Learning Module Research Articles

ScSCC: A swapped contrastive learning‐based clustering method for single‐cell gene expression data

AbstractCell clustering plays a pivotal role in deciphering the intricacies of cell types, facilitating subsequent cell annotation endeavors within scRNA‐seq data analysis. In this paper, we propose a novel swapped contrastive clustering algorithm for scRNA‐seq data called scSCC. scSCC combines two contrastive learning modules, namely the instance contrastive learning module and the swapped prediction module, to extract clustering‐friendly cell representations. Through the combination of swapped prediction module and instance contrastive learning module, scSCC can retrieve disentangled cell representations and amplify the clustering signals in the latent space, leading to satisfactory clustering performance. Different from existing contrastive‐learning‐based scRNA‐seq data clustering algorithms, the swapped prediction module of scSCC injects clustering signals to the latent space through some clustering prototypes. The swapped prediction module encourages cells of the same cluster to gravitate toward the common clustering prototype and naturally stay away from other prototypes in the latent space, hence cell representations obtained by scSCC are more clustering‐friendly compared to other algorithms. Experimental results on real scRNA‐seq datasets show that scSCC achieves improved clustering performance compared with the benchmark methods. The ablation study on two contrastive modules exhibits the promotion by the combination of instance learning module and swapped prediction module. The source codes are available at the GitHub website (EnchantedJoy/scSCC).

Unified Text-Image Space Alignment with Cross-Modal Prompting in CLIP for UDA

Unsupervised Domain Adaptation (UDA) aims to transfer models trained on a labeled source domain to an unlabeled target domain. Due to the excellent generalization ability of Vision Language Models (VLMs) such as CLIP in downstream tasks, most recent methods apply CLIP to UDA tasks through learning domain-specific text prompts for source and target domains separately. However, these methods fail to dynamically adjust image features based on the characteristics of their respective domains, thereby limiting their alignment with domain-specific text prompts in CLIP’s joint space, which is a key factor in improving classification performance in the target domain. To bridge this gap, we propose a Unified Text-Image Space Alignment with Cross-Modal Prompting (UTISA) framework for UDA. Firstly, we introduce a cross-modal prompt learning (CMP) module to generate domain-specific image prompts and layer-specific image prompts for the visual branch to encode domain-specific knowledge globally and locally. Secondly, under the guidance of image prompts, we introduce a domain-aware multi-layer feature fusion (DMF) module to construct multi-layer domain features for each domain and enhance the image features with these multi-layer domain features, which enables the image features to better reflect the characteristics of their respective domains, thereby promoting their alignment with domain-specific text prompts. Moreover, we introduce a perturbation-driven regularization (PDR) mechanism for the target domain to enhance the robustness and generalization of the model. The experiments demonstrate that UTISA achieves the best performance on three mainstream UDA benchmarks, including 87.9% on OfficeHome, 90.9% on VisDA-2017, 62.4% on DomainNet.

Class-aware multi-level attention learning for semi-supervised breast cancer diagnosis under imbalanced label distribution.

Breast cancer affects a significant number of patients worldwide, and early diagnosis is critical for improving cure rates and prognosis. Deep learning-based breast cancer classification algorithms have substantially alleviated the burden on medical personnel. However, existing breast cancer diagnosis models face notable limitations which are challenging to obtain in clinical settings, such as reliance on a large volume of labeled samples, an inability to comprehensively extract features from breast cancer images, and susceptibility to overfitting on account of imbalanced class distribution. Therefore, we propose the class-aware multi-level attention learning model focused on semi-supervised breast cancer diagnosis to effectively reduce the dependency on extensive data annotation. Additionally, we develop the multi-level fusion attention learning module, which integrates multiple mutual attention components across different layers, allowing the model to precisely identify critical regions for lesion categorization. Finally, we design the class-aware adaptive pseudo-labeling module which adaptively predicts category distribution in unlabeled data, and directs the model to focus on underrepresented categories, ensuring a balanced learning process. Experimental results on the BACH dataset demonstrate that our proposed model achieves an accuracy of 86.7% with only 40% labeled microscopic data, showcasing its outstanding contribution to semi-supervised breast cancer diagnosis.

RL-Cervix.Net: A Hybrid Lightweight Model Integrating Reinforcement Learning for Cervical Cell Classification

Background: Reinforcement learning (RL) represents a significant advancement in artificial intelligence (AI), particularly for complex sequential decision-making challenges. Its capability to iteratively refine decisions makes it ideal for applications in medicine, such as the detection of cervical cancer; a major cause of mortality among women globally. The Pap smear test, a crucial diagnostic tool for cervical cancer, benefits from enhancements in AI, facilitating the development of automated diagnostic systems that improve screening effectiveness. This research introduces RL-Cervix.Net, a hybrid model integrating RL with convolutional neural network (CNN) technologies, aimed at elevating the precision and efficiency of cervical cancer screenings. Methods: RL-Cervix.Net combines the robust ResNet-50 architecture with a reinforcement learning module tailored for the unique challenges of cytological image analysis. The model was trained and validated using three extensive public datasets to ensure its effectiveness under realistic conditions. A novel application of RL for dynamic feature refinement and adjustment based on reward functions was employed to optimize the detection capabilities of the model. Results: The innovative integration of RL into the CNN framework allowed RL-Cervix.Net to achieve an unprecedented classification accuracy of 99.98% in identifying atypical cells indicative of cervical lesions. The model demonstrated superior accuracy and interpretability compared to existing methods, addressing variability and complexities inherent in cytological images. Conclusions: The RL-Cervix.Net model marks a significant breakthrough in the application of AI for medical diagnostics, particularly in the early detection of cervical cancer. By significantly improving diagnostic accuracy and efficiency, RL-Cervix.Net has the potential to enhance patient outcomes through earlier and more precise identification of the disease, ultimately contributing to reduced mortality rates and improved healthcare delivery.

Problematika Guru Dalam Menerapkan Kurikulum Merdeka Belajar Di Sekolah Dasar Swasta Jakarta

This study aims to determine the implementation of the independent learning curriculum, the problems of teachers in implementing the independent learning curriculum and how teachers overcome the problems of implementing the independent learning curriculum at SD Tirtamarta- BPK Penabur. This study uses a qualitative approach with a descriptive research type. The subjects of this study were the School, teachers and students of SD Tirtamarta- BPK Penabur. Data collection was carried out through observation, interviews and documentation. The data analysis techniques used were power reduction, data presentation and drawing conclusions. Based on the results of the study, it was concluded that the implementation of the independent curriculum had been implemented at SD Tirtamarta- BPK Penabur with project-based learning activities, diagnostic, formative and summative assessments and subject-based learning, science, report cards, ATP and teaching modules worked on in groups. The problems experienced by teachers in implementing the independent learning curriculum in planning, implementing and assessing learning were the difficulty of analyzing CP, formulating TP and compiling ATP and teaching modules, determining learning methods and strategies and lack of time allocation for project-based learning.

Features of real-time modeling of steelmaking processes using unreal engine 5

This paper explores the features of real-time modeling of steelmaking processes using Unreal Engine 5, focusing on the visualization and simulation of continuous casting machine (CCM) operations. The study highlights the advantages of interactive and virtual reality (VR) technologies in the training and optimization of metallurgical processes, providing a safer and more cost-effective alternative to traditional training methods. A detailed approach to 3D modeling of CCM components is presented, including the tundish, mold, secondary cooling system, and roller sections, with a focus on their realistic visualization and optimization for real-time performance. The implementation of physical simulations using Chaos Physics and Niagara in Unreal Engine 5 allows for an accurate rep-resentation of molten steel flow, solidification dynamics, and temperature gradients. Addi-tionally, gamification elements have been integrated to enhance user interaction, enabling students and engineers to explore various operational parameters in a controlled virtual envi-ronment. The research also assesses the efficiency of digital simulation techniques compared to conventional learning methods in metallurgical training programs. The use of VR-based in-teractive learning modules improves engagement and comprehension by allowing users to manipulate casting parameters and observe the impact on the final product in real-time. The paper concludes that Unreal Engine 5 provides an effective platform for modeling complex industrial processes, offering high-quality visual representation, dynamic interaction, and immersive training experiences for metallurgical professionals. Future work will focus on enhancing AI-driven adaptive learning, expanding the simu-lation scope to include additional steelmaking stages, and integrating augmented reality (AR) tools to bridge the gap between virtual training and real-world applications.

Abstract DP9: Enhancing Nursing Education to Improve Stroke Patient Education and Outcomes

Background: Stroke is the fifth leading cause of death in Minnesota and in the United States. Stroke is a leading cause of long-term disabilities and is largely preventable (80%). Evidence indicates that patients receiving care from stroke specialized nurses, including nurse-led stroke prevention, show better outcomes. Stroke program leaders at a Primary Stroke Center identified this as an opportunity to improve stroke program education, quality of nursing care, and stroke patient outcomes. Purpose: The primary focus area was improving stroke-patient education to include individualized stroke risk factors. As our project progressed, the initial goal remained. The secondary focus was to see if the effect of in-person, hands-on (escape room) nurse education would improve individualized stroke-patient education compliance compared to eLearning and poster board education. Methods: All stroke cases from June 2023 - June 2024 were audited for documentation of individualized stroke-patient education. Utilizing stroke guidelines and scientific statements, a standardized practice of individualizing stroke-patient education was developed. Staff learning modules were developed including eLearning, unit poster board, and in-person learning (escape room). Throughout the project, completion rates of individualized stroke-patient education were compared. A nursing survey was conducted to evaluate confidence and knowledge post multimodality learning modules. Results: Prior to initiation of nursing learning modules, 24 hospitalized stroke-patient charts from June and July 2023 were audited for individualized stroke education (40%). After initiation of nursing learning modules, 183 hospitalized stroke-patient charts from August 2023 to June 2024 were audited for individualized stroke education (poster board 44%, eLearning 45%, escape room 75%). The post nursing learning module survey responses demonstrated improved learners’ knowledge for the interactive escape room (poster board average-good 77%, eLearning average-good 83%, escape room good-excellent 89%). Conclusion: By identifying educational gaps, we were able to enhance nursing education; which improved stroke-patient education and outcomes. In conclusion, nursing confidence and knowledge increased with the in-person escape room education. This suggests engaging in-person learning could be more effective in developing nursing knowledge, improving patient care, and positive patient outcomes.

Weakly supervised multi-modal contrastive learning framework for predicting the HER2 scores in breast cancer.

Human epidermal growth factor receptor 2 (HER2) is an important biomarker for prognosis and prediction of treatment response in breast cancer (BC). HER2 scoring is typically evaluated by pathologist microscopic observation on immunohistochemistry (IHC) images, which is labor-intensive and results in observational biases among different pathologists. Most existing methods generally use hand-crafted features or deep learning models in unimodal (hematoxylin and eosin (H&E) or IHC) to predict HER2 scores through supervised or weakly supervised learning. Consequently, the information from different modalities is not effectively integrated into feature learning which can help improve HER2 scoring performance. In this paper, we propose a novel weakly supervised multi-modal contrastive learning (WSMCL) framework to predict the HER2 scores in BC at the whole slide image (WSI) level. It aims to leverage multi-modal (H&E and IHC) joint learning under the weak supervision of WSI label to achieve the HER2 score prediction. Specifically, the patch features within H&E and IHC WSIs are respectively extracted and then the multi-head self-attention (MHSA) is used to explore the global dependencies of the patches within each modality. The patch features corresponding to top-k and bottom-k attention scores generated by MHSA in each modality are selected as the candidates for multi-modal joint learning. Particularly, a multi-modal attentive contrastive learning (MACL) module is designed to guarantee the semantic alignment of the candidate features from different modalities. Extensive experiments demonstrate the proposed WSMCL has the better HER2 scoring performance and outperforms the state-of-the-art methods. The code is available at https://github.com/HFUT-miaLab/WSMCL.

Enabling Multi-Step Forecasting with Structured State Space Learning Module

Data-driven soft sensor incorporated with the model predictive control (MPC) algorithms facilitating product quality and cost control are of imperative importance in industrial processes. However, the widely used one-step forecasting method can not incorporate with MPC and therefore restricts the practical usage of soft sensor. Multi-step forecasting introduces long-term dependencies problems yet has not been effectively resolved within traditional model structure. To address this problem, this paper proposes the deep learning network architecture named Extended State Space Learning Module (ESSLM). ESSLM extends the nonlinear mapping architecture of deep learning based on state space and retains state transfer matrices to characterize the dynamics of the system. ESSLM distinguishes itself from explicit network architectures such as gated RNNs by addressing the long-term dependencies problems through an implicit initialization method, and the MLP and RNN algorithms can be regarded as the manifestation of ESSLM in special cases. ESSLM characterizes the latent space as the coefficients of the orthogonal basis functions so that the input data can be encoded into a high-dimensional feature space with minimal information loss which efficiently achieves multi-step forecasting and give greater utility and practical significance.

Abstract TP104: Bridging the Education Gap: Strengthening Stroke Rehabilitation in the Midwest

Introduction: Post-acute care (PAC) facilities across Iowa, Montana, Nebraska, and North Dakota are engaged in the American Heart Association’s Mission: Lifeline Stroke PAC Initiative, a quality improvement initiative focused on enhancing stroke rehabilitation through evidence-based practice. Participants across these states collaborate to improve patient care, identify gaps in current practices, and enhance stroke specific education. Purpose: In a 2023 survey of healthcare professionals (HCPs) in Iowa, 48% (n=120) of respondents identified professional education as a missing resource that could help them provide rehabilitation care to stroke survivors. Through the initiative, education was developed to address this identified need. Methods: A variety of free educational opportunities were offered throughout the Mission: Lifeline Stroke Initiatives. These included in-person stroke conferences and workshops, virtual learning modules, online courses, and discipline specific rehabilitation certifications. 760 HCPs participated in 12 different educational offerings across Midwest states. Attendees were surveyed to assess the perceived benefits of the offerings, utility to their current role, and improvements in knowledge and confidence providing rehabilitation care to stroke patients. Results: Following the completion of the educational offerings, 93 attendees completed a post-survey. The large majority of respondents reported they were extremely or mostly confident in identifying signs of a stroke (95%), describing common deficits after a stroke (96%), and implementing evidenced-based practices to address these deficits (90%). Most respondents agreed the trainings improved their knowledge of the relationship between treatment and stroke rehab patient outcomes (91%) and their understanding about key issues related to working with stroke rehab patients (91%). Overall, 78% of respondents reported they always or often use the knowledge and skills learned from the trainings and 95% were satisfied with the content of the trainings. Respondents also identified additional educational needs for cognitive rehabilitation strategies, aphasia interventions, caregiver education, community integration, and general stroke rehabilitation. Conclusions: Mission: Lifeline Stroke educational offerings benefitted HCPs by improving their knowledge and confidence to provide care to stroke patients. Education on additional stroke specific topics were also identified.

Feasibility and acceptability of a virtual learning module for navigating angry conversations in clinical encounters

BackgroundThe use of technology in medical education has been increasing with more students exposed to some form of online learning or tutorials, under the umbrella of virtual learning (VL). Many programmes, particularly those involving virtual reality, have centred on practical skills, such as surgical techniques or anatomical knowledge, rather than communication. The study presented here examined the feasibility and acceptability of a VL module developed to aid communication when handling angry patients and their relatives.MethodsParticipants were 4th and 5th year medical students at the Brighton and Sussex Medical School. Students were randomly allocated to receive training about having angry conversations in a clinical setting via virtual reality headset or desktop application. Prior to the intervention, everyone completed the SE12 self-efficacy questionnaire, a 5-item confidence measure, and free-response study specific survey. Following the module, they completed another study specific survey, with fixed and free responses, the confidence measure, along with the UTAUT2 questionnaire on acceptance and use of technology. Quantitative data was analysed descriptively, conceptual content analysis was applied to free responses. Participants received a £25 voucher for their time.ResultsTwenty students took part in the project. Scores on the SE12 did not differ significantly between intervention arms. Confidence improved across all five categories - recognising responses that diffuse or exacerbate anger, identifying anger signals, remaining calm in hostile situations, moving forward with empathy, and applying techniques to different situations. Responses to the UTAUT2 indicated acceptance of VL, including the psychological safety it provides. Nineteen categories for free text responses were developed via content analysis. Participants spoke frequently about the challenges of navigating anger. There was initial apprehension VL would not feel realistic, though this was largely reversed post-intervention. Students expressed preference for a combination of VL, whichever modality, and face-to-face teaching, recognising benefits of both.ConclusionStudents found the training to be acceptable, providing them with tangible skills. There should be a consideration as to how to incorporate VL, with a mix of face-to-face practice for added realism.Trial RegistrationClinical trial number not applicable.

A critical multicultural education learning module for English language teacher education programs

As classrooms become increasingly diverse in terms of culture and language, many teacher education programs have started to incorporate critical multicultural education (CME). Within the framework of a qualitative research design, the aim of this study was to construct a concise learning module that introduces CME to pre-service English language teachers and encourages them to adopt a critical perspective in their teaching. Developing this module involved a meticulous four-stage process. In the initial two stages, the module was formulated through iterative brainstorming sessions, and in the latter stages, it was implemented with groups of English language teaching students. The overall findings indicate that the module provided various benefits including fostering positive attitudes towards diversity, promoting an understanding of social issues, and raising awareness about educational goals, the ideologies behind education, and the political position of teaching. It is evident that the implementation of CME in teacher education positively impacts pre-service teachers by instilling more favorable attitudes toward diversity. This study highlights the significant influence of CME courses on pre-service teachers.

Denoising Multi-Level Cross-Attention and Contrastive Learning for Chest Radiology Report Generation.

Chest radiology report generation plays a vital role in supporting diagnosis, alleviating physician workload, and reducing the risk of misdiagnosis. However, significant challenges persist: (1) Data bias and background noise in chest images often obscure subtle lesion details, leading models to generate similar reports; (2) the distinct modal spaces of radiology images and reports weaken the semantic correlation between detailed visual lesion features and report sentences; (3) generated reports often lack crucial patient background and disease extent details, impeding report quality and accuracy. To address these challenges, this paper proposes a novel approach for generating chest radiology reports, utilizing denoising multi-level cross-attention and multi-level contrastive. The proposed method first involves sequential encoding of frontal and lateral radiology images into a visual extractor to enhance semantic coherence across image patches and improve visual feature representation. The enhanced visual features are processed through denoising multi-level cross-attention, which effectively suppresses noise and highlights the details of subtle lesions. Secondly, a multi-level contrastive learning module leverages contrastive among images, text, and disease labels to distinguish positive samples from negative ones, thereby strengthening the semantic correlation between detailed visual lesion features and report sentences. Finally, relevant knowledge is incorporated into the report generator to enhance the description of patient lesion details. Comparative experiments were performed against other state-of-the-art methods on the IU-Xray and MIMIC-CXR datasets, demonstrating that the proposed method significantly improves model performance. Additionally, ablation studies confirm that each module contributes to enhancing the quality of generated reports.

Implementation of a Malignant Hematology Education Intervention and its Impact on Hematology Nurse Practitioner Knowledge and Self-Efficacy to Practice

Background: Nurse practitioners (NPs) entering the malignant hematology specialty often lack hematology-specific knowledge needed for practice; many have reported they want and need more hematology education. Deficiencies in hematology education stem from the minimal amount of hematology content included in NP programs and during job orientation. Knowledge deficits among NPs are associated with unpreparedness to practice and feelings of anxiety, insecurity, inadequacy, and guilt. Self-efficacy (SE) is a correlate to NP knowledge acquisition and competency development. Purpose: This was a process improvement effort to examine the impact of a malignant hematology education module on NP knowledge and SE to practice in malignant hematology. Methods: A convenience sample of 11 NP participants were recruited during onboarding to a hematology department in a tertiary care cancer hospital in Southern California. Participants completed an online learning module containing education about hematological malignancies. A pretest and posttest design using questionnaires was employed for data collection. Knowledge and SE scores obtained before and after the intervention were compared to assess for improvement. Results: Posttest NP knowledge scores increased by a mean of 2.4 points (20%; mean pretest: 7.1/12, posttest: 9.5/12), p < .05. Posttest scores for NP SE were similar to baseline (mean pretest: 32.6/40 points, posttest: 32.3/40), p > .05. Participants reported that the intervention was helpful in the onboarding process. Implications/Conclusion: The findings help establish the feasibility of a malignant hematology learning module in increasing knowledge for NPs during onboarding. Integrating an education module into NP job training may increase knowledge and preparedness to practice in malignant hematology.

Multiscale Spatial–Spectral Dense Residual Attention Fusion Network for Spectral Reconstruction from Multispectral Images

Spectral reconstruction (SR) from multispectral images (MSIs) is a crucial task in remote sensing image processing, aiming to enhance the spectral resolution of MSIs to produce hyperspectral images (HSIs). However, most existing deep learning-based SR methods primarily focus on deeper network architectures, often overlooking the importance of extracting multiscale spatial and spectral features in the MSIs. To bridge this gap, this paper proposes a multiscale spatial–spectral dense residual attention fusion network (MS2Net) for SR. Specifically, considering the multiscale nature of the land-cover types in the MSIs, a three-dimensional multiscale hierarchical residual module is designed and embedded in the head of the proposed MS2Net to extract spatial and spectral multiscale features. Subsequently, we employ a two-pathway architecture to extract deep spatial and spectral features. Both pathways are constructed with a single-shot dense residual module for efficient feature learning and a residual composite soft attention module to enhance salient spatial and spectral features. Finally, the spatial and spectral features extracted from the different pathways are integrated using an adaptive weighted feature fusion module to reconstruct HSIs. Extensive experiments on both simulated and real-world datasets demonstrate that the proposed MS2Net achieves superior performance compared to state-of-the-art SR methods. Moreover, classification experiments on the reconstructed HSIs show that the proposed MS2Net-reconstructed HSIs achieve classification accuracy that is comparable to that of real HSIs.

P-964. Case-Based Learning Outcomes for Pre-Exposure Prophylaxis Education for the Preventative Medicine Plus One Maternal Health Track

Abstract Background As HIV prevention methods continue to develop with new agents added for Pre-Exposure Prophylaxis (PrEP). Prep uptake has remained low among US women since its approval in 2012. One major barrier to increasing PrEP uptake remains lack of knowledge of PrEP among patients and providers and lack of access to PrEP care in regions with high HIV prevalence. This study aims to increase PrEP knowledge and medical competence among Maternal Health Track Preventive Medicine residents at a single academic center in Southern California. Pre-Test and Post-Test Scores and % Improvement A pre-test was given at the start of each module. The same questions were made into a post-test after the curriculum was completed. Open and closed book portions were offered, and the % improvement was recorded. Methods Three Preventative Medicine Fellows from Family Medicine and Obstetrics and Gynecology backgrounds enrolled in multidisciplinary curriculum involving women’s health. A series of twelve case-based lectures were completed from July 2023 through December 2023. Each four-hour session included a pre-test, lecture, and a problem-based learning module. Simultaneous, they provided direct care for women in a combined clinic for PrEP and HIV care weekly for 3-6 months. In March 2024, the pre-test questions were combined into 47 post-test questions. Results Two of the three fellows had improvement in the closed book portion of the exam with a mean improvement of 9.5%. However, all of them had significant improvement for the open book exam with an average improvement of 16.2%. This suggests that although the memory retention of the material can wane with time, the ability to use resources to address clinically challenging questions has been strengthened. Of note, the questions were designed to simulate the American Academy of HIV Medicine (AAHIV) certification exam, which is an open-book test. Conclusion The combination of lectures, self-assessments, case-based discussions, and clinical experience has enhanced the fellows’ knowledge of PrEP care for women based on the improvement of exam scores after the curriculum. This will enhance their ability to provide multidisciplinary preventative care for women in underserved areas. Disclosures Kristin A. Carr, MD, Preventive Medicine Plus One Maternal Health Track: Grant/Research Support

MSEI-ENet: A Multi-scale EEG-Inception Integrated Encoder Network for Motor Imagery EEG Decoding

Background: Due to complex signal characteristics and distinct individual differences, the decoding of a motor imagery electroencephalogram (MI-EEG) is limited by the unsatisfactory performance of suboptimal traditional models. Methods: A subject-independent model named MSEI-ENet is proposed for multiple-task MI-EEG decoding. It employs a specially designed multi-scale structure EEG-inception module (MSEI) for comprehensive feature learning. The encoder module further helps to detect discriminative information by its multi-head self-attention layer with a larger receptive field, which enhances feature representation and improves recognition efficacy. Results: The experimental results on Competition IV dataset 2a showed that our proposed model yielded an overall accuracy of 94.30%, MF1 score of 94.31%, and Kappa of 0.92. Conclusions: A performance comparison with state-of-the-art methods demonstrated the effectiveness and generalizability of the proposed model on challenging multi-task MI-EEG decoding.

DEVELOPING FLIPBOOK-BASED DIGITAL MODULE OF AKIDAH AKHLAK SUBJECT TO INCREASE STUDENTS’ LEARNING MOTIVATION

Today's education requires teachers to provide learning that can utilize technology appropriately because, in this era of globalization, technology has become very important and cannot be separated from daily human life. So, teachers need to apply appropriate technology to assist the process of learning and improve learning efficacy, including using electronic devices. The Flipbook-based digital learning module is an achievement system that has been selected in light of developing a more efficient and effective education system. On that basis, the central theme of using a teaching system with Flipbook-based modules is to increase student learning motivation so that students become motivated and enthusiastic and can actively participate in learning. This research uses the R&D (Research & Development) research method. Analysis, design, development, implementation, and evaluation are the five phases that make up the ADDIE development model. The research results show that the Flipbook-based digital learning module is very suitable for use, with media expert assessment results of 92.5 percent and material experts 100 percent. Additionally, the findings of the student's learning motivation questionnaire indicate a relationship between digital learning modules and students' motivation to study.

Fostering Students' Competencies in Sustainability in Vocational Agriculture Education in Burundi: Effects of a Participatory Integrated Planning (PIP) Module

ABSTRACTDeveloping students' sustainable agriculture competencies requires the alignment of curriculum content to encompass the full complexity of the dynamics of agriculture. However, empirical evidence from sustainable agriculture research indicates that actions oriented towards fostering students' competencies are scarce. To advance understanding, this study monitored and evaluated the development of students' sustainability competencies for smallholder farming during the implementation of a Participatory Integrated Planning (PIP) module—A sustainability learning programme that elaborates on cross‐cutting issues of long‐term integrated planning, which is normally not taught. A total number of 104 students from two vocational agricultural schools (N1 = 63; N2 = 41) were involved in the PIP learning module that was implemented from April 2022 to June 2023. Survey data on the development of the students' competencies were collected at the beginning (April 2022), in‐between (December 2022) and end (June 2023) of the implementation of the PIP module. A one‐way repeated measures MANOVA test was used to evaluate the effect of the PIP module on the development of students' competencies. Findings revealed that the PIP module significantly contributed to fostering all intended competencies, including facilitator of change, leadership, innovation and creativity, planning, system thinking, stewardship, auto‐determination and interdisciplinarity. Improvement of these competencies was much higher in the earlier stage of the learning process compared to the later stage, implying that sustainability learning requires going beyond ‘business‐as‐usual’ to retain the interest of students and competencies development. Findings also revealed that competencies development levels differ depending on schools' context, implying that sustainability learning programmes should be flexible and adaptive for learning to occur in a real‐context environment. These findings have practical implications for policies related to designing and upscaling sustainability learning programmes in vocational agriculture schools.

Comparative Analysis of the Effectiveness of Informatics Course Learning Utilizing Chatgpt

This study examined the effectiveness of conventional teaching methods and ChatGPT in an introductory Algorithms and Programming course at the university level. ChatGPT, an AI-based NLP technology, assisted students in understanding course material through automated responses. However, its effectiveness relative to conventional methods required further evaluation, particularly concerning motivation, interaction, self-regulation, instructional structure, and the instructor's role. Using a sample of 10 students for pretest-posttest analysis, 38 respondents for the User Experience Questionnaire (UEQ), and accuracy analysis via prompt engineering, the results revealed that conventional methods better enhanced motivation and interaction. ChatGPT demonstrated strengths in attractiveness (1.982) and efficiency (2.053) but scored lower in accuracy (1.395) and novelty (1.053). Prompt engineering significantly improved response accuracy when tailored to learning modules, highlighting the importance of precise inputs. The findings suggested that while ChatGPT excelled as a supplementary tool, it was less effective as a standalone teaching method. This study contributed to the growing field of educational technology by providing insights into the integration of AI tools in learning environments.