Interactive Information Research Articles

DEVELOPMENT OF AUGMENTED REALITY (AR) IPAS LEARNING MEDIA TO IMPROVE CRITICAL THINKING SKILLS OF ELEMENTARY SCHOOL STUDENTS

This study aims to: 1) analyse the characteristics and quality of Class IV AR IPAS media products based on reviewer assessments; 2) implement AR IPAS media development products to improve the critical thinking skills of Class IV students; 3) analyze the effectiveness of AR IPAS media in improving critical thinking skills of Class IV students. The method used is the ADDIE RD model with the stages of Analysis, Design, Development, Implementation, and Evaluation. Data were collected through observation, interviews, documentation, validation sheets, assessments, and tests. Data analysis was carried out using qualitative and quantitative descriptive techniques. The results showed that AR IPAS media integrates the real and virtual worlds, and provides interactive and real-time information in 2D and 3D, made with the Assemblr Edu application. This media is effective, easy to operate, and affordable to make. However, AR media is sensitive to changes in viewing angle and requires sufficient memory capacity on the device used. The implementation of AR IPAS media at SD Negeri Nanggulan and MI Al-Huda was conducted in five meetings, students were enthusiastic and actively involved, allowing direct concept exploration. AR IPAS media products are proven to be effective, proven through paired sample t-tests on two classes showing significance p 0.05, indicating a significant difference in students' critical thinking skills before and after using AR media.

Digital Information Exchange Between the Public and Researchers in Health Studies: Scoping Review.

Information exchange regarding the scope and content of health studies is becoming increasingly important. Digital methods, including study websites, can facilitate such an exchange. This scoping review aimed to describe how digital information exchange occurs between the public and researchers in health studies. This scoping review was prospectively registered and adheres to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines. Eligibility was defined using the population (public and researchers), concept (digital information exchange), and context (health studies) framework. Bibliographic databases (MEDLINE, PsycINFO, CINAHL, and Web of Science), bibliographies of the included studies, and Google Scholar were searched up to February 2024. Studies published in peer-reviewed journals were screened for inclusion based on the title, abstract, and full text. Data items charted from studies included bibliographic and PCC (Population, Concept, and Context) characteristics. Data were processed into categories that inductively emerged from the data and were synthesized into main themes using descriptive statistics. Overall, 4072 records were screened, and 18 studies published between 2010 and 2021 were included. All studies evaluated or assessed the preferences for digital information exchange. The target populations included the public (mainly adults with any or specific diseases), researchers, or both. The digital information exchange methods included websites, emails, forums, platforms, social media, and portals. Interactivity (ie, if digital information exchange is or should be active or passive) was addressed in half of the studies. Exchange content included health information or data with the aim to inform, recruit, link, or gather innovative research ideas from participants in health studies. We identified 7 facilitators and 9 barriers to digital information exchange. The main facilitators were the consideration of any stakeholder perspectives and needs to clarify expectations and responsibilities, the use of modern or low-cost communication technologies and public-oriented language, and continuous communication of the health study process. The main barriers were that information exchange was not planned or not feasible due to inadequate resources, highly complex technical language was used, and ethical concerns (eg, breach of anonymity if study participants are brought together) were raised. Evidence gaps indicate that new studies should assess the methods and the receiver (ie, public) preferences and needs that are required to deliver and facilitate interactive digital information exchange. Few studies addressing digital information exchange in health studies could be identified in this review. There was little focus on interactivity in such an exchange. Digital information exchange was associated with more barriers than facilitators, suggesting that more effort is required to improve such an exchange between the public and researchers. Future studies should investigate interactive digital methods and the receiver preferences and needs required for such an exchange.

Lightweight Safety Activity Recognition Algorithm for Railway Field Personnel Based on Portable Cards

