problem Based Learning (PBL) is a learning approach that is often used in education in recent years. The purpose of this article is to find out how junior high school students perceive PBL model in science class and the benefits of PBL model in science learning. To find data related to junior high school students' perceptions of problem-based learning models in science classes, the researcher conducted a systematic review of several relevant sources with a period from 2014 to 2023, following the principles of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). Based on the research results, it is known that the problem solving ability of students who learn using the PBL model is better than students who learn with conventional models. The results of this study prove PBL as an effective learning method for science compared to conventional teaching methods. PBL method is perceived to improve students' knowledge and communication skills. Based on several research results that have been conducted, it can be concluded that the PBL model has a positive impact on learning.

Opportunities and Challenges of the Merdeka Curriculum in the Islamic Religious Education Subject at SMA Negeri 1 Sakra, East Lombok

Education in Indonesia is undergoing significant transformation in tandem with technological advancements, with the Merdeka Curriculum emerging as a response to the demands of the modern era, particularly in the context of Industry 4.0. This curriculum introduces flexibility and innovation in learning approaches but also encounters challenges such as resource limitations and the necessity for teacher training. At SMA Negeri 1 Sakra in East Lombok, the implementation of the Merdeka Curriculum since the 2022/2023 academic year has presented opportunities for more varied and inclusive teaching methods, particularly in Islamic Religious Education (PAI), while also confronting challenges in balancing theological and practical aspects.This research aims to explore the opportunities and challenges of the Merdeka Curriculum within the Islamic Religious Education subject at SMA Negeri 1 Sakra, East Lombok Regency. Utilizing a qualitative approach with a descriptive naturalistic method, the study focuses on direct observation and a holistic understanding of the phenomena at SMA Negeri 1 Sakra, East Lombok. Data collection is conducted through observation, interviews, and documentation, employing interactive analysis based on the Miles and Huberman model to examine the implementation of the Merdeka Curriculum.The findings of this study reveal that the opportunities perceived by SMA Negeri 1 Sakra, East Lombok, after implementing the Merdeka Curriculum include enhanced enthusiasm among educators to develop their capacities, increased student engagement and activity in the classroom, and the school's ability to cultivate a learning ecosystem that is more student-centered. However, the school also faces challenges, such as the need to shift the mindset of both teachers and students, the task of harmonizing diverse Islamic values among students and teachers, the difficulty of creating enjoyable and high-quality learning experiences, and the issue of resource constraints.

Collaborative co-design to reimagine teaching practices in hybrid spaces

Following the Future Made in Australia Act, which prioritises quality and equitable education (Albanese, 2024), there is a pressing need for regional universities to provide equal educational opportunities for students regardless of their geographical locations. Furthermore, as outlined in the Australian Universities Accord Final Report, advances in pedagogy are required to fulfil the potential of high-quality online and hybrid learning modes, as well as deliver teaching which supports more diverse student cohorts (Australian Government, 2024). Hybrid learning spaces have the potential to bridge educational disparities by offering accessibility and flexibility to both regional and metropolitan students, allowing them to pursue their education without compromising their other responsibilities (Lightner & Lightner-Laws, 2016). However, significant adjustments to traditional face to face student engagement activities are crucial to the success of hybrid teaching and learning (Gamage et al., 2022). University teachers often have preconceived ideas about these spaces and lack the time, skills, and knowledge to adapt their teaching practices effectively (Bøjer & Brøns, 2022; Raes et al., 2020). Additionally, students struggle to adapt their learning approaches in these spaces, leading to limited engagement and interaction with teachers and peers (Raes et al., 2020). Effective implementation requires support for both teachers and students to reimagine their routines, practices, and expectations in teaching and learning (Bøjer & Brøns, 2022; Fawns et al., 2022; Gamage et al., 2022). To address this challenge, a research project was conducted that involved four phases: (1) Co-design workshop, (2) Resource development, (3) Data collection, (4) Analysis and outcomes. The first and second phases will be discussed in this poster presentation. In the first phase, the research team participated in a collaborative co-design workshop to develop a professional development (PD) module and practical guidelines for teachers. Collaborative co-design is a participatory process where designers and users work together using co-design tools, techniques, and scaffolds to provide expert perspectives on a problem (Bøjer & Brøns, 2022; Sanders and Stappers, 2007). In this process, researchers, teachers, and students collaborated to generate ideas and knowledge for the content of a PD module and the practical guidelines. The workshop was theoretically grounded in Moore’s (1989) three types of interactions and Radcliffe’s (2009) Pedagogy-Space-Technology (PST) framework. The active involvement of teachers and students allowed the research team to explore challenges between learning spaces, technology and teaching practices, focusing on student agency and engagement. Crucial to co-design, the research participants interchanged roles during the workshop, shifting between users, designers, and facilitators depending on individual creativity and expertise in specific areas (Bøjer & Brøns, 2022; Sanders and Stappers, 2007). The design solution of a PD module and practical guidelines was developed and provided teachers with strategies for effective hybrid teaching, addressing both technological challenges and appropriate pedagogies. The future direction of this project is to examine the effectiveness of hybrid learning and teaching by exploring the engagement between teachers and students online and on campus. The findings advance pedagogy by promoting equitable engagement for online and on-campus students, ensuring consistent learning experiences and bridging geographical gaps.

Challenging dominant perspectives of digital literacy in Western higher education

The 21st-century work environment, for which universities prepare graduates, demands multi-dimensional and interdisciplinary competencies infused with digital literacy skills essential for sustainable employability (Baruch & Sullivan, 2022; Gürbüz, Bakker, Demerouti et al., 2023). Our increasingly digitalised society reinforces the need for learners to acquire digital competencies as part of their education (Gutiérrez-Ángel, Sánchez-García, Mercader-Rubio et al., 2022). Digital literacy, defined as the ability to access, manage, understand, integrate, communicate, evaluate, and create information safely and appropriately through digital technologies, is still an evolving concept (UNESCO, 2018). With the mandate to produce graduates capable of performing to expected standards in ever-changing work environments, higher education institutions continuously define the digital competencies of their graduates and how to develop them. This involves ongoing discussions about what students should learn to be digitally competent (curriculum), how they should be taught and supported (pedagogy), and the associated progression in learning (assessment) (Ilomäki, Lakkala, Kallunki et al., 2023). However, Western perspectives on learning and teaching in higher education often dominate these discussions. Given that most students in Australian universities come from diverse socio-cultural backgrounds, there is a need to address this gap in the literature by considering alternative perspectives on curriculum, pedagogy, and assessments for digital literacy in higher education. This study presents findings from a critical integrative review of the literature (Whittemore & Knafl, 2005). The aim is to identify alternative perspectives in defining the digital competencies critical in preparing students for diverse socio-cultural work environments. To achieve this, a well-known meta-framework of digital literacy (Tran, Jung, Unangst et al., 2023; Martínez, Sádaba and Serrano-Puche, 2021) will be used to map digital competencies identified in global literature, exploring and challenging the concept of a digitally competent university graduate in today’s diverse world. The integrative review will also identify teaching and learning approaches (pedagogies) for developing students’ digital competencies from socio-cultural perspectives, as discussed in the literature (Mahn and John-Steiner, 2012). Peer-reviewed articles, reports, and e-book chapters will be searched in global literature, with inclusion criteria focusing on keywords such as digital literacies, digital competencies, socio-cultural learning theory, teaching pedagogies, curriculum design, and learner support. Databases will include ProQuest, Scopus, and Web of Science. Relevant data will be extracted using an Excel sheet, and thematic and interpretive analyses will be employed to synthesise the data critically. The findings will provide insights into how socio-cultural differences influence the valued aspects of digital literacy and how they are taught or assessed in higher education. This research will also offer critical insights into how teachers can inclusively support diverse students, considering unique ways to recognise prior educational experiences, learning preferences, and needs. The final output will be a conceptual framework for teaching and learning digital literacy in Western high education from socio-cultural perspectives. Ultimately, this research aims to provide a much-needed international perspective on what it means to be a digitally competent graduate in Australia and globally.

Experts' Cognition-Driven Ensemble Deep Learning for External Validation of Predicting Pathological Complete Response to Neoadjuvant Chemotherapy from Histological Images in Breast Cancer

In breast cancer, neoadjuvant chemotherapy (NAC) provides a standard treatment option for patients who have locally advanced cancer and some large operable tumors. A patient will have better prognosis when he has achieved a pathological complete response (pCR) with the treatment of NAC. There has been a trend to directly predict pCR to NAC from histological images based on deep learning (DL). However, the DL-based predictive models numerically have better performances in internal validation than in external validation. In this paper, we aim to alleviate this situation with an intrinsic approach. We propose an experts’ cognition-driven ensemble deep learning (ECDEDL) approach. Taking the cognition of both pathology and artificial intelligence experts into consideration to improve the generalization of the predictive model to the external validation, ECDEDL can intrinsically approximate the working paradigm of a human being which will refer to his various working experiences to make decisions. ECDEDL was validated with 695 WSIs collected from the same center as the primary dataset to develop the predictive model and perform the internal validation, and was also validated with 340 WSIs collected from other three centers as the external dataset to perform the external validation. In external validation, ECDEDL improves the AUCs of pCR prediction from 61.52(59.80-63.26) to 67.75(66.74-68.80) and the Accuracies of pCR prediction from 56.09(49.39-62.79) to 71.01(69.44-72.58). ECDEDL was quite effective for external validation of predicting pCR to NAC from histological images in breast cancer, numerically approximating the internal validation.

Novel reinforcement learning technique based parameter estimation for proton exchange membrane fuel cell model

Proton Exchange Membrane Fuel Cells (PEMFCs) offer a clean and sustainable alternative to traditional engines. PEMFCs play a vital role in progressing hydrogen-based energy solutions. Accurate modeling of PEMFC performance is essential for enhancing their efficiency. This paper introduces a novel reinforcement learning (RL) approach for estimating PEMFC parameters, addressing the challenges of the complex and nonlinear dynamics of the PEMFCs. The proposed RL method minimizes the sum of squared errors between measured and simulated voltages and provides an adaptive and self-improving RL-based Estimation that learns continuously from system feedback. The RL-based approach demonstrates superior accuracy and performance compared with traditional metaheuristic techniques. It has been validated through theoretical and experimental comparisons and tested on commercial PEMFCs, including the Temasek 1 kW, the 6 kW Nedstack PS6, and the Horizon H-12 12 W. The dataset used in this study comes from experimental data. This research contributes to the precise modeling of PEMFCs, improving their efficiency, and developing wider adoption of PEMFCs in sustainable energy solutions.

Enhancing type 2 diabetes mellitus prediction by integrating metabolomics and tree-based boosting approaches

BackgroundType 2 diabetes mellitus (T2DM) is a global health problem characterized by insulin resistance and hyperglycemia. Early detection and accurate prediction of T2DM is crucial for effective management and prevention. This study explores the integration of machine learning (ML) and explainable artificial intelligence (XAI) approaches based on metabolomics panel data to identify biomarkers and develop predictive models for T2DM.MethodsMetabolomics data from T2DM (n = 31) and healthy controls (n = 34) were analyzed for biomarker discovery (mostly amino acids, fatty acids, and purines) and T2DM prediction. Feature selection was performed using the least absolute shrinkage and selection operator (LASSO) regression to enhance the model’s accuracy and interpretability. Advanced three tree-based ML algorithms (KTBoost: Kernel-Tree Boosting; XGBoost: eXtreme Gradient Boosting; NGBoost: Natural Gradient Boosting) were employed to predict T2DM using these biomarkers. The SHapley Additive exPlanations (SHAP) method was used to explain the effects of metabolomics biomarkers on the prediction of the model.ResultsThe study identified multiple metabolites associated with T2DM, where LASSO feature selection highlighted important biomarkers. KTBoost [Accuracy: 0.938; CI: (0.880-0.997), Sensitivity: 0.971; CI: (0.847-0.999), Area under the Curve (AUC): 0.965; CI: (0.937-0.994)] demonstrated its effectiveness in using complex metabolomics data for T2DM prediction and achieved better performance than other models. According to KTBoost’s SHAP, high levels of phenylactate (pla) and taurine metabolites, as well as low concentrations of cysteine, laspartate, and lcysteate, are strongly associated with the presence of T2DM.ConclusionThe integration of metabolomics profiling and XAI offers a promising approach to predicting T2DM. The use of tree-based algorithms, in particular KTBoost, provides a robust framework for analyzing complex datasets and improves the prediction accuracy of T2DM onset. Future research should focus on validating these biomarkers and models in larger, more diverse populations to solidify their clinical utility.