In recent years, railway construction has been plagued by frequent safety accidents, with workers’ safety during the construction process remaining a major concern. To mitigate this issue, intelligent monitoring of railway workers’ activity has been proposed as a means of improving the safety coefficient of construction. Human activity recognition (HAR) based on wearable devices holds significant application value in areas such as health monitoring, motion analysis, and intelligent assistance. Recently, convolutional neural networks (CNNs) have gained extensive adoption and demonstrated outstanding performance in HAR. However, current HAR research still faces some challenges, including problems with establishing spatial–temporal dependencies and addressing the demand for lightweight models. To address the above issues, we propose a lightweight dual-stream convolution model (LDSC) based on deformable convolution and hierarchical segmentation. The model adaptively captures significant variations in sensor readings over time from portable cards of railway personnel through a temporal stream and learns the interactive information among sensor channels over a spatial stream. LDSC consists of three lightweight convolutional modules that combine deep convolution and point convolution to reduce model parameters, thus meeting the demand for a lightweight model. Experiments and ablation studies are conducted on three available datasets (UCI-HAR, UniMiB-SHAR, and WISDM) to evaluate the proposed model. The experimental results indicate that our model outperforms existing state-of-the-art methods in terms of recognition accuracy, validating the effectiveness and feasibility of LDSC. In addition, theoretical analysis and ablation experiments demonstrate that the proposed LDSC embodies lightweight characteristics.

Development of a Web-Based e-Portal for Freeform Surfaced Lens Design and Manufacturing and Its Implementation Perspectives

In modern freeform surfaced optics manufacturing, ultraprecision machining through single-point diamond turning (SPDT) plays a crucial role due to its ability to meet the high accuracy demands of optical design and stringent surface quality requirements of the final optic. The process involves meticulous steps, including optic surface modeling and analysis, optic design, machining toolpath generation, and manufacturing. This paper presents an integrated approach to customized precision design and the manufacturing of freeform surfaced varifocal lenses through a web-based e-portal. The approach implements an e-portal-driven manufacturing system that seamlessly integrates lens design, modeling and analysis, toolpath generation for ultraprecision machining, mass personalized customization, and service delivery. The e-portal is specifically designed to meet the stringent demands of personalized mass customization, and to offer a highly interactive and transparent experience for the lens users. By using Shiny and R-script programming for platform development and combining COMSOL Multiphysics for the ray tracing simulation, the e-portal leverages open-source technologies to provide manufacturing service agility, responsiveness, and accessibility. Furthermore, the integration of R-script and Shiny programming allows for advanced interactive information processing, which also enables the e-portal-driven manufacturing system to be well suited for personalized complex products such as freeform surfaced lenses.

Empowering Global AMR Research Community: Interactive GIS dashboards for AMR data analysis and informed decision-making

Background Antimicrobial Resistance (AMR) poses a global public health challenge, necessitating advanced tools to support data analysis, and visualization. This study introduces interactive Geographic Information System (GIS) dashboards as innovative platforms for AMR data analysis and visualization, offering comprehensive insights into resistance patterns, and geographic distribution across multiple countries, with a specific focus on Africa. Methods Three GIS dashboards were developed to address key objectives. The first integrates over 860,000 ATLAS data points from 83 countries, providing an interactive platform. Users can filter data by variables such as country, year, and region, enhancing data accessibility and visualization. The second dashboard focuses on the ATLAS dataset for Kenya and Uganda, incorporating detailed variables such as species, sample sources, and resistance phenotypes. The third involves Kampala, Uganda, to fill data gaps, enabling localized analyses through interactive features like geographic mapping and sample breakdowns by year. Results The dashboards demonstrated significant utility in visualizing and analyzing AMR data. The global dashboard effectively highlighted geographical trends and exposed critical data gaps, such as the scarcity of AMR records from Africa. The Kenya and Uganda dashboard revealed alarming resistance patterns, including the ineffectiveness of Ceftriaxone, Erythromycin, Levofloxacin, and Ampicillin in combating E. coli isolates. Notably, the Kampala-specific dashboard utilized simulated data to illustrate the potential of this tool for in-depth regional analysis of antibiotic resistance. This simulation showcased the dashboard's capability to incorporate real-time filtering, provide detailed insights into sample sources, and track trends. Across all dashboards, the interactivity significantly streamlined the identification of key trends, simplifying data interpretation for diverse users and supporting informed decision-making processes. Conclusions Interactive GIS dashboards offer a more effective alternative to spreadsheets for AMR data analysis. While spreadsheets provide a basic structure for organizing and managing data, dashboards deliver dynamic, user-friendly interfaces that enable advanced visualization, real-time updates, and intuitive data exploration. Dashboards also enhance accessibility and communication by presenting complex datasets in a more interactive and visually engaging format, making them more efficient for analyzing and interpreting AMR data.

Exploring the Current State of Information Literacy Education: A Systematic Literature Review

Information Literacy Education is a skill taught to students of various educational levels, from elementary school to college and even postgraduate programs. This research aims to present the current state of Information Literacy Education, identify the dominant research themes, the methods of teaching and learning used, the context of learning in which the skill is taught, and recommends future research development in related fields. A systematic literature review was conducted using large electronic journal databases such as Taylor and Francis, and Science Direct, which obtained 37,633 articles. Furthermore, The PICOC and PRISMA protocols were applied to analyze 17 relevant articles and identify 14 themes in learning, 14 learning methods, 12 research methods used, and 7 scientific fields in the implementation of Information Literacy Learning from 2017 to 2021. These results provide direction for future research and contribute to enhanced information literacy curriculum development, effective integration of information literacy in specific disciplines, transitioning to online information literacy instruction, development of engaging and interactive information literacy modules, assessing learning outcomes and informing instructional strategies, collaboration and professional development in information literacy, diverse and effective teaching methods, and advancements in research methods for information literacy.

Strengthening Responsible Journalism Through Self-Efficacious Learning-Oriented Media Literacy Interventions

This article investigates the challenges journalism professionals face in a rapidly changing digital media environment, proposing that a “processual” and human-centered perspective might offer valuable insights into developing resilient professionalism. The article builds its argument on theories of transmediality and hybridization in digital media ecosystems and the socio-psychological development toward accountable communication and responsible professionalism. It specifically looks at future journalists as active learners to whom media literacy interventions may offer new insights into the mental processes in professional decision-making. It tests these ideas in an experimental study with journalism students, where the lateral reading approach was applied within the framework of learning skills for information verification. Results from the thematic analysis of students’ reflexive assessments of their practice reveal norms illustrative of a self-efficacious learning process: Students’ answers demonstrate empowering and perseverance-directed approaches. As argued, these norms are geared toward imposing a higher media awareness and self-regulatory capacity, which is critical for accountable decision-making in transmedial and highly interactive digital information environments.

Improving skeleton-based action recognition with interactive object information

Improving skeleton-based action recognition with interactive object information

Classroom Application of English Language and Literature Wisdom based on Artificial Intelligence Technology

In order to improve the ability of intelligent classroom management and dynamic management of English language and literature in streaming media environment, this paper proposes an intelligent classroom application of English language and literature based on artificial intelligence technology. To construct the information interaction resource scheduling and fitness control model of English language and literature smart classroom in streaming media environment, this paper restructures the situation and context information of language application, and extract the course parameter setting information of smart classroom teaching mode. The sparsity feature decomposition is carried out on the personalized assignment set assigned by the smart platform, and the recognition evaluation decision of the smart classroom teaching mode is realized according to the coding results of interactive information features. The control of information interaction is achieved based on the evaluation results concerning the level of classroom participation and the effectiveness of learning. The changes of students’ satisfaction with teaching design and students’ participation in class were analyzed. The operation data were divided into decision-making data and computing resource pool data to achieve intelligent classroom design of English language and literature. The experimental results show that the proposed method has good adaptability in the intelligent classroom design of English language and literature, the level of students’ participation and satisfaction is higher, and the utilization rate of classroom teaching resources is higher.

Quality Assessment of a Medical Prescription for Antibiotic Therapy

Objective: Our study aimed to assess the quality of antibiotic prescriptions and identify potential opportunities for improvement in antibiotic use in our setting. Methodology: This is how we conducted a prospective cross-sectional descriptive study which concerned 404 prescriptions collected in the pharmacy of the HGR mwangeji Results: The most prescribed antibiotic families were beta lactams with 60.15%, followed by imidazoles with 11.88% and fluoroquinolones or 8.1%; they were more prescribed in INN with 70% in the following galenic forms: tablets 46.8% followed by bottles 17.1% and finally suspensions with 12.6% with an unspecified route of administration in 80.9%. The dosage, dose, total quantity and duration of treatment were specified respectively as follows: 86.9%; 62.4%; 98.3% and 80.4%. Most of the prescriptions bore the name and first name of the patients 82.9%; The age of the patient was indicated at 75.2%; Almost all of the prescriptions did not have the weight of the patient indicated (97%); Almost all of them bore the sex of the patient (97.6%); Not all prescribers wore the patient's size (100%). In 84.4% of cases, prescribers had indicated their name on the prescriptions; A large number of prescription slips did not bear the prescriber status 86.6%; Almost all prescribers had indicated their place of practice 99.5%; 97% of prescribers had not indicated their contact details on the prescriptions; The minority 18.6% of prescribers had not signed on the prescriptions; The order number is shown on the majority of prescription slips 83.7%. Conclusion: The results of this study remain relevant. They will help improve training programs for health workers by establishing interactive information for practitioners, through visits and intensive supervision by more experienced managers in the hierarchy, this would be likely to optimize the quality of practitioners' prescriptions ; so that they better master the principles of prescribing antibiotics . Keywords: Prescription, Antibiotics, evaluation, quality, medical, antibiotic therapy

Enhancing Community Cooperation through Network Governance and Smart Specialisation to Mitigate Social Conflicts

Introduction: The relevance of the research topic is primarily in introducing the concepts of network management and smart specialisation to strengthen social ties, avoid social conflicts and increase the competitiveness of communities through interactive information exchange and active participation of stakeholders. Objectives: The main aim of writing a scientific article is to substantiate the need to apply network management practices and smart specialisation within the framework of community cooperation to prevent social conflicts. Methods: During the study, several general scientific methods of cognition were used, including induction, synergy, comparative analysis, deductive and analytical methods, and systematisation and generalisation. Results: The study identified areas for effectively implementing network management principles in Ukraine, including developing a unified methodology for evaluating results, a long-term register of initiatives and partnership mechanisms, strengthening territorial communities, adapting the project approach, and increasing its maturity. In addition, priority areas for improving the current strategy of smart specialisation of Ukraine's regions and ways to optimise community cooperation have been identified, including introducing a data collection system, coordinating participants, integrating technologies, and developing communication strategies. Conclusions: Thus, the developed measures to optimise community cooperation, based on network management and smart specialisation of regions, allow for meeting society's current needs and strengthening the community population's trust in state institutions, ensuring their effectiveness and sustainable development of the region of Ukraine.

DENSE CLUSTERS OF RAW SEWAGE LOCATIONS ON MINNA NEIGHBOURHOOD GIS MAPS ARE POINTERS TO VERITABLE URBAN DECAY AND MOSQUITO-BREEDING GROUNDS

Background: The lack of comprehensive baseline data on raw sewage pollution in Minna, Nigeria's Niger State capital, represents a significant gap in public health information. This deficiency necessitates a systematic study to establish a database documenting sewage pollution patterns in Minna's urban environment. Aim: To employ georeferencing tools and descriptive observations for identifying and mapping potential mosquito breeding sites related to sewage discharge within Minna's built-up areas and to develop an interactive Geographic Information System (GIS) map as an environmental audit tool for public health officials. Methods: The study area was divided into five sectors: Greater Bosso, Minna Central, Greater Maitumbi, Tunga, and Greater Chanchaga. Following initial site familiarization and GIS equipment testing, systematic surveys were conducted. Field teams documented locations of household sewage discharge points, collecting geographic coordinates, temporal data (date, time, weather conditions), and site characteristics. Each identified location was photographed and recorded in standardized data sheets, including household information where available. Results: Using ArcGIS®10.8 software, comprehensive sewage pollution layers were created for each sector by integrating collected field data with Minna's township built-up and settlement shapefiles. The mapping revealed distinct pollution patterns and clusters across different neighborhoods. Discussion: Analysis of the spatial distribution showed a clear correlation between socioeconomic status and sewage management practices, with higher concentrations of improper sewage discharge in low-income areas. Conclusions: This database serves as a valuable resource for public health interventions targeting mosquito breeding grounds. Regular monitoring through periodic sewage pollution audits and expanded geographical coverage is recommended for improved urban health management.

EVIDENCE FROM REPAYMENT OF ACCOUNTS PAYABLE: QUANTITATIVE AND QUALITATIVE INDICATORS

The article examines the criminological characteristics of economic crime in general, as well as the example of the crime, liability for which is provided for by Art. 242 of the Criminal Code of the Republic of Belarus (hereinafter referred to as the Criminal Code of the Republic of Belarus) "Evasion of repayment of accounts payable", analyzes qualitative and quantitative indicators. The subject of the study is the level, structure, dynamics of economic crime, the study is based on official data of the interactive information and analytical system for the dissemination of official statistical information. The level of economic crime in the Republic of Belarus for 2009 - 2023 was studied, the percentage of economic crime from the total number of socially dangerous acts was derived. The structure of economic crime was analyzed, the percentage of evasion of repayment of accounts payable from the total number of crimes encroaching on the economic security of the Republic of Belarus was identified.

The potential of GPTs for enhanced information access and user services at academic libraries

Generative pre-trained transformers (GPTs) hold the promise of revolutionizing information access and user services. The purpose of this study is to examine the possible uses and applications of GPTs in academic libraries. One important objective is to investigate how GPTs can improve information access and user services within academic library environments. In this research, 10 information specialists from academic libraries in Oman were interviewed using a semi-structured interview design. The results highlighted an in-depth understanding of the underlying potential of GPTs in academic libraries and their challenges. GPTs can enhance users’ experience by facilitating an interactive user-based language information retrieval model. They can also support library technical operations. However, there are also challenges related to information quality and reliability, as well as ethical considerations mainly related to user data privacy and security. These insights, coming from information specialists, can contribute to advancing future integration frameworks that enforce the regulation, integration, and use of information.

An investigation into the relationship amongst local government employees and their performances towards the distribution of services: study of Umhlathuze Local Municipality

This study's objective was to investigate the relationship amongst the local government employees and their performances towards the distribution of services. Many employees face different challenges in their workplaces, however in a situation like service delivery it’s important for employees to be able to work well together and have a positive relationship amongst each other and avoid political conflict that may end up affecting the citizens as they are their priority. The researcher acknowledges that there are many challenges faced by local government which affect the effective service delivery, these include toxic work relationships among employees, corruption and mismanagement of funds, red tape etc. The study was based on the current legal, theoretical and conceptual perspective that applies to the protocol of the working relationship amongst employees and service provision. Diverse perspectives were used to determine whether the employees of UMhlathuze Local Municipality have the positive relationship amongst each other without the interference of the political conflict, also for the community members under UMhlathuze Local Municipality to have a voice on whether the employees are distributing services rightfully or not. This study has adopted qualitative methods for getting in-depth and interactive information from the participants. The recommendations are also made for UMhlathuze Municipality, including the theoretical framework of service delivery and the working protocol of employees within an organization.

Research on Interactive Efficiency and Experience of Intelligent Education Platform

To explore online user behavioral habits from the perspectives of cognitive efficiency and emotional experience, as well as to provide objective and quantitative physiological evidence for the design of intelligent online education platforms and the technological application of Education 4.0. In this paper, 45 subjects were selected and recruited using a combination of eye-tracking and questionnaires, with a male-to-female ratio of 1:2. The results show that: (1) Intelligent interaction modes significantly impact online users’ positive emotional arousal compared to traditional interaction modes; (2) Diversified presentation of interactive information can help online learners understand collaborators’ opinions more quickly and optimize the experience of online peer Interaction; (3) Multi-channel interaction can provide learners with a better emotional experience while improving the efficiency of interacting with instructional information. This paper provides an objective design basis for intelligent platforms like online teaching and conference software in the context of information technology.

Assessing the Value of Information in an Augmented Reality City Experience

Information is an experience good, meaning that its value emerges upon use and varies based on individual perceptions. Augmented reality (AR) is a technology designed to deliver immersive informational experiences. This study investigates the impact of AR on the perceived value of information by people experiencing information as consumers or producers. Using Google Maps, we developed an AR mobile map called Tour-It-Yourself (TIY) for city navigation. The TIY app facilitates content consumption and production around local points of interest engagingly and interactively. We discuss the development of the research tool and results from two user studies (N = 37 teenagers and N = 51 students, respectively). Further, we highlight the economic valuation aspect by examining the effects of AR on Willingness-to-Pay and Willingness-to-Accept payment for information. The results indicate a significant main effect of participant experience (consumer/producer) and AR technology on the perceived evaluation of information. Moreover, the addition of AR technology to information improves knowledge regarding points of interest for producers. These findings contribute to our understanding of how AR shapes information value perception and have implications for designing interactive location-based information systems, benefiting tourism, education, and business, as well as fostering a sense of belonging in cities.

Utilization of Augmented Reality as a Media Introduction to Pests in Rice Plants Using the Marker Base Method

Indonesia is an agrarian country where rice is a major commodity and a primary food source for the Indonesian population. To encourage the interest and knowledge of the younger generation as farmers, innovation is needed as a basic learning medium about agriculture, specifically rice. This research aims to develop a mobile application based on Marker-Based Augmented Reality (AR) to enhance the effectiveness of pest identification in rice plants. The application is designed to provide visual and interactive information to farmers about the types of pests that frequently attack rice plants, along with their characteristics. Data on rice pests are collected through literature reviews and interviews with several rice farmers. The application utilizes markers that can be recognized by the mobile device's camera to display 3D models and information about the pests. Testing results show that the application works well as expected, and the 3D object information regarding rice pests, along with their characteristics, can be displayed. Additionally, this application is expected to contribute to reducing crop losses due to pest attacks by improving farmers' quick and accurate responses, and it can serve as an effective and innovative tool in the field of agriculture, particularly in pest identification in rice plants.

A Scoping Study on AI Affordances in Early Childhood Education: Mapping the Global Landscape, Identifying Research Gaps, and Charting Future Research Directions

Artificial intelligence (AI), manifested in the forms of technologies, systems, tools, and applications, has advanced rapidly, especially in recent years. It has permeated many aspects of human behavior and nearly all sectors of society, such as healthcare and education. In the context of early childhood education (ECE), AI has afforded valuable opportunities that directly and indirectly enhance children’s learning and development. While there are already two existing reviews of the literature on AI in ECE, they show either a lack of descriptive information concerning selected studies or inconsistencies between inclusion/exclusion criteria and selected studies, thereby raising concerns about their rigor. Representing a more methodologically rigorous effort and a significant contribution to the field of AI in ECE, this scoping study aimed to achieve three main goals: (1) “mapping” the global landscape of the current extent, range, and nature of relevant studies on the affordances of AI for use in ECE, (2) identifying potential research gaps, and (3) charting future research directions. Specifically, it addressed this overarching research question: What is the global landscape of the current state of knowledge concerning the affordances of AI for use in ECE? Specifically, the state of knowledge here refers to three aspects: (1) extent, (2) range, and (3) nature. First, regarding the extent aspect, the empirical knowledge was derived from 18 research articles in 11 countries and 16 peer-reviewed academic journals between 2005 and 2023, with 14 of these articles published in the past four years (2020–2023). Second, with respect to the range of study populations, it covered 15,081 children in early childhood (ages 2 to 8 years) across these 11 countries. Third, thematic analysis of these studies revealed four areas of AI affordances: (1) AI as tangible and intangible tools for interactive learning and information retrieval, (2) AI as technology for predicting/classifying children’s conditions, (3) AI as the object for learning by adapting to and personalizing children’s learning, and (4) AI as the subject for children's learning about it. Based on these findings, this scoping review identified three research gaps for future studies: (1) interviewing and/or surveying education stakeholders (parents, educators, policymakers) to explore the affordances of appropriate AI for use with, by, and for children bearing ethical considerations; (2) conducting group comparisons to investigate contextual factors contributing to the “AI divide” among children from different socioeconomic backgrounds; and (3) comparing sociocultural influences on AI use in ECE across cultures.

Strip and boundary detection multi-task learning network for segmentation of meibomian glands.

Automatic segmentation of meibomian glands in near-infrared meibography images is basis of morphological parameter analysis, which plays a crucial role in facilitating the diagnosis of meibomian gland dysfunction (MGD). The special strip shape and the adhesion between glands make the automatic segmentation of meibomian glands very challenging. A strip and boundary detection multi-task learning network (SBD-MTLNet) based on encoder-decoder structure is proposed to realize the automatic segmentation of meibomian glands. A strip mixed attention module (SMAM) is proposed to enhance the network's ability to recognize the strip shape of glands. To alleviate the problem of adhesion between glands, a boundary detection auxiliary network (BDA-Net) is proposed, which introduces boundary features to assist gland segmentation. A self-adaptive interactive information fusion module (SIIFM) based on reverse attention mechanism is proposed to realize information complementation between meibomian gland segmentation and boundary detection tasks. The proposed SBD-MTLNet has been evaluated on an in-house dataset (453 images) and a public dataset MGD-1K (1000 images). Due to the limited number of images, a five-fold cross validation strategy is adopted. Average dice coefficient of the proposed SBD-MTLNet reaches 81.08% and 84.32% on the in-house dataset and the public one, respectively. Comprehensive experimental results demonstrate the effectiveness the proposed SBD-MTLNet, outperforming other state-of-the-art methods. The proposed SBD-MTLNet can focus more on the shape characteristics of the meibomian glands and the boundary contour information between the adjacent glands via multi-task learning strategy. The segmentation results of the proposed method can be used for the quantitative morphological characteristics analysis of meibomian glands, which has potential for the auxiliary diagnosis of MGD in clinic